Journal of Mechanical Engineering 2011 5

57 (2011) 5 1

Platnica SV-JME 57(2011)5_kor1.ai 1 17.5.2011 7:52:01

Since 1955

C

M

Y

CM

MY

CY

CMY

Journal of Mechanical Engineering - Strojniški vestnik

K

5 year 2011 volume 57 no.

Platnica SV-JME 57(2011)5_kor1.ai 2 17.5.2011 7:52:14

Strojniški vestnik – Journal of Mechanical Engineering (SV-JME) Aim and Scope The international journal publishes original and (mini)review articles covering the concepts of materials science, mechanics, kinematics, thermodynamics, energy and environment, mechatronics and robotics, fluid mechanics, tribology, cybernetics, industrial engineering and structural analysis.

The journal follows new trends and progress proven practice in the mechanical engineering and also in the closely related sciences as are electrical, civil and process engineering, medicine, microbiology, ecology, agriculture, transport systems, aviation, and others, thus creating a unique forum for interdisciplinary or multidisciplinary dialogue.

The international conferences selected papers are welcome for publishing as a special issue of SV-JME with invited co-editor(s).

Editor in Chief Vincenc Butala University of Ljubljana Faculty of Mechanical Engineering, Slovenia Co-Editor Borut Buchmeister University of Maribor Faculty of Mechanical Engineering, Slovenia Technical Editor Pika Škraba University of Ljubljana Faculty of Mechanical Engineering, Slovenia

C

M

Y

CM

MY

CY

CMY

K

Editorial Office University of Ljubljana (UL) Faculty of Mechanical Engineering SV-JME Aškerčeva 6, SI-1000 Ljubljana, Slovenia Phone: 386-(0)1-4771 137 Fax: 386-(0)1-2518 567 E-mail: info@sv-jme.eu http://www.sv-jme.eu Founders and Publishers University of Ljubljana (UL) Faculty of Mechanical Engineering, Slovenia University of Maribor (UM) Faculty of Mechanical Engineering, Slovenia Association of Mechanical Engineers of Slovenia Chamber of Commerce and Industry of Slovenia Metal Processing Industry Association Cover: EOS Formiga P100 SLS machine (at the background), Selective Laser Sintering process (small photo above) and Additive Manufacturing products (small photo below).

Image courtesy: Intelligent Manufacturing Systems Laboratory, Production Engineering Institute, Faculty of Mechanical Engineering, University of Maribor

ISSN 0039-2480 © 2011 Strojniški vestnik - Journal of Mechanical Engineering. All rights reserved. SV-JME is indexed / abstracted in: SCI-Expanded, Compendex, Inspec, ProQuest-CSA, SCOPUS, TEMA. The list of the remaining bases, in which SV-JME is indexed, is available on the website. The journal is subsidized by Slovenian Book Agency.

International Editorial Board Koshi Adachi, Graduate School of Engineering,Tohoku University, Japan Bikramjit Basu, Indian Institute of Technology, Kanpur, India Anton Bergant, Litostroj Power, Slovenia Franci Čuš, UM, Faculty of Mech. Engineering, Slovenia Narendra B. Dahotre, University of Tennessee, Knoxville, USA Matija Fajdiga, UL, Faculty of Mech. Engineering, Slovenia Imre Felde, Bay Zoltan Inst. for Mater. Sci. and Techn., Hungary Jože Flašker, UM, Faculty of Mech. Engineering, Slovenia Bernard Franković, Faculty of Engineering Rijeka, Croatia Janez Grum, UL, Faculty of Mech. Engineering, Slovenia Imre Horvath, Delft University of Technology, Netherlands Julius Kaplunov, Brunel University, West London, UK Milan Kljajin, J.J. Strossmayer University of Osijek, Croatia Janez Kopač, UL, Faculty of Mech. Engineering, Slovenia Franc Kosel, UL, Faculty of Mech. Engineering, Slovenia Thomas Lübben, University of Bremen, Germany Janez Možina, UL, Faculty of Mech. Engineering, Slovenia Miroslav Plančak, University of Novi Sad, Serbia Brian Prasad, California Institute of Technology, Pasadena, USA Bernd Sauer, University of Kaiserlautern, Germany Brane Širok, UL, Faculty of Mech. Engineering, Slovenia Leopold Škerget, UM, Faculty of Mech. Engineering, Slovenia George E. Totten, Portland State University, USA Nikos C. Tsourveloudis, Technical University of Crete, Greece Toma Udiljak, University of Zagreb, Croatia Arkady Voloshin, Lehigh University, Bethlehem, USA President of Publishing Council Jože Duhovnik UL, Faculty of Mechanical Engineering, Slovenia Print Tiskarna Present d.o.o., Ižanska cesta 383, Ljubljana, Slovenia General information Strojniški vestnik – The Journal of Mechanical Engineering is published in 11 issues per year (July and August is a double issue).

Institutional prices include print & online access: institutional subscription price and foreign subscription €100,00 (the price of a single issue is €10,00); general public subscription and student subscription €50,00 (the price of a single issue is €5,00). Prices are exclusive of tax. Delivery is included in the price. The recipient is responsible for paying any import duties or taxes.

Legal title passes to the customer on dispatch by our distributor.

Single issues from current and recent volumes are available at the current single-issue price. To order the journal, please complete the form on our website. For submissions, subscriptions and all other information please visit: http://en.sv-jme.eu/.

You can advertise on the inner and outer side of the back cover of the magazine. The authors of the published papers are invited to send photos or pictures with short explanation for cover content.

We would like to thank the reviewers who have taken part in the peer-review process.

Strojniški vestnik - Journal of Mechanical Engineering is also available on http://www.sv-jme.eu, where you access also to papers’ supplements, such as simulations, etc.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5 Contents

Contents Strojniški vestnik - Journal of Mechanical Engineering volume 57, (2011), number 5 Ljubljana, May 2011 ISSN 0039-2480 Published monthly

Papers Domen Šeruga, Matija Fajdiga, Marko Nagode: Creep Damage Calculation for Thermo Mechanical Fatigue Uroš Zupanc, Janez Grum: Surface Integrity of Shot Peened Aluminium Alloy 7075-T651 Uroš Trdan, José Luis Ocaña, Janez Grum: Surface Modification of Aluminium Alloys with Laser Shock Processing Edvard Detiček, Uroš Župerl: An Intelligent Electro-Hydraulic Servo Drive Positioning Ugljesa Bugaric, Dusan Petrovic, Zoran Petrovic, Miroslav Pajcin, Gordana Markovic-Petrovic: Determining the Capacity of Unloading Bulk Cargo Terminal Using Queuing Theory Fatih Taylan, Oğuz Çolak, Mehmet Cengiz Kayacan: Investigation of TiN Coated CBN and CBN Cutting Tool Performance in Hard Milling Application Branko Tadic, Djordje Vukelic, Janko Hodolic, Slobodan Mitrovic, Milan Eric: ConservativeForce-Controlled Feed Drive System for Down Milling Instructions for Authors

371 379 385 394 405 417 425 440

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 371-378 DOI:10.5545/sv-jme.2010.108

Paper received: 06.05.2010 Paper accepted: 07.03.2011

Creep Damage Calculation for Thermo Mechanical Fatigue Šeruga, D. – Fajdiga, M. – Nagode, M. Domen Šeruga* – Matija Fajdiga – Marko Nagode University of Ljubljana, Faculty of Mechanical Engineering, Slovenia

For the needs of the creep damage prediction of thermo mechanically loaded components, software has been developed. It enables master curve determination using the time-temperature parameters and creep damage calculation using Robinson’s damage accumulation rule and simple time integration. The developed software makes it possible to calculate fatigue and creep damage, respectively. In the article the most used time-temperature parameters are introduced, a fast and user-friendly master curve determination is presented and an example of creep damage on a real set of data with a simple temperature-stress history is shown. ©2011 Journal of Mechanical Engineering. All rights reserved.

Keywords: creep, damage model, parametric methods, damage accumulation, Larson-Miller, Manson-Brown, Manson-Haferd, Orr-Sherby-Dorn, master curves 0 INTRODUCTION Automotive components, gas and steam turbines, power plants and other products that operate at high temperature are subjected to creep. The lifetime of these components depends on thermal and mechanical loading. Typical examples are start-up, full load, partial load and shut-down [1]. The creep damage can no longer be neglected when the loading temperature exceeds the creep temperature typically determined as 40% of the melting temperature of the material [2] to [4]. Thus one of the critical factors in determining the lifetime of components is also their creep resistance. Due to the thermal loads materials slowly but constantly creep even at low mechanical loading, so rupture is possible.

Rupture can be defined by some limit value of strain or by actual rupture depending on the type of component. It can be shown that the creep damage is determined by knowing the time to rupture depending on stress level and temperature.

1 MASTER CURVES The criterion for creep damage calculation is given by master curves which represent time to rupture depending on stress level and temperature (Fig. 1). Each point on a master curve represents a complete creep-rupture test at constant temperature and stress level. As expected, higher test temperatures shift master curves towards lower stress levels and shorter times to rupture.

Master curve determination at lower stress levels and lower test temperatures usually means tests of very long duration, thus as a rule master curves in these areas are determined by using parametric extrapolation methods.

Fig. 1. Master curves 2 CREEP DAMAGE ACCUMULATION RULE Although the creep damage at controlled test conditions is relatively easy to obtain, components rarely operate under constant conditions. The most frequently used approach to creep damage assessment under variable thermo mechanical loading is to calculate the time for which the component is subjected to some

*Corr. Author’s Address: University of Ljubljana, Faculty of Mechanical Engineering, Aškerčeva 6, 1000 Ljubljana, Slovenia, domen.seruga@fs.uni-lj.si

371

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 371-378

loading. Robinson proposed a linear creep damage accumulation rule [5]:

ni

mij

Dc = ∑ ∑ i =0 j =0

∆tij (σ ij , Tij ) trij (σ ij , Tij )

, (1)

where Δtij and trij are the actual times under some loading and temperature and the corresponding time to rupture at the same loading and temperature, respectively. Index i represents the number of the time step in load history and index j represents additional subdivisions of time step i.

Loading temperature T has to be higher or equal to creep temperature Tc and loading stress σ has to be higher than the temperature dependent elastic limit of the material k(T) otherwise, no creep occurs [2] to [4], (Fig. 2). If the loading temperature is beneath the limit temperature for which material data is available, then the material data for the limit temperatures is taken into account. The damage accumulation in a single time step is independent of the previously accumulated damage. When the sum of individual creep damages reaches the defined limit value (usually 1), creep rupture occurs. Robinson’s damage accumulation rule is the most widely accepted one [1], [3], [4], [6] to [8].

different creep relations used to calculate creep damage due to compressive stresses [4], [9] and [10]. The creep relation can allow either tensile creep only if:

0 < tr < ∞ if σ > k(T) and T > Tc,

otherwise tr = ∞, or tensile-compressive creep if:

0 < tr < ∞ if |σ|>|k(T)| and T > Tc, otherwise tr = ∞, or compressive healing if: –∞ < tr < 0 if σ < –k(T) and T > Tc, 0 < tr < ∞ if σ > k(T) and T > Tc, otherwise tr = ∞.

The creep healing is possible only if Dc(t) > 0. Moreover, the tensile-compressive creep is larger than or equal to the tensile creep that is larger than or equal to the compressive healing.

The appropriate creep relation depends on the material. Some materials tend to heal under compressive loading while for other materials creep damage continues to grow regardless of the direction of the loading. The influence of the creep relation on the creep damage accumulation is shown in Fig. 3. The temperature is supposed to be constant, while stress changes from tension to compression at t1. Before t1 the creep relation does not affect Dc. However, after t1 the tensilecompressive creep results in the highest Dc, while compressive healing in the lowest.

Fig. 3. Influence of the creep relation upon the creep damage accumulation Fig. 2. Creep damage calculation Creep damage is calculated as a simple integration over all the time increments according to Eq. (1) (Fig. 2). However, there are three 372

3 SPECIMENS, MATERIAL AND TESTING Master curves and creep damage calculation are performed on a real set of creep – rupture data. They were gained from an available existing database [11].

Šeruga, D. – Fajdiga, M. – Nagode, M.

StrojniĹĄki vestnik - Journal of Mechanical Engineering 57(2011)5, 371-378

3.1 Material The material under investigation is 5Cr0.5Mo steel. It is suitable for components operating at high temperatures. Specimens with geometry of 6 mm in diameter and 30 mm in gauge length were taken longitudinally from the boiler tubes at the middle of the wall thickness.

3.2 Testing Material is tested at temperatures of 500, 550, 600 and 650 ÂşC and at various stress levels with standard creep-rupture testing [ASTM E 139 â&#x20AC;&#x201C; 00]. Every test is performed at a single temperature and a single load up to rupture. The specimen is loaded with a constant load. Time to rupture, area reduction at rupture and elongation at rupture are recorded.

3.3 Test Results Test results are shown in Fig. 4. For extrapolation purposes one set of data is chosen.

Since the main goal is to determine master curves from the shortest creep-rupture tests, results at the highest stresses and the highest temperatures

are selected. Results gained by extrapolation are compared with 4 comparison sets of experimental data. Each comparison set represents the results of creep-rupture testing at a distinct test temperature.

4 TIME-TEMPERATURE PARAMETERS Time-temperature parameters represent a parametric method for determining time to rupture. The basic idea of every parametric method is to predict long-term creep behaviour by compensating time with temperature, i.e. to obtain long-term creep estimates from short-term tests at higher temperature at the same stress [12].

Time-temperature compensation parameters are introduced to establish a model for the estimation of long-term creep properties. Mostly used compensation parameters are introduced below.

The assumption made at every extrapolation is that no change of metallurgical microstructure in the extrapolation space occurs. By applying proposed constitutive equations at every described parameter, a master curve is obtained (Fig. 1).

Usually, a second degree polynomial sufficiently describes the master curve [13]. Coefficients used for every parameter are obtained by the leastsquares method and are presented in the tables for

Fig. 4. Creep-rupture test results [11]; one set of data is used for extrapolation purposes, four sets of experimental data are used for comparison purposes Creep Damage Calculation for Thermo Mechanical Fatigue

373

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 371-378

the data used. Master curves for the Larson-Miller parameter for four test temperatures are shown in Fig. 5. They are obtained exclusively from the data used for extrapolation (Fig. 4). However, it is possible to predict time to rupture for every stress and temperature.

4.1 Larson-Miller Parameter The LM parameter [14] assumes that the logarithm of the time to rupture is inversely proportional to temperature.

LM = T(logtr + C), (2)

log tr = −C +

1 a0 + a1 log σ + a2 log 2 σ . (3) T

(

)

Table 3. MH coefficients for used test results logta -29.9

Ta 2242

a0 -0.023

a1 -0.006

a2 0.003

4.4 Orr-Sherby-Dorn Parameter The OSD parameter [17] assumes that the logarithm of the time to rupture is equal to the reciprocal value of temperature and risen for a constant.

OSD = log tr −

log tr =

A , T

(8)

A + a0 + a1 log σ + a2 log 2 σ . (9) T

Table 4. OSD coefficients for used test results Table 1. LM coefficients for used test results C 19

a0 18123

a1 6715

A 19535

a2 -3264

a0 -22.4

a1 9.4

a2 -4.2

4.2 Manson-Brown Parameter The MB parameter [15] assumes that the logarithm of the time to rupture is a power function of temperature.

MB =

log tr − log ta , (4) q (T − Ta )

log tr = log ta +

(

(5)

)

q

Table 2. MB coefficients for used test results logta -36.4

Ta q 666.5 -0.143

a0 79.9

a1 18.2

a2 -8.4

4.3 Manson-Haferd Parameter The MH parameter [16] assumes that the logarithm of the time to rupture is proportional to temperature.

MH =

Fig. 5. Master curve determination with the LM parameter; coefficients are gained exclusively from the data used for extrapolation

+ (T − Ta ) a0 + a1 log σ + a2 log 2 σ .

log tr = log ta +

(

log tr − log ta , T − Ta

(6)

)

+ (T − Ta ) a0 + a1 log σ + a2 log 2 σ .

374

(7)

4.5 Comparison of Time-Temperature Parameters Time-temperature parameters are compared with the goodness of master curves’ fit for four comparison sets. Each comparison set has actual time to rupture values xai (Fig. 4).

Predicted time to rupture values xpi are calculated from the data used for extrapolation (Fig. 4). E[xa] is the mean of actual time to rupture data for each comparison set. The residual sum of squares:

RSS =

∑ (x i

ai

)

2

− xpi , (10)

is compared to the total sum of squares:

Šeruga, D. – Fajdiga, M. – Nagode, M.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 371-378

TSS =

∑ (x

ai

i

− E [ xa ]) , (11) 2

for each comparison set. By comparing the values RSS and TSS, a value for goodness of fit M can be introduced, denoted by: M = 1 − RSS = 1 − TSS

∑ (x − x ) ∑ ( x − E [ x ]) 2

ai

i

i

pi

ai

2

. (12)

a

If the predicted results are compared to the data used for extrapolation the M value will coincide with the R2 value [18]. However, the obtained R2 value could be negative for two different sets of data and its validity is therefore questionable (i.e. comparison set 1 and data used for extrapolation Fig. 4). The M value cannot be used to predict whether a model is good or not.

But since it is the same measure for all compared parameters it gives an insight into which parameters are better than others. If the M value is near 1, the model seems to be acceptable for that comparison set. If it is 0 or even negative the model appears to be useless in that particular area.

Goodness of fit calculated by the M value is given numerically in Table 5 and graphically in Fig. 6.

the OSD parameter lies somewhere in between.

The M value shows that the MB parameter is good only if the stress area is being extrapolated (comparison sets 1 and 2). If the MB parameter is used also for extrapolating results in the temperature area, the actual and predicted results deviate significantly (comparison sets 3 and 4).

For extrapolation purposes a set of data at two highest temperatures and three highest stress levels were chosen. By this choice the creep-rupture testing would be the shortest. It can be proven that two stress levels or one test temperature alone are not sufficient for a reliable prediction of the time to rupture at unknown temperatures and unknown stress levels. If more test temperatures and more stress levels were taken into account for assessment of the time-temperature parameters the matching between actual and predicted times to rupture would improve. Despite less conservative predictions with the LM parameter its usage is more satisfactory as it is still one of the most used time-temperature parameters for determining the master curves [19] and [20].

Table 5. Calculated M values for comparison sets with different time-temperature parameters T [ºC]

650

600

550

500

Par.

Comp. set 1

Comp. set 2

Comp. set 3

Comp. set 4

0.936 0.929 0.931 0.918

0.922 0.873 0.887 0.796

0.718 -1.871 0.951 0.872

0.810 -109.2 0.201 0.847

LM MB MH OSD

Master curves gained by different timetemperature parameters look very similar at first sight but a detailed analysis (in this case with the M value) reveals a significant difference in assessment of the time to rupture. Assessment of the coefficients represents a significant challenge regardless of which time-temperature parameter is used.

The M value is comparable for LM-, MHand OSD parameters (comparison sets 1 to 4). It can be noticed that the master curves are the most conservative when using the MH parameter, the least conservative when using the LM parameter;

Fig. 6. Calculated M values for comparison sets with different time-temperature parameters (negative values are plotted as 0) 5 DETERMINATION OF COEFFICIENTS Since the values of unknown coefficients for the time-temperature parameters are so different (Tables 1 to 4) and their calculation can be time consuming it is reasonable to introduce

Creep Damage Calculation for Thermo Mechanical Fatigue

375

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 371-378

a general form of the master curve equation.

A second order polynomial is sufficient for the description of the master curve, thus the time to rupture using the LM parameter (Eq. (3)) can be written in the general form:

log tr = b0 + b1 log σ + b2 log 2 σ , (13)

where

b0 = −C +

a0 , T

b1 =

a1 T

and

b2 =

a2 .

T

Coefficients b0, b1 and b2 can be obtained easily by drawing a trend line through the test data in any commercial software. The presented software calculates C, a0, a1 and a2 with the least-squares method;

Fig. 7. Master curves with a temperature step ΔT = 10 ºC over the whole test area (500 to 650 ºC)

2

a0    log trij + C − T −  n m j 2   , (14) S = ∑∑   a a i =1 j =1 2 1 2  − log σ i + log σ i  Tj  Tj 

where n and m are the number of stress levels and the number of test temperatures, respectively.

If Eq. (13) is taken into account Eq. (14) can be rewritten in the form: 2

 b0 j + b1 j log σ i + b2 j log 2 σ i + C −    S 2 = ∑ ∑  a0 a1 a  . (15) − log σ i + 2 log 2 σ i i =1 j =1  −  Tj  Tj Tj  n

Equalling partial derivatives of Eq. (15) with zero:

∂S 2 = 0; ∂C

∂S 2 = 0; ∂a0

∂S 2 = 0; ∂a1

∂S 2 = 0; (16) ∂a2

the LM coefficients C, a0, a1 and a2 are obtained.

The same procedure may be used for every presented time-temperature parameter. Master curves with a temperature step ΔT = 10 ºC over the whole test area (500 to 650 ºC) are depicted in Fig. 7.

6 EXAMPLE OF CREEP DAMAGE CALCULATION Creep damage calculation with Robinson’s damage accumulation rule and master curves determined by the LM parameter (Fig. 7) is performed on a simple temperature-stress history (Fig. 8).

376

Fig. 8. A simple temperature-stress history

m

The loading temperature is kept at 650 ºC at the beginning, decreases to 550 ºC after 150 h and increases back to 650 ºC after 200 h. Tensile loading stress rises to 10 MPa and remains constant for 150 h, then it turns into compressive direction with the same value. Creep damage calculation is performed online between successive time steps. Every time step is divided into smaller time sub steps. Their sizes depend on the stress or temperature jump in the actual time step. Creep damage is calculated as integration over all time steps and is depicted in Fig. 9. The thick line shows creep damage changing and the thin line shows the fatigue damage growing. The creep contribution to the damage of the thermo mechanically loaded component can be seen easily from the 20th to the 100th hour where the fatigue damage remains constant due to constant loading conditions whereas creep damage constantly grows. From the 150th hour on, the influence of the creep relation rule for the compressive stresses can be noticed. Here, the creep damage can increase, decrease or remain constant.

Šeruga, D. – Fajdiga, M. – Nagode, M.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 371-378

Fig. 9. Damage calculation for the simple temperature-stress history 7 SUMMARY AND CONCLUSIONS The developed software makes separate and quick calculations of the creep damage possible if temperature dependent material parameters and temperature-stress history are known. For determination of the material parameters only a few standard creep-rupture tests at different test temperatures and different stress levels are required. Extrapolation of the material parameters is possible using any of the introduced time-temperature parameters. The input material data are always coefficients for every known master curve and in the software they are appropriately transformed regarding the selected time-temperature parameter. The next step in the development of the creep damage calculation will be a search for a generalized time-temperature parameter which would embrace the introduced time-temperature parameters into a single form.

Once achieved, the creep damage calculation will be compared to the actual creep damage of the specimens at controlled loading conditions. The introduced programme code is also integrated into the commercial software LMS Virtual.Lab. 8 REFERENCES [1] Granacher, J., Scholz, A., Möhlig, H. (2000).

Behaviour of heat resistant power plant steels undergoing variable long term loading conditions. Mat.-wiss. u. Werkstofftech., vol.

31, p. 29-37.

[2] Lemaitre, J., Chaboche, J.L. (2000).

Mechanics of solid materials. Cambridge University Press, Cambridge.

[3] Wiswanathan, R. (1989).

Damage Mechanisms and Life Assessment of High-Temperature Components.

ASM International.

[4] Kropiwnicki, J., Hack, M. (2006). Improved calculation of damage due creep by more accurate time to rupture data presentation.

Progressive Technologies, Engines and Mechanisms in Mechanical Engineering: Scientific Journal of the International Baltic Assocciation of Mechanical Engineers, Kaliningrad.

[5] Robinson, E.L. (1938). Effect of temperature variation on the creep strength of steels.

Trans. ASME, vol. 160, p. 253-259.

[6] Nagode, M., Hack, M. (2004). An online algorithm for temperature influenced fatigue-life estimation: stress-life approach.

Int. J. Fatigue, vol. 26, p. 163-171.

[7] Sabour, M.H., Bhat, R.B. (2008). Lifetime prediction in creep-fatigue environment.

Materials Science Poland, vol. 26, p. 563584.

[8] Ainsworth, R.A., Budden, P.J. (1994).

Design and assessment of components subjected to creep. Journal of strain analysis, vol. 29, p. 43-50.

[9] Wada, Y., Aoto, K., Ueno, F. (1996). Creepfatigue evaluation method for type 304 and 316FRSS. International Atomic Energy Agency, p. 750-086.

[10] Jaske, C.E., Mindlin, H., Perrin, J.S. (1973).

Combined low-cycle fatigue and stress relaxation of alloy 800 and type 304 stainless steel at elevated temperatures. Fatigue at Elevated Temperature, ASTM STP 520, p.

365-376.

[11] N. I. for Material Science, Creep Database.

Retrieved on 15.9.2008 from https://tsuge. nims.go.jp/top/creep.html.

[12] Reti, T., Felde, I., Grum, J., Colas, R., Sarmiento, G.S., Moita de Deus, A. (2010).

Extension of isothermal time-temperature parameters to non-isothermal conditions: Application to the simulation of rapid tempering. Strojniški vestnik - Journal of Mechanical Engineering, vol. 56, no. 2, p.

84-92.

[13] Manson, S.S., Mendelson, A. (1959).

Optimization of parametric constants for

Creep Damage Calculation for Thermo Mechanical Fatigue

377

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 371-378

creep-rupture data by means of least squares.

NASA MEMO 3-10-59E. [14] Larson, F.R., Miller, E.J. (1952). Timetemperature relationship for rupture and creep stresses. Trans. ASME, vol. 74, p. 765775.

[15] Manson, S.S., Brown, W.F. (1953). Timetemperature stress relations for correlation and extrapolation of stress rupture data.

Proc. ASTM, vol. 53, p. 683-719.

[16] Manson, S.S., Haferd, A.M. (1953). A linear time-temperature relation for extrapolation of creep and stress rupture data. NACA TN 2890.

[17] Orr, R.L., Sherby, O.D., Dorn, J.E. (1954).

Correlations of rupture data for metals at elevated temperatures. Trans. ASM, vol. 46, p. 113-118.

378

[18] Motulsky, H., Christopoulos, A. (2004).

Fitting models to biological data using linear and nonlinear regression: a practical guide to curve fitting. Oxford University Press, Oxford.

[19] Kandare, E., Feih, S., Lattimer, B.Y., Mouritz, A.P. (2010). Larson–Miller failure modeling of aluminum in fire. Metallurgical and materials transactions A, vol. 41, p.

3091-3099.

[20] Bueno, L.O., Sordi, V.L. (2008). Creep behaviour of Fe-Mn-Al steel from 500 C to 800 C. Part 2: Aspects of rupture strength and parametric analysis. Materials Science and Engineering A, vol. 483-484, p. 560563.

Šeruga, D. – Fajdiga, M. – Nagode, M.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 379-384 DOI:10.5545/sv-jme.2010.142

Paper received: 02.06.2010 Paper accepted: 29.10.2010

Surface Integrity of Shot Peened Aluminium Alloy 7075-T651 Zupanc, U. – Grum, J. Uroš Zupanc1, Janez Grum2,* 1 Slovenian Welding Institute, Slovenia 2 University of Ljubljana, Faculty of Mechanical Engineering, Slovenia

The objective of the present study was to investigate the effect of surface hardening by shotpeening (SP) on fatigue properties of high-strength aluminium alloy 7075-T651. The paper describes the effects of SP treatment by presenting analyses of surface roughness measurement, microhardness profiles, microstructure changes, residual stresses and material bending fatigue resistance. The obtained results show a favourable influence of SP treatment on fatigue properties as induced compressive residual stresses and hardened surface layer retarded the initiation of fatigue cracks. SP treatment nearly doubled the cycles to failure at the higher applied stresses when compared to the untreated specimens. The fatigue limit of the SP-treated specimens increased to 218 MPa at 107 cycles. The experimental data confirmed an increase of fatigue strength after SP treatment due to the compressive residual stress ability to influence fatigue crack nucleation. Increased resistance to plastic deformation and the residual stress profiles provided a corresponding fatigue crack closure. ©2011 Journal of Mechanical Engineering. All rights reserved.

Keywords: aluminium, shot peening, surface roughness, residual stresses, fatigue testing 0 INTRODUCTION Shot peening (SP), i.e. bombarding a surface with spherical shot or beads, is a surface treatment process aimed at increasing material’s fatigue strength. Intense elastoplastic deformation in the surface layer increases material fatigue properties by strain hardening an inducing favourable compressive residual stresses [1].

Modern approaches using FEM made it possible to understand interactions if key SP parameters on actual residual stress variations in hardened surface layer [2]. To achieve optimum SP treatment, monitoring of process parameters is essential [3] and [4]. A variety of inspection and research applications were also performed on SP-treated structural high-strength aluminium alloys as these alloys show a favourable ratio of its mechanical properties to its specific weight [4] to [7]. SPtreated specimens made of precipitation hardened aluminium alloys showed better resistance to fatigue crack initiation by a factor of 1.2 to 6 [8] to [11]. In fatigue testing an analysis of surface integrity after SP is important to provide optimum strain hardening properties. The objective of the present study was to investigate the surface roughness properties and microstructural changes on fatigue properties of SP treated high-strength

aluminium 7075-T651. The effects of surface elastoplastic deformation by SP were analysed and quantified.

1 EXPERIMENTAL PROCEDURE 1.1 Material A wrought plate of high-strength, precipitation hardened, aluminium alloy 7075T651 of 20 mm in thickness was delivered with the chemical composition (in wt. %): Al-5.78Zn2.56Mg-1.62Cu-0.21Cr-0.05Mn-0.04Ti-0.09Si0.18Fe. Mechanical properties of the tested material were: Rm = 585 MPa, Rp0.2 = 532 MPa and A50 = 12%. Specimens for fatigue testing were prepared in a long traverse (LT) direction (Fig. 1). As-machined specimens were ultrasonically cleaned in ethanol.

Fig. 1. Fatigue specimen details and dimensions in [mm]

*Corr. Author’s Address: University of Ljubljana, Faculty of Mechanical Engineering, Aškerčeva 6, 1000 Ljubljana, Slovenia, janez.grum@fs.uni-lj.si

379

StrojniĹĄki vestnik - Journal of Mechanical Engineering 57(2011)5, 379-384

1.2 SP Treatment The specimens were SP-treated from all sides at the Metal Improvement Company in Germany using an air-blast machine. Cast steel shot MI-170H with a hardness of 55 HRC and a nominal diameter of 0.40 mm was chosen.

In order to avoid medium collision, the angle of nozzle inclination was shifted by 10° with regard to the vertical axis. A constant specimen distance from the nozzle of around 120 mm was maintained. Surface coverage was set to 150%.

SP intensity was quantified by means of the standardized Almen measurement. The residual compressive stress from the on-side peening causes the standardized Almen strip to bend or arc convexly towards the peened side. The Almen strip arc height is a function of the energy of the shot stream and is very repeatable. Other details of Almen intensity measurements can be found in [12] and [13]. By different settings of air pressure, mass flow and a nozzle distance two comparative Almen intensity values of 8A and 12A were achieved.

thin hardened layer. Three different research combinations were evaluated: a) prior to SP treatment (as-machined specimens); b) SP treatment in Almen intensity 8 A; and c) SP treatment in Almen intensity 12 A. Measurement of high-resolution surface roughness was made with a Taylor Hubson Form Talysurf Series 2 device. The residual-stress measurements were made with a semi destructive hole-drilling method in accordance with ASTM 837 [14].

Bending fatigue testing of the specimens was carried out with a Rumul Cracktronic device at room temperature. A constant amplitude bending stress was applied in the range of the maximum applied stresses, i.e. those ranging between 15 and 65% of delivered-material tensile strength Rm. The testing resonant stress frequency was 107 Hz using a sinusoidal waveform at a stress ratio R of 0.05. A criterion of specimen failure was a drop of inherent oscillation by more than 3%, where fatigue cracks occurred in a depth of up to 4 mm. In the present study a run-out criterion as a limit of fatigue strength was set at 10 million cycles. Fractured surfaces of all fatigued specimen were further evaluated using a scanning electron microscope (SEM).

2 EXPERIMENTAL RESULTS 2.1 Surface Roughness

Fig. 2. Surface profiles of SP-treated specimens 1.3 Surface Integrity Characterization The evaluation of the tested specimens comprised surface properties, microscopic analysis and residual-stress measurement in the 380

The surface conditions after intense plastic deformation of the hardened material can be described by surface roughness changes, which depend on the chosen SP parameters. Data on the surface roughness properties are important in predicting material fatigue resistances as roughen areas might represent local stress concentrations.

In Fig. 2 comparative surface roughness rofiles are presented. Prior SP treatment of specimens was grinding using emery paper of #1000 (Fig. 2a) showing nearly uniform surface properties.

The SP treated specimens were affected by shot indentations and material elastoplastic flow along the shot during impact period and material elastic relaxation after impact. Because of a degree of coverage of 150% the traces of SP dimples got partially blurred.

Key measured surface roughness properties within the assessment length of 2

Zupanc, U. â&#x20AC;&#x201C; Grum, J.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 379-384

mm, i.e., calculated arithmetic average surface roughness (Ra), an average peak to valley height (Rt) and Dp the average peak to peak distance (Dp) are presented in Table 1. In order to evaluate fatigue life in it it is important to evaluate surface distortion. Changes in surface roughness due to SP treatment increased notch sensitivity of the treated material. Geometrical notch stress concentration factor (Kt) due to surface roughness can be estimated as in [15]: Kt = 1 + 4(Rr / Dp)1.3. (1)

Both SP intensities show similar stress concentration values Kt (Table 1).

Table 1. Surface roughness properties

Prior to SP

Ra [µm] 0.32

Rt [µm] 2.59

Dp [µm] -

Kt Eq. (1) -

Almen 8A

4.57

30.5

193.12

1.36

Almen 12A

5.81

40.4

217.20

1.45

Treatment

2.2 Microscopic Analysis In the analysis of the surface condition after SP treatment, microscopic cross-section examinations of specimens were performed and are presented in Fig. 3. For etching Keller’s reagent was used.

Fig 3. Microstructure cross-sectional analysis; a) specimen prior SP treatment, b) SP-treated specimen – Almen 12A, c) marked area at higher magnification Microsectional image of as-received material showing a lined-up orientation of precipitated phases due to rolling is presented for further comparison (Fig. 3a). Light gray rounded irregular particles of Al2CuMg eutectic

phase and darker globular MgZn2 eutectic particles were evaluated [16]. On SP treated specimens a distinctive material flow in the surface layer could be observed. Changes in the orientation of the Al2CuMg phases precipitated in the aluminium matrix were noticed (Fig. 3b).

Individual local areas of the specimens treated at the maximum peening intensity, i.e. 12A, were subject to excessive plastic deformation. As a critical local plastic deformation was exceeded, cracks occurred in the surface layer (Fig. 3c).

The presence of such surface defects can exert an unfavourable influence on a dynamically loaded material as such local stress concentrations often initiate fatigue cracks. Higher SP intensities are not recommended. The defects appearing at the surface could be, assuming the same peening intensity, alternatively alleviated by using larger shots. Contamination of treated material due to inadequate shot precipitation during the operation may occur as well. Shot residual may be introduced into the specimen surface as a foreign body, which can in case of using steel shots result in lower corrosion resistance. Surface image (Fig. 4a) and corresponding microscopic cross-section of the contaminated SP-treated aluminium (Fig. 4b) are presented. Dimensions of the steel residual are of the order of magnitude of 30 × 50 µm2.

Critical depths of the foreign bodies also present a risk of the initiation of fatigue cracks. With peening intensities of 8 A the residuals penetration depth was up to 30 µm, and at maximum intensity of 12 A the residuals penetration depth even up to 50 µm. An energy dispersive X-ray spectroscopy (EDS) analysis of the chemical elements present at the marked by an arrow in Fig. 4b is given in Table 2. An area with a predominating Fe fraction representing the foreign inclusion of SP steel media was confirmed. a)

b)

Fig 4. Inclusion of steel shot; a) after SP treatment (Almen 12A) at surface, b) in corresponding cross-section analysis

Surface Integrity of Shot Peened Aluminium Alloy 7075-T651

381

StrojniĹĄki vestnik - Journal of Mechanical Engineering 57(2011)5, 379-384

Table 2. The EDS analysis of the chemical elements at contaminated surface Element Concentration [in wt. %]

Al 1.69

2.3 Microhardness Through-depth microhardness measurements of the thin surface layer shown in Fig. 5 were performed to determine hardness variation profiles. After SP treatment, hardness increased by 20 to 25% with reference to the hardness of the as-received material was observed.

With the SP intensity of 12 A, maximum values of microhardness after SP treatment ranged between 215 and 220 HV0.3.

With the SP intensity of 8A, microhardness was lower due to a lower Hertzian pressure at the impact area. Thus, the maximum values ranged between 200 and 205 HV0.3. A depth of the hardened layer can be inferred from a hardness variation, a starting point being hardness of the substrate. After SP treatment of the surface, the measured depth of the hardened layer ranged around 350 mm.

Si 1.38

Ca 1.76

Mn 2.15

Fe 91.98

Zn 1.04

Prior to SP treatment the as-machined specimens showed residual stresses in the surface layer amounting to around Âą50 MPa, induced most probably due to specimen preparation.

A relatively small-magnitude measured stress of the as-machined specimens was neglected in further evaluation. Maximum compressive residual stresses at the surface ranged between -200 and -165 MPa resp. depending on the peening intensity used.

Fig 6. Comparison of residual stress profiles using different SP intensities

Fig. 5. Microhardness variations in cross section of SP-treated specimens 2.4 Residual Stresses The residual-stress variations in the thin surface layer were used to analyse the depth of elastoplastic deformation due to SP treatment.

Maximum residual stresses obtained at the surface layer have an important influence on material fatigue properties. Fig. 6 shows calculated residual-stress variations in dependence of depth.

382

Fig. 7. Fatigue life for SP-treated aluminium 7075-T651 With the higher intensity stronger residualstress relief could be noticed at the surface possible due to Bauschinger effect [1]. Higher intensity resulted in increased residual stress values and higher depths. With the intensity of Almen 8A a maximum value of residual stresses

Zupanc, U. â&#x20AC;&#x201C; Grum, J.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 379-384

after SP treatment amounted to -294 MPa, and with the intensity of Almen 12 A to -312 MPa. The maximum residual stresses occurred at a depth of 100 μm. With regard to the mechanical properties of the delivered 7075-T651 aluminium alloy, the induced residual stresses after SP treatment ranged around 50% of the material tensile strength (Rm). Furthermore, different gradients of throughdepth residual-stress increases could be noticed.

With the intensity of 12 A calculated comparative gradients of the residual-stress increase to the depth of the maximum stresses amounted to 1.47 MPa/µm (147 MPa/100 µm) and with the intensity of 8 A to 0.94 MPa/µm (94 MPa/100 µm). The depths of the induced residual stresses measured up to 400 µm.

2.5 Bending Fatigue Testing In dynamic loading, material fatigue occurs at loads considerably lower than tensile strength of the material. Changing loads are related to fatigue cracks at the material surface, which influence the real lifetime of machine components. The semilogarithmic S-N curves generated for the fatigued specimens in different research combinations are shown in Fig. 7. Fatigue results for the asmachined specimens presented a baseline for further comparison.

A favourable influence of SP treatment on material fatigue resistance was found. SP treatment nearly doubled the maximal cycles to failure at the higher applied stresses when compared to baseline. The fatigue limit of the SP-treated specimens increased to 218 MPa at 107 cycles.

The experimental data confirmed an increase of fatigue strength of the SP-treated material due to the compressive residual stress ability to influence

a) Prior to SP (200 MPa)

fatigue crack nucleation. Strain hardening by SP retarded crack propagation. Increased resistance to plastic deformation and the residual stress profiles provided a corresponding fatigue crack closure and thus prolonged fatigue life.

In a material fatigue evaluation it is important to analyse spots of fatigue crack propagation. To evaluate crack initiation the fractured surfaces of fatigued specimens were examined. Typical SEM images of fracture surfaces in low and high applied bending stress regimes are shown in Fig. 8. Crack initiation sites are marked by an arrow. On as-machined specimen fatigued at a maximal applied stress of 200 MPa crack nucleation was found at ~50 µm below the surface (Fig. 8a). Fatigue crack initiations of SP treated specimens were observed at much greater depths compared to the as-machined specimens.

Typical fatigue crack initiation depths of SP treated specimens were found between 250 and 370 µm below the surface, depending on applied fatigue stresses (Fig. 8b and c). The compressive residual stress layer pushed the crack region deeper beneath the surface. The subsurface fatigue bending stress exceeded the critical values below the surface hardened layer, so fatigue cracks initiated at much greater depths.

3 CONCLUSIONS To determine the effects of fatigue properties of the SP-treated aluminium alloy 7075-T651, a series of tests were performed. The research results demonstrate a positive effect of the SP treatment of structural elements exposed to dynamic loads. In order to provide a higher fatigue strength of the material, adequate surface integrity after SP should be ensured. Based on the

b) Almen 8A (219 MPa)

c) Almen 12A (278 MPa)

Fig. 8. Typical fracture surfaces of fatigue-tested as-machined and SP treated specimens Surface Integrity of Shot Peened Aluminium Alloy 7075-T651

383

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 379-384

study of the influence of the SP treatment on the fatigue resistance of aluminium alloy 7075-T651 the following conclusions can be drawn; a) after SP treatment the increase of surface roughness should be expected. Possible surface cracks and SP medium residual introduced into the surface may ocour and represent a potential risk of fatigue crack initiation under dynamic loads; b) a favourable influence of SP treatment on material increased fatigue resistance was found. SP treatment nearly doubled the cycles to failure at the higher applied stresses when compared to the untreated specimens. The fatigue limit of the SP-treated specimens increased to 218 MPa at 107 cycles; c) the experimental data confirmed an increase of fatigue strength after SP treatment due to the compressive residual stress ability to influence fatigue crack nucleation.

Increased resistance to plastic deformation and the residual stress profiles provided a corresponding fatigue crack closure. The compressive residual stress layer also pushed the fatigue crack region deeper beneath the surface.

4 REFERENCES [1] Schulze, V. (2006). Modern mechanical surface treatment. Wiley-VCH Verlag GmbH, Weinheim.

[2] Hong, T., Ooi, J.Y., Shaw, B. (2008). A numerical simulation to relate the shot peening parameters to the induced residual stresses. Eng. Fa. Anal., vol. 15, p. 10971110.

[3] George, P.M., Pillai, N., Shah, N. (2004).

Optimization of shot peening parameters using Taguchi technique. J. of Mat. Proc.

Techn., vol. 153-154, p. 925-930.

[4] Guagliano, M. (2001). Relating Almen intensity to residual stresses induced by shot peening: a numerical approach. Mat. Proc.

Tech., vol. 110, p. 277-286.

[5] Grum, J. (2008). Surface integrity after shot peening applied to a precipitation hardened aluminium alloy. Metal Finishing News, vol.

9, p. 54-56.

384

[6] Žagar, S., Grum, J. (2011). Surface integrity after mechanical hardening of various aluminium alloys. Strojniški vestnik Journal of Mechanical Engineering, vol.

57, no. 4, p. 334-344, DOI:10.5545/svjme.2010.092.

[7] Zupanc, U., Grum, J. (2010). Effect of pitting corrosion on fatigue performance of shot-peened aluminium alloy 7075-T651. J. Mat. Proc. Tech., vol. 9, p. 1197-1202.

[8] Sharp, P.K., Clark, G. (2001). The effect of peening on the fatigue life of 7050 aluminium alloy. Reserch Report DSTO-RR-0208. from: http://hdl.handle. net/1947/3292, accesed on 2010-10-14.

[9] Rodopoulos, C.A., Curtis, S.A., Rios, E.R., SolisRomero, J. (2004). Optimization of the fatigue resistance of 2024-T351 aluminium alloys by controlled shot peeningmethodoloy, results and analysis. Int. J. Fatigue, vol. 26, p. 849-856.

[10] Benedetti, M., Fontanari, V., Scardi, P., Ricardo, C.L.A., Bandini, M. (2009).

Reverse bending fatigue of shot peened 7075-T651 aluminium alloy. The role of residual stress relaxation. Int J Fatigue, vol.

31, p. 1225-1236.

[11] Benedetti, M., Bortolamedi, T., Fontanari, V., Frendo, F. (2004). Bending fatigue behaviour of differently shot peened Al 6082 T5 alloy. Int. J. Fatigue, vol. 26, p. 889-897.

[12] AMS-S-13165 (1997). Shot-peening of Metal Parts. SAE, Warrendale.

[13] Shot Peening Applications, 9th edition (2005). Metal Improvement Company, Paramus.

[14] ASTM Standard E837-08. (2008). Standard test method for determining residual stresses by the hole-drilling strain-gage method.

ASTM International, West Conshohocken.

[15] Li, J.K., Mei, Y., Wang, D., Wang, R. (1992). An analysis of stress concentration caused by shot peening and its application in predicting fatigue strength. Fatigue Fract.

Eng.Mater Struct., vol. 152, p. 1271-1279.

[16] Voort, G.F.V. (2004). ASM handbook: Volume 9: Metallography and microstructures. ASM International, West Conshohocken.

Zupanc, U. – Grum, J.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 385-393 DOI:10.5545/sv-jme.2010.119

Paper received: 25.05.2010 Paper accepted: 06.10.2010

Surface Modification of Aluminium Alloys with Laser Shock Processing Trdan, U. – Ocaña, J.L. – Grum, J. Uroš Trdan1 – José Luis Ocaña2 – Janez Grum1,* 1 University of Ljubljana, Faculty of Mechanical Engineering, Slovenia 2 Centro Láser U.P.M., Universidad Politécnica de Madrid, Spain

An adequate residual stress variation and frequently also improved corrosion resistance of a material are key requirements for usability of numerous machine components in various applications. The aim of the investigation conducted was to determine optimum Laser Shock Processing (LSP) parameters for aluminium specimens in order to obtain the desired residual stress variation and improved corrosion resistance. LSP treatment was performed with a Q-switched Nd:YAG laser with a wavelength of 1064 nm. In order to statistically confirm the optimum process parameters, a factorial design was applied, in which the first experimental factor was pulse density, i.e. 900 and 2500 pulses/cm2, the second factor was the type of material used, i.e. aluminium alloys AlMgSiPb and AlSi1MgMn and the third factor was the direction of LSP surface sweep, i.e. longitudinal and transversal direction.

The experiments made confirmed a characteristic influence of the first factor representing different pulse densities. An analysis of residual stresses confirmed that in processing with 2500 pulses/cm2 the highest compressive residual stresses were obtained. Potentiodynamic corrosion testing confirmed that the higher pulse density resulted in a stronger shift of pitting potential, which provided higher corrosion resistance. ©2011 Journal of Mechanical Engineering. All rights reserved.

Keywords: Laser Shock Processing, surface roughness, residual stresses, microhardness, pitting corrosion, analysis of variance 0 INTRODUCTION Laser Shock Processing (LSP) is an innovative surface treatment, with which mostly a Q-switched Nd:YAG laser with short pulses of several ns and with a power density, in the pulse peak, of as much as several tens of GW/ cm2 is used. In contrast to conventional Shot Peening (SP), which provides kinetic energy of hard particles [1] and [2], LSP is based on plasma generation at the moment of the interaction of laser light with a workpiece material, which produces shock impact waves and elasto-plastic shifts of atomic planes in the material [3] and [4].

Due to the shock waves generated, the dislocation density considerably increases.

Consequently, fatigue resistance also notably increases.

Zhang and Yao [5] applied LSP to different types of steels, aluminium alloys, and titanium alloys. LSP produces shocks of motive quantity, which produce considerable densification of dislocations and generation of compressive

residual stresses of high gradient [4]. In practice, technologists and engineers frequently require the introduction of compressive residual stresses since it improves the fatigue resistance of a material [6] and [7].

Commercially available laser sources and major advancement of laser engineering have permitted various industrial applications of LSP. LSP is frequently applied to exacting components, particularly in the aircraft industry for treatment of the most demanding components such as turbine spades of an aircraft F-16 Falcon and a bombardier F-22 Rockwell [8].

In numerous studies Sano et al. [9] confirmed the applicability of LSP, particularly because of improved material resistance to stresscorrosion cracking (SCC). LSP was carried out at specimens made of stainless steel SUS304 having a size of 10 × 50 × 2 mm, pulse duration being 8 ns and a degree of overlapping of 7000 pulses/cm2. Corrosion tests carried out in a vapour chamber for 500 hours confirmed an influence of LSP on increased material resistance to SCC.

*Corr. Author’s Address: University of Ljubljana, Faculty of Mechanical Engineering, Aškerčeva 6, 1000 Ljubljana, Slovenia, janez.grum@fs.uni-lj.si

385

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 385-393

The same specimens were also used to perform a microscopic analysis, which confirmed the presence of cracks in the specimens in the asdelivered state whereas in the LSP-treated specimens no cracks were found.

Hong and Chengye [10] studied the effects of individual LSP parameters such as laser-beam power density, laser-spot size, pulse duration, and pulse density, i.e. degree of overlapping.

It was found that a correct choice of the processing parameters provided desired variations of hardness and residual stresses in the surface of machine components which, in turn, provided improved material fatigue resistance.

1 EXPERIMENTAL The aim of the investigation conducted was an analysis of the influence of pulse density per area unit of LSP of two types of aluminium alloys with reference to the surface profile and roughness obtained, and variations of residual stresses and microhardness. Furthermore, an analysis of a change of corrosion resistance with potentiodynamic corrosion testing and confirmation of an improved surface condition on SEM followed.

In order to determine the optimum LSP parameters, a statistical evaluation of experimental data with the factorial design was chosen. The factorial design is particularly useful when an influence of at least two factors to an output response of the experimental process is treated [11]. Grum and Slabe [12] confirmed the factorial design as a suitable method for a rapid choice of optimum heat-treatment conditions with lasersurfaced specimens.

1.1 Experiment and Factorial Design Fig. 1 shows a block diagram of the experimental process and the responses concerned.

The comprehensive analysis of the specimens is to provide the most favourable surface integrity condition.

An evaluation of the LSP parameters chosen was performed with following response characteristics: • Specimen surface profile prior to and after processing expressed by an average arithmetic surface roughness - Ra. • A profile of minimal principal residual rs stresses ( σ min ) in the thin surface layer determined by the hole drilling relaxation method • Microhardness variation in the thin surface layer of the specimen material – HV0.2.

• Corrosion resistance expressed by a pitting potential and a number of pits formed at the material surface.

1.2 Material Selection For a comparison of the effect of shock waves after LSP two aluminium alloys, i.e.

AlMgSiPb and AlSi1MgMn, were chosen.

Alloys were in the precipitation-hardened state T-651. The alloys were subjected to preliminary homogenization at a temperature of 540 °C, then quenched to ambient temperature, subjected to tensile loading with a 2% strain, and subjected to artificial ageing at a temperature T of 160 °C for 10 hours.

The procedure chosen for the preparation of the alloy produces a large number of precipitates, i.e. intermetallic phases, which result in higher

Fig. 1. Block diagram of experimental process 386

Trdan, U. – Ocaña, J.L. – Grum, J.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 385-393

Table 1. Chemical composition of the investigated alloys (wt %) Alloy Mg AlMgSiPb 0.6-1.2 AlSi1MgMn 0.6-1.2

Si 0.6-1.2 0.7-1.3

Pb 0.4-2.0 -

Mn 0.4-1.0 0.4-1.0

material hardness and high ultimate tensile strength and yield stress (Rm = 350 MPa, Rp0.2 = 320 MPa).

Chemical compositions of the alloys concerned are given in Table 1.

After the T-651 treatment 8 mm thick discshaped specimens were cut from a cylindrical drawn rod having a diameter of 40 mm. In order to ensure the same initial state for all the specimens, cutting of the rod was carried out very carefully and under the same conditions.

An adequate specimen preparation thus prevented the surface overheating and a change of microstructure and mechanical properties. In this way specimens for LSP with initial residual stresses as small as possible were provided.

1.2 Laser Shock Processing Setup LSP treatment of the specimens was performed in the Centro Láser U.P.M. Ctra deValencia, Madrid, Spain with a Q-switched Nd:YAG laser with a wavelength λ of 1.064 μm and a power density of 10.75 GW/cm2. Two levels of pulse density (900 and 2500 pulses/cm2) were chosen, laser pulse duration tp of 10 ns being uniform with a repetition of 10 Hz.

Fig. 2. Presentation of LSP experimental setup Fig. 2 shows the presentation of LSP experimental setup, the specimen being clamped

Fe 0.5 0.5

Cr 0.3 0.25

Zn 0.3 0.2

Ti 0.2 0.14

Bi 0.7 -

Cu 0.1 0.1

in a movable computer-aided x-y table and submerged in water.

At the interaction of laser light with the material surface high-energy plasma will due to a high temperature generate at an extremely small surface area. Due to the confining medium, high pressure will occur at the specimen surface, which produces spreading of shock waves across the specimen.

The same state of treatment was provided under all processing conditions with a laserbeam sweep across the specimen surface. When processed with 900 pulses/cm2, overlapping pitch among individual pulses equalled 0.33 mm, and with 2500 pulses/cm2 0.22 mm.

2 RESULTS AND DISCUSSION 2.1 Surface Roughness Analysis Due to the preliminary preparation of the specimens with the cutter and LSP surface sweep direction, it was decided to establish surface profiles in the longitudinal (L) and transverse (T) directions, by average values of the mean arithmetic roughness Ra L and RaT and by three profile measurements. For a comparison and analysis of specimen surface roughness, a measuring length l m of 8 mm was chosen. Measurements were made with a profile meter Surtronic 3+, product of Taylor/Hobson Pneumo, using a Gaussian filter, cut-off 0.8 mm Fig. 3 shows topographic images of the surface that confirm an extreme dependence of surface roughness on LSP conditions. The increase in surface roughness is a consequence of numerous laser-beam interactions with the specimen surface due to the overlapping of tracks and cumulative action of shock waves at the interaction point.

The roughness Ra of the measured initial specimen, without laser peening amounts to 0.72 μm in the L direction and 0.81 μm in the T direction. At the specimen treated with 900 pulses/ cm2 Ra amounts to 3.74 μm in L direction and 6.0

Surface Modification of Aluminium Alloys with Laser Shock Processing

387

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 385-393

b) c) a) Fig 3. Topographic images and specimen surface profiles of alloy AlSi1MgMn; a) initial state, without LSP, b) LSP with 900 pulses/cm2 and c) LSP with 2500 pulses/cm2

μm in T direction. The highest roughness increase was noticed at the specimen, which was treated with 2500 pulses/cm2, i.e. Ra L = 5.36 μm and RaT = 9.11 μm.

From a comparison of the surface topographies it can be inferred that the specimen surfaces after LSP differ in crater sizes. With the higher pulse density, the size of the surface craters occurring ranges between 100 and 200 μm and is by factor 2 greater than with the lower pulse density, with which crater diameters range between 50 μm and 100 μm.

The calculated values of the mean arithmetic surface roughness Ra were verified also with an analysis of variance (Table 2). The analysis of variance confirmed that the pulse density and roughness profile measurement in the longitudinal (L) and transverse (T) directions exerted a significant influence on the surface condition after LSP. The influence of individual factors and their interactions is shown with a pie chart in Fig. 4. The influence of an individual factor is determined by: SS factor ⋅100 [ % ] , (1) Factor effect = SStotal − SSerror where SSfactor is the deviation square sum of the calculated factor [μm2] and SStotal is the total deviation square sum [μm2].

From Fig. 4 it can be inferred that the greatest influence on the final surface roughness Ra is exerted by pulse density, i.e. with an influence fraction of 76.23%; then follows the direction of roughness measurement with an influence fraction of 8.96%.

The smallest influence is exerted by the type of material, i.e. 1.51%, which is a very low fraction of influence.

388

A: Pulse density - 76.23% B: Material - 1.51% C: Direction - 8.96% AB Interaction - 1.04% AC Interaction - 6.76% BC Interaction - 0.3% ABC Interaction - 0.54% Error - 4.68%

Fig. 4. Pie chart showing influence of factors 2.2 Residual Stresses Analysis Knowledge of the residual-stress variation in the thin surface layer of a material gives an insight into material condition. In practice, instead of long-term material fatigue testing, measurement of residual stresses is preferred as high compressive residual stresses in the surface result in improved fatigue resistance of the material and better resistance to fatigue cracks.

Strain measurements and calculations of residual stresses in the surface layer were based on the relaxation hole-drilling method in accordance with the ASTM standard [13] and using measuring resistance rosettes CEA-06-062-UM and device RS-200 Milling Guide, Vishay Group.

Fig. 5 shows variations of the minimal principal residual stresses. From the residualstress variation it can be inferred that the values of the minimal compressive residual stresses (σmin) in the specimens prior to LSP are ideal since they amount to around 0 MPa.

Trdan, U. – Ocaña, J.L. – Grum, J.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 385-393

Table 2. Results of the three-factor analysis of variance of mean arithmetic roughness Ra Source of variation Pulse density (A) Material (B) Direction (C) Interaction AB Interaction AC Interaction BC Interaction ABC Error Total

νi [ / ] 2 1 1 2 2 1 2 24 -

SSi [μm2] 229.557 4.551 26.971 3.119 20.356 0.890 1.615 14.081 301.140

MSi [μm2] 114.778 4.551 26.971 1.560 10.178 0.890 0.808 0.587 -

σ min [MPa]

100 0 -100 -200 -300

AlSi1MgMn

-400 0

a)

0.2

0.4

0.6

Depth z [mm]

0.8

1

σ min [MPa]

b)

0 -100 -200 AlMgSiPb

-400 0

0.2

initial state 900 pulses/cm² 2500 pulses/cm²

0.4

0.6

Depth z [mm]

Fν,24, 0.01[ / ] 9.34 14.03 14.03 9.34 9.34 14.03 9.34 -

P[/] <0.0001 0.0103 <0.0001 0.0906 <0.0001 0.23 0.2716 -

0.033 mm whereas with th pulse density of 2500 pulses/cm2, they are a little higher, i.e. -337 MPa. A similar variation of the minimal principal residual stresses after LSP treatment can also be noted with aluminium alloy AlSi1MgMn. In this case with 900 pulses/cm2 the highest stresses obtained amount to -242 MPa, and with 2500 pulses/cm2 to as much as -317 MPa. 2.3 Microhardness Analysis

100

-300

F0 [ / ] 195.628 7.757 45.969 2.658 17.348 1.517 1.377 -

0.8

1

Fig. 5. Variation of minimal principal residual stresses (σmin); a) AlSi1MgMn, b) AlMgSiPb Such a variation confirms that the heat treatment T-651 is adequate for the initial material state. In alloy AlMgSiPb residual stresses range between -10 and +30 MPa and in alloy AlSi1MgMn between ± 20 MPa. The analysis of the principal residual stresses after LSP treatment of the specimens with a power density of 10.75 GW/cm2 in alloy AlMgSiPb confirmed that the influence of pulse density is important. In this alloy with the pulse density of 900 pulses/cm2 compressive residual stresses of -314 MPa are obtained in a depth of

The microhardness variations prior to and after LSP of aluminium alloys AlMgSiPb and AlSi1MgMn were measured using the Vickers method with a load of 200 g (HV0.2), as at least ten measurements are required to establish a suitable microhardness profile in the hardened layer and, consequently, a reliable microhardness variation.

Microhardness was measured with vertical and horizontal staggering between the two lines at a distance of a triple diagonal of an indentation between two adjacent indentations.

Fig. 6 shows through-depth microhardness variations in the hardened layer using different pulse densities. The microhardness variations are within the boundaries of expectation and confirm the LSP effects with regard to different pulse densities.

The highest microhardness value after LSP was measured at the specimen surface of alloy AlMgSiPb after treatment with 2500 pulses/cm2.

It amounted to 123 HV0.2.

After treatment with 900 pulses/cm2 microhardness of 119 HV0.2 was achieved at the specimen surface of alloy AlMgSiPb. In comparison to the untreated material (92 HV0.2), the microhardness increased by as little as 12% whereas after treatment with 2500 pulses/cm2

Surface Modification of Aluminium Alloys with Laser Shock Processing

389

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 385-393

Microhardness Microhardness HV HV0.2 0.2

120 110 100 90

Longitudinal

T ransversal

80 0

a)a)

125

250

375

500

Depth [µm] 130

Microhardness HV0.2 0,2

130

900 pulses/cm²

110 100 90

Longitudinal

T ransversal

80 0

125

c)

110 100 90

Longitudinal

250

375

0

500

125

250

375

500

Depth [µm] 2500 pulses/cm²

120 110 100 90

Longitudinal

80 0

d)

Depth [µm]

Transversal

80

130

900 pulses/cm²

120

2500 pulses/cm²

120

b) Microhardness HV0.2 0,2

Microhardness Microhardness HV0.2 0.2

130

125

T ransversal

250

375

500

Depth [µm]

Fig. 6. Microhardness variation in thin surface layer; a) AlSi1MgMn – 900 pulses/cm2, b) AlSi1MgMn – 2500 pulses/cm2, c) AlMgSiPb – 900 pulses/cm2, d) AlMgSiPb – 2500 pulses/cm2 hardness increased by as much as 19.8%. The same tendency of increasing microhardness at the surface was recorded in the same range also at the specimens of alloy AlSi1MgMn.

Microhardness HV 0.2

112 110 108 106 104 102

AlSi1MgMn AlMgSiPb

100 98 900

2500

Pulse density [pulses/cm²] Fig. 7. Charcteristic influence of factors and interactions

390

The analysis of variance of the measured microhardness was used to verify the influence of pulse overlapping density and material. Fig. 7 shows characteristic influences of the factors of pulse density and material type on the average value of microhardness.

The results show obvious nonparallelism, i.e. a line intersection, which represents a strong interaction of the factors. The diagram additionally confirms the analysis of variance, which states a significant influence of the interaction of factors AB. 2.4 Pitting Corrosion Analysis Aluminium alloys are frequently used in industrial applications due to their low density, and excellent corrosion resistance. In the presence of chloride ions the protective effect of a passive/ oxide film at the surface of aluminium alloys is drastically reduced, which results in serious corrosion damages in the form of small surface

Trdan, U. – Ocaña, J.L. – Grum, J.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 385-393

Potential [V SCE]

-0.4 -0,4 -0.8 -0,8

Material

Increase of the passivation potential (Epass)

-1.2 -1,2 -1.6 -1,6

AlMgSiPb AlMgSiPb

-2 -50

a)

Table 3. Pitting potentials from electrochemical corrosion polarization tests

Increase of Epitt with higher pulse density

-30

-10

10

30

AlSi1MgMn 50

2

Pulse density [pulses/cm2] No LSP 900 2500 No LSP 900 2500

Epitt [mVSCE] -1166 -947 -899 -782 -720 -662

Δ Epitt [mVSCE] 0 +219 +267 0 +62 +120

Current density [mA/cm ]

Potential [V SCE]

-0.4 -0,4 -0.8 -0,8 -1.2 -1,2

No LSP 900 pulses/cm² 2500 pulses/cm²

-1.6 -1,6

AlSi1MgMn -2

b) b)

-15

-10

-5

0

5

Current density [mA/cm2]

Fig. 8. Potentiodynamic polarization curves; a)AlMgSiP, b) AlSi1MgMn

a1) No LSP

pits. This type of corrosion is called pitting corrosion.

Corrosion resistance of the aluminium alloys was tested with potentiodynamic polarisation tests in a 3.5% NaCl water solution.

Anodic potentiodynamic polarisation tests were performed with Voltalab 21 potentiostat and corrosion cell CEC/TH, Radiometer Analytical.

The data were registered with a scan rate of potential of 10 mV/s, with the potential range from -2000 to -500 mVSCE. From the variation of polarisation curves in Fig. 8 it can be inferred that with an increasing

a2) 900 pulses/cm2

a3) 2500 pulses/cm2

b1) No LSP b2) 900 pulses/cm2 b3) 2500 pulses/cm2 Fig. 9. SEM macrographs of the specimen surfaces; a) AlSi1MgMn, b) AlMgSiPb Surface Modification of Aluminium Alloys with Laser Shock Processing

391

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 385-393

laser-pulse density the passivation (Epass) and pitting potential (Epitt) will also increase.

Table 3 shows the results obtained in the potentiodynamic polarization tests of the specimens prior to and after LSP. Alloy AlMgSiPb treated with 900 pulses/cm2 showed an increase in pitting potential of 219 mV and after 2500 pulses/ cm2 an increase in the pitting potential of 267 mV in comparison with the same material in the asdelivered state, was established.

An increase in pitting potential after LSP was noted also with alloy AlSi1MgMn. For an additional confirmation of improved corrosion resistance, the specimens were verified also with the SEM microscope.

In accordance with an ASTM standard for the preparation of specimens after corrosion testing [14], the specimen surfaces were subjected to a preliminary cleaning action in nitric acid HNO3.

With all the specimens treated, the same cleaning time, i.e. 2 min, was used. Fig. 9 shows SEM images of the specimen surfaces prior to and after LSP with an additional corrosion test.

From the surface images it can be assessed that with both aluminium alloys the largest number of corrosion damages (pits) at the specimen surfaces occurs in the as-delivered state. Whereas the surfaces of the specimens which were subjected to a preliminary LSP confirm that with a higher pulse density the number of pits will reduce. At the specimen surfaces after corrosion tests also a corrosion product (CP) near pits, which mostly consists of Al(OH)3 is visible.

A metallographic analysis of the specimens after corrosion testing confirmed that alloy AlMgSiPb shows higher corrosion resistance since after corrosion testing there is a smaller number of pits than at alloy AlSi1MgMn. 3 CONCLUSIONS The results obtained in the investigation permit the following conclusions: The factorial design of the input parameters confirmed that aluminium alloy AlMgSiPb is a more suitable material with a smaller increase in surface roughness Ra after LSP treatment with both pulse densities.

392

The analysis of residual stresses confirmed that after LSP treatment the specimens show higher compressive residual stresses with the higher pulse density. After treating alloy AlMgSiPb with 2500 pulses/cm2, compressive residual stresses of -337 MPa are obtained just beneath the surface in a depth of 0.033 mm.

Corrosion testing confirmed that the intensity of pitting corrosion attack decreases with the increase in pulse density. With both alloys an increase in pitting potential with higher pulse density was confirmed.

The metallographic analysis of the specimens after corrosion testing confirmed alloy AlMgSiPb as better corrosion resistant material with a smaller number of pits than with aluminium alloy AlSi1MgMn. 4 REFERENCES [1] Zupanc U., Grum, J. (2011). Surface integrity of shot peened of 7075-T651 aluminium alloy.

Strojniški vestnik - Journal of Mechanical Engineering, vol. 57, no. 5, p. 379-384.

[2] Žagar, S., Grum, J. (in press). Surface integrity after mechanical hardening of various aluminium alloys. Strojniški vestnik - Journal of Mechanical Engineering, DOI:10.5545/sv-jme.2010.092.

[3] Clauer, A.H. (1996). Surface performance of titanium. Gregory, J.K., Rack, H.J., Eylon, D. (eds.). The Metal Society of AIME, Warrendale, p. 217-230.

[4] Grum, J., Trdan, U., Hill, M.R. (2008). Laser shock processing of ENAW 6082 aluminium alloy surface. Materials Science Forum, vol.

589, p. 379-384.

[5] Zhang, W., Yao, Y.L. (2002). Micro scale laser shock processing of metallic components.

Journal of Manufacuring Science and Engineering, vol. 124, p. 369-378.

[6] Hammersley, G., Hackel, L.A., Harris, F. (2000). Surface prestressing to improve fatigue strenght of components by laser shot peening. Optics and Lasers in Engineering, vol. 34, p. 327-337.

[7] Ocaña, J.L., Molpeceres, C., Porro, J.A., Gómez, G., Morales, M. (2004).

Experimental assessment of the influence of irradiation parameters on surface deformation

Trdan, U. – Ocaña, J.L. – Grum, J.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 385-393

and residual stresses in laser shock processed metallic alloys. Appl. Surf. Sci., vol. 238, p.

501-505.

[8] Ding, K., Ye, L. (2006). Laser shock peening - Performance and process simulation.

CRC Press, Woodhead Publishing Limited Cambridge.

[9] Sano, Y., Obata, M., Kubo, T., Mukai, N., Yoda, M., Masaki, K., Ochi, Y. (2006).

Retardation of crack initiation and growth in austenitic stainless steels by laser peening without protective coating. Materials Science and Engineering A, vol. 417, p. 334-340.

[10] Hong, Z., Chengye, Y. (1998). Laser shock processing of 2024-T62 aluminum alloy.

Materials Science and Engineering, A257, p.

322-327.

[11] Montgomery, D.C. (2001). Design and Analysis of Experiments. Wiley, New York.

[12] Grum, J., Slabe, J.M. (2004). The use of factorial design and response surface methodology for fast determination of optimal heat treatment conditions of different Ni–Co–Mo surfaced layers. Journal of Materials Processing Technology, vol. 155156, p. 2026-2032.

[13] ASTM Standard (1995). Standard Test Method for Determing Residual Stress by the Hole Drilling Relaxation method, ASTM E 837-01, ASTM Int., West Conshohocken.

[14] ASTM Standard (2003). Standard Practice for Preparing, Cleaning, and Evaluating Corrosion Test Specimens (G1-03), ASTM Int., West Conshohocken.

Surface Modification of Aluminium Alloys with Laser Shock Processing

393

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 394-404 DOI:10.5545/sv-jme.2010.081

Paper received: 14.03.2010 Paper accepted: 23.03.2011

An Intelligent Electro-Hydraulic Servo Drive Positioning Detiček, E. – Župerl, U. Edvard Detiček* - Uroš Župerl University of Maribor, Faculty of Mechanical Engineering, Slovenia

The goal of the research is to develop a closed loop control system for position control of electrohydraulic servo drive useful for practical application. The research was performed based on the theory of adaptive feedback systems. The proposed new control structure consists of three parts. The first part is actually a feed-forward digital filter in the form of an inverse model of the system, which enables a perfect reference tracking behaviour. The second part is a position feedback fuzzy logic controller designed as a self-learning system, which possesses an ability of automatic tuning and is able to cope with parameter and load changes of the control loop. The third part is the so called switching integrator involved to improve steady state errors and ensure precise final positioning. The effectiveness is proven by laboratory experimental investigations. ©2011 Journal of Mechanical Engineering. All rights reserved.

Keywords: servo-hydraulic, linear actuators, position control, nonlinear and fuzzy control 0 INTRODUCTION High-speed/high accuracy positioning is a key element in modern mechatronics systems such as computer controlled machine tools, metal-forming machines, manipulators, injection moulding plastic machines, assembly and transport devices, material testing devices, etc.

To obtain this rapid movement, hydraulic drives controlled by resistance in impressed pressure network are used very often. In addition to the excellent dynamics, the main advantage of such systems is the parallel use of multiple drives fed by one pressure net combined with energy recycling into the hydraulic accumulators. In particular, the hydraulic cylinders should be stressed at this point due to their ability of direct transformation of hydraulic energy into linear movements and forces. Fluid power actuators are characterized by their high power density and excellent dynamic response. The hydraulic actuators in particular are capable of very high output power levels combined with very compact drive unit dimensions, with very good positioning speed being achieved. They are ideally suited to many high dynamic drive applications in modern machines and mechatronics systems.

The servo-hydraulic cylinder system is also a mechatronic system [1], which can be divided into the electronic part and the mechanical part. It is quite normal nowadays for cylinders to be equipped with electronically controlled 394

proportional and servo valves, as well as with position transducers and force sensors, creating together closed control loops. To fulfil another demand of the above mentioned machines, namely the accurate and precise position and force control, appropriate control strategies are needed.

From the control engineering point of view, a good reference tracking for the step, ramp, and sinusoidal reference signals, as well as rejection on external disturbances, have to be achieved.

However, the disadvantages of hydraulic systems such as the low damping ratio, nonlinearities of valve cylinder combination, friction forces, and fluid compressibility need to be overcome. This is successfully obtainable only by implementation of modern control theory and sophisticated digital control algorithms [2]. Backe [2] summarized the development of hydraulic applications from 1975 to 1995 and forecast the need for advanced control techniques.

In other words, to meet the requirements of modern mechatronics systems the electrohydraulic drive systems have to possess increasingly more intelligence. The objective of many researchers is therefore to develop an algorithm which would be able to control the drive without any a-priori knowledge of geometrical, operating or any other parameters of the system.

A contribution to achieve the above objective is presented in this paper. A very promising tool to achieve such desired goals is the fuzzy logic control. Fuzzy control technology is based on

*Corr. Author’s Address: University of Maribor, Faculty of Mechanical Engineering, Smetanova 17, Maribor, Slovenia, edvard.deticek@uni-mb.si

StrojniĹĄki vestnik - Journal of Mechanical Engineering 57(2011)5, 394-404

the linguistic variables Ë&#x2014; the human thought process; the input signal is fuzzified at first, then goes through the fuzzy reasoning process under operational experience and expert knowledge, and finally the control signal is defuzzified and sent out [3] and [4]. In the literature [5] to [7], servo-hydraulic system control has been executed by fuzzy controller successfully; however, which control parameters to use to satisfy the performance requirement is determined through trial and error method. Berger [8] presented the idea of designing fuzzy controllers according to the classification of the controlled systems. Thus, fuzzy controllers can be classified as Fuzzy-PID, Fuzzy-PI and Fuzzy-PD controllers. Yamazaki [9] and Mudi [10] tested the self-tuning of PI controllers using fuzzy logic and demonstrated that fuzzy control was able to reduce the initial cost of building an automatic tuning system. An optimal-tuning PID control has also been devoted to a proportional control system including a 4/3-way proportional valve and a differential cylinder with computer data acquisition system [11] and [12]. Branco [13] has demonstrated the feasibility of using fuzzy control to reduce the influence of unmodeled nonlinearities and parameter uncertainties in hydraulic systems.

Jelali [14] summarized recent developments in nonlinear identification, nonlinear control and the application of both to hydraulic servosystems. Kim [15] proposed an experimental

optimization algorithm to determine the effective control parameters for an electro-hydraulic position control system. In his algorithm the variations of the system dynamics can be properly compensated. Xiang performed a design study of adaptive Fuzzy PD controller for pneumatic servo-system [16].

This paper presents a new hybrid-fuzzy control strategy for position control of the electrohydraulic linear drive. An adaptability is obtained by fuzzy logic controller designed as a selflearning system, while the reference tracking and position accuracy are improved by conventional control measures such as an inverse feed-forward controller and switching integrator. The algorithm is experimentally investigated and implemented on the hydraulic device for testing mechanical constructions-load simulator.

1 ADAPTIVE CONTROL STRUCTURE There are clear design goals for electrohydraulic drive control system to achieve fast response to step command signals with minimal overshoot (dead-beat response) and approximate zero steady state errors for step, ramp and sinusoidal reference inputs. Those goals have to be obtainable in a rough industrial environment, where significant system behaviour changes may occur due to nonlinear characteristics of system elements, temperature changes, noise

Fig. 1. Shematic presentation of electro-hydraulic servo drive An Intelligent Electro-Hydraulic Servo Drive Positioning

395

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 394-404

and other environmental influences. Therefore, the controller has to be able to adapt itself to the changing properties of controlled process and its signals.

1.1 Description of Electro-Hydraulic Servo Drive Dynamics The schematic of electro-hydraulic servo system is depicted in Fig. 1. It is composed of a single rod cylinder, a 4/3 way servo valve, a sliding load mass, a position transducer, an electronic interface and a control computer.

Since the high response servo valve is used it is assumed that the valve spool displacement xV is directly proportional to the electric current i from the electronic amplifier driven by control input voltage u.

The relation is described by the following equation where KV is valve spool gain:

xV(t) = KV u(t) .

(1) The servo valve second stage output flows QA, QB, by symetric zero overlapping spool and pump flow Q0 (rated), supply pressure p0, tank pressure pT ≈ 0, according to Bernoullie’s equation are [14]: for xV ≥ 0 : QA = QB = −

Q0

xV max Q0

xV max

xV sign( p0 − p A ) xV sign( pB )

p0 − p A , p0

pB , p0

(2)

for xV < 0 : QA = QB = −

Q0

xV max Q0

xV max

xV sign( p A )

pA , p0

xV sign( p0 − pB )

p0 − pB . p0

Following the continuity condition, with regard to oil compressibility E(p), the pressure changes in both cylinder chambers, can be expressed as:

dp A (t ) E ( pA ) = QA ( p A , xV , t ) − dt V0 A + AA y (t ) 

396

− AA

(3)

dy (t ) − K Li ( p A (t ) − pB (t )) ] , dt

dpB (t ) E ( pB ) dy (t ) = [ AB dt − QB ( pB , xV , t ) (4) dt V0 B − AB y (t ) + K Li ( p A (t ) − pB (t )) ] ,

where AA, AB are piston areas, pA - left chamber pressure, pB - right chamber pressure, V0A, V0B dead volumes, KLi - leakage coefficient, and y is the piston rod position e.g. load position.

The dynamics of the moving parts of the system is expressed by the force equilibrium: m

d2y = p A (t ) AA − pB (t ) AB − FR (t ) − FL (t ), (5) dt 2

where m is the moving mass, FR(t) is friction force and FL(t) is external load force.

Finally, by rearrangement of the above equations and introduction of some new terms, such as velocity gain Vs(pA, pB, y), eigen frequency ω(pA, pB, y) and damping D(pA, pB, y), leads to a common mathematical expression: d 3 y (t ) d 2 y (t ) + 2 D( p A , pB , y )ω ( p A , pB , y ) + 3 dt dt 2 +ω 2 ( p A , pB , y ) dy (t ) = ω 2 ( p A , pB , y ) ⋅ (6) dt dF ⋅Vs ( p A , pB , y ) KV u (t ) − L . dt m After linearization of Eq. (6) in particular operating point, using Laplace transformation, the transfer function of the entire system can be derived. It generally has the form of a third order system (second order system and integrator), which numerical parameter varies among different operating points. A major simplification is done by neglecting the second order term (neglected oil compressibility and inertial forces) and representing the system by transfer function of an integrator:

G= (s)

y ( s ) VS KV K I = = . (7) u (s) s s

It should be pointed out that integrator gain KI significantly depends on piston movement direction.

1.2 Adaptive Control Structure A successfully applied control strategy based on conventional control theory is depicted

Detiček, E. – Župerl, U.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 394-404

in Fig. 2. It consists of a digital self-tuning, parameter adaptive controller to cope with parameter changes, a cancellation feed-forward controller to minimize steady-state errors of ramp and sinusoidal reference inputs and a switching integrator to improve precise final positioning of hydraulic drive.

Fig. 2. Adaptive control structure for position control of electro-hydraulic drive It is important to obtain the digital parameter adaptive controller, to be realizable on a control computer, inside a sampling interval, which needs to be small enough to represent electro-hydraulic servo drive dynamics well (in our case T = 2 ms). Therefore, the self-tuning approach is used and a simplified mathematical model Eq. (7) is taken to be the identification model of the system. However, it shows that all changes are reflected on load pressure changes and therefore on KI as well. In that way it is the function, not only of systems state variables, servo valve rated flow, piston area and supply pressure but also of inertial, friction and external forces.

Putting Eq. (7) in to discrete form it gives: Gm ( z ) = (1 − z −1 ) z (

TK I b1 KI )= = . (8) −1 2 s 1− z 1 − z −1

Therefore, a realizable, digital parameter adaptive (self tunning) controller, with only one unknown parameter b1 is obtained using recursive least square estimation (Iserm). The equation for unknown parameter vector estimation has the general form:  ( k + 1) = Θ  (k ) + γ (k )e(k + 1), Θ

(9) with correction vector γ(k) : γ (k ) =

P(k )Ψ (k + 1) , (10) Ψ (k + 1) P(k )Ψ (k + 1) + λ (k + 1) T

The final form of an adaptive P-controller with initial model value bm (Koek) is: u AC (k ) =

   bm (k ) b1 (k ), b1 (k ) ≥ b1 min , b1 (k ) =  . (12) b1 (k ) b1 (k − 1), else

The computer control algorithm also contains a decision logic to separate computing of (bm,left , b 1,left ,bm,right , b 1, right ), depending on the movement direction. The effectiveness of the above adaptive P-controller is experimentally proven by step change of supply pressure (see Fig. 9). Supply pressure change has been avoided from any further test, while this is not a part of normal operating conditions.

The feed-forward controller (UNbeha) is realized on the basis of the inverse model Gm-1(z).

It has non-causal form but is realizable in control computer for a partial compensation of steadystate errors by ramp and sinusoidal reference signals:

w′(k ) =

1 [ w(k + 1) + w(k )] . (13) bm

It also depends on movement direction (bm,left , bm,right ). Switching integrator in form: uSI (k − 1) + ∆uSI (k ), if emin ≤ e(k ) ≤ emax , (14) , 0 else 

uSI (k ) = 

is switched on to improve final position accuracy.

It is switched off just before the desired position arrives to prevent limit cycle oscillations.

Among the several advantages of the above control structure the main disadvantage is the inability to obtain automatic first start because the initial value of the estimated parameter must be known in advance.

To overcome this problem our further investigations lead to the replacement of the conventional adaptive controller by appropriate fuzzy logic self learning (self organizing) algorithm.

2 FUZZY POSITION CONTROL OF ELECTRO-HYDRAULIC LINEAR DRIVE

where Ψ is the signal vector and P is the covariance matrix, with λ - forgetting factor:

2.1 Fuzzy Control

P(k + 1) = 1

Fuzzy logic, developed by [3] in 1960s, is much closer to human reasoning and natural

λ (k + 1)  I − γ (k )Ψ T (k + 1)  P(k )

. (11)

An Intelligent Electro-Hydraulic Servo Drive Positioning

397

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 394-404

language. In contrast to the traditional set theory, where elements are only, either complete members of a set, or complete non-members of a set. It provides an effective means of capturing the approximate and inexact nature of the real world. Fuzzy sets have no precise criterion for membership.

To distinguish between members that are more probable than those which are less probable in belonging to the set, the grade of membership denoted by μ which lies in the range of 0 to 1 is used. The fuzzy sets can have different shapes such as triangular, trapezoidal, bell-shape, etc.

The elements of the fuzzy set are taken from the universe of all elements.

A principal characteristic of fuzzy control is that it works according to linguistic rules such as “if the room temperature is low then increase heating” rather than obeying mathematical models and functional relationship. The fuzzy controller consists of four main elements: a fuzzifier, an inference engine, a rule base and a defuzzifier.

The first step of the fuzzy controller design is to select the number of controller inputs and corresponding number of input membership functions required for the fuzzification process.

The fuzzification is the process of translating crisp input values into fuzzy linguistic values through the use of membership functions. In other words, determining how much each real value belongs to each fuzzy set using the corresponding membership value.

The fuzzy rule base is simply a data base of the desired control rules for the system. It is most equivalent to the controller of a traditional control system and formalizes the designer’s “expert knowledge” of what control output should result from a given combination of system states, expressed in a linguistic manner. It often takes the form of a truth table consisting of rules constructed in the following form: IF “condition 1” AND/ OR “condition 2” THEN “consequence 1”. In fuzzy logic terminology, the statement following the IF condition is known as the “premise”, “antecedent” or “condition”. The corresponding statement following THEN is known as the “conclusion” or “consequent”. The inference engine is the heart of a fuzzy logic controller.

It acts as the bridge between the fuzzification input stage and defuzzification output stage of 398

the controller, translating the designer’s desired control rules from linguistic representation to a numeric computation. The inference engine can be divided into three elements. The first step is known as “aggregation”. During this step the premise (IF statement) of each rule in the rule base is calculated using the fuzzified controller inputs. With fuzzification, each condition in the premise is assigned a degree of membership in the corresponding input fuzzy set. It should be noted that the traditional formulation of logical statements such as AND/OR has been modified to accommodate the use of membership functions.

In particular, the result of an AND operation is often defined as either the minimum or a product of two fuzzy values compared. Similarly, an OR operation is often defined as either the maximum or probabilistic sum. Any premise with a value greater than zero means that its corresponding rule is active. The second step of the inference process is known as “composition” or “implication”, where the consequent (THEN statement) of each rule is created using the premises calculated in the first step. The output of the composition step is not a single value for each rule in the rule base, but rather one modified output fuzzy set for each rule. These modified output sets are known as “implied” fuzzy sets. The modification is controlled by the premise calculated in the aggregation step. There are two fundamental methods of creating the fuzzy sets that are the results of composition min and prod. The min operation truncates the output fuzzy set at the value of the premise while the prod operation scales the output fuzzy set according to the premise. The third and final step of the inference process is known as “accumulation” or “results aggregation”. In this step the implied fuzzy sets that are the output of the composition process are combined into an accumulated fuzzy set, which is the input to the defuzzification. Defuzzification is the process of converting the fuzzy output set which is the result of inference of the process into a discrete number. Two fundamental methods are known as the Mean of Maxima and the Center of gravity. The latter, also known as the Center of Area method involves two steps: • multiply the membership degree μ(xi) of each element i by the singleton value xi of the membership function.

Detiček, E. – Župerl, U.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 394-404

•

sum the values obtained in the first step and divide total by the sum of the output membership degrees.

2.2 Fuzzy PD Controller For stability reasons the general form of PD-controller was used at the beginning:

u = f (e, e) . (15)

By fuzzification of two input variables (e, e ) and one output variable (u) of the controller, a two- dimensional rule base has been constructed.

Linguistic variables e, e , and u are shown in Fig. 3. The universe of discourse of each variable is selected in range 1:α, where α=AA/AB is piston area ratio.

Fig. 3. Linguistic variables represented as fuzzy variables; a) e, b) e , c) u The rule base or decision table, shown in Fig. 4, contains linguistic rules of the form: IF (A and B ) THEN C. In the present context the antecedents A and B refer to linguistic statements about the error between set point and actual position of the moving mass and the rate of change of this error. The logic conclusion of each rule is about the servo valve opening. Each non zero entry in the decision table corresponds to one rule. Two axes or universes of discourse

of the decision table are for error e and its change e respectively. Each axis has 13 entries from -6 through 0 to +6. The points 1 to 6 can be taken to mean: very small, small, medium, quite large and very large. The fuzzy sets have, for computational simplicity, a triangular shape with μ - values of 0, 0.1, 0.2, ..., 0.9, 1. The consequent of each rule in the matrix also has one of 13 values and represents the contribution of each rule to the output action set. The collective effect of all rules can be given as: (rule 1) ELSE (rule 2) ELSE (rule 3) ELSE...

“ELSE” can be regarded as an inclusive OR function or MAX function. Finally, the centre of gravity of the output action set was used, as the method of defuzzification, to obtain unique real output.

From Fig. 4 it can be seen that the rule base is diagonally symmetrical and the rules along the diagonal change equally.

Fig. 4. Rule base or decision table The consequence is a linear dynamic behaviour of the controller. In other words, the fuzzy controller emulates an exactly conventional PD-controller. This was practically proved by experiments. Due to fuzzyffication of measurement signals and robustness of fuzzy rules, the fuzzy controller is less sensitive to noise and small parameter variations. It is therefore, to a certain extent, more robust than the conventional one.

An Intelligent Electro-Hydraulic Servo Drive Positioning

399

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 394-404

Fig. 5. a) Self-organizing mechanism of fuzzy logic controller; b) Self-learning mechanism of fuzzy logic controller On the other hand, it is possible to obtain nonlinear characteristics of the fuzzy PDcontroller by modification of the symmetrical rule base.

The following paragraph describes an attempt of automatic rule base creation.

2.3 Fuzzy Self-Learning Controller The theory of fuzzy control seems to be a suitable mathematical tool for both modelling and control of complex systems [3]. However, the primitive form of fuzzy control sometimes fails in dealing with these complex systems mainly because it lacks enough adaptability. Several researchers have devoted themselves to developing fuzzy controller with more “intelligence” [1], [17] and [18]. Unfortunately, most of them are difficult for real time implementation.

Therefore, we decided to implement the so called reinforcement learning, proposed in [9] and [19].

Self-learning fuzzy logic controller tries to emulate human decision making behaviour and learning, namely, an ability to make rules and to modify them based on experience. It has a 400

hierarchical structure, which consists of two rule bases.

The first one is constructed as a base of instructions and a base of performances simultaneously. It creates and modifies the main rule base according to the desired overall performance of the system. At the first start of the controller, the main rule base contains zero values and it begins to fill according to base of instructions.

The quality of control action is checked by a comparison of present values of control error e and e with expected values from the base of performances. The linguistic terms in the base of performances show how the rules must be changed. The question is which rule must be changed, if we know that the present e and e are caused by a past control action. The solution is determining the time constant τ, which express that the control result at nth discrete moment of time is a consequence of control action activated at the (n ‒ τ)th discrete moment in the past. Determination of τ is done on the base of heuristic knowledge of dynamic behaviour of the control process.

As the investigations show, the selection of the τ is not critical and can be made inside a certain range of values. The self learning fuzzy controller

Detiček, E. – Župerl, U.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 394-404

therefore, uses control computer memory buffer to save past information and contains the mechanism for changing rules (Fig. 5a) or creating new rules (Fig. 5b), if they do not yet exist. It can be said that the controller also possesses the ability to self organize.

Let us suppose the above mentioned learning process is convergent, then the general rule base become constant after certain number of experiments. As long as the changes in system parameters are not significant, the rule base stays constant; otherwise a new adaptation is needed.

The self-learning and self-organizing mechanism is explained comprehensively in Fig. 5, with only three linguistic terms for the sake of simplicity.

Fig. 6 shows the original forms of rule base and base of instructions at the beginning of the control process. The rule base and the base of instructions has integer entries ranging from -6 to +6. These entries are positive or negative reinforcements, which are given to the corresponding (e, e ) pair in the rule base.

Reinforcements also implicate rules creation, in case they do not yet exist. Fig. 6b shows the matrix developed by [9], although the other versions could be used. Learning proceeds as follows. Let us assume a delay of τ samples (in our case τ = 4) on controlled system, then the present e and e values are mainly the consequences of controller output from τ samples ago. Each past state consisting of a triplet (eτ, e τ, uτ) is stored in a control computer “history” buffer. As the present values of e and e point to the value p in the performance matrix, then the rule entry for eτ, e τ in the rule base become uτ,+p, whatever the rule entry may have been up to the present time.

The self-organizing controller comprises both the blank rule base fuzzy controller and the above learning process.

3 EXPERIMENTAL RESULTS

Fig. 6. a) Rule base or decision table, b) base of instructions

The laboratory experiments were performed on laboratory testing equipment shown in Fig. 7. First the self tuning parametric adaptive controller, according to Eq. (12), was tested. It was compared with an ordinary P-controller (Fig. 8a), by ramp reference signals.

Adaptive ability was proven by a sudden supply pressure drop of 20%, at the time of 0.3 s (Fig. 8b). The next experiments were step responses of fuzzy logic controller. First the selforganizing fuzzy logic controller was started with zero rule base (Fig. 9a), while the learning process was currently involved in the following tests (Fig. 9b).

The rule base becomes mostly constant after the second and third step. Finally such an ”experienced” fuzzy controller, together with feed-forward controller Eq. (13) and switching integrator Eq. (14) were joined in, so called, “hybrid” control structure to obtain a ramp response of the hydraulic drive (Fig. 10).

An Intelligent Electro-Hydraulic Servo Drive Positioning

401

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 394-404

a)

b)

Fig. 7. a) Structure of the testing system (cylinder40/28 mm, slider mass 70 to 300 kg, lin. pot. 300 mm ±1%), b) Servo valve (MOOGD-769-233, Q = 19 l/min)

a)

b)

Fig. 8. a) P-controller ramp response by 20% supply pressure drop, b) Self-tuning controller ramp response by 20% supply pressure drop

Fig. 9. a) First step of learning process, b) Third step of learning process The proposed hybrid fuzzy control structure is shown in Fig. 11.

The experiments show that the selforganizing fuzzy controller enables a successful control of electro-hydraulic drive without any a-prior knowledge of the system dynamics in the 402

first starting conditions [20] and [21]. It is also able to cope with non-linearity and parameter changes and therefore, possess a certain level of adaptability as well. However, the reference tracking behaviour [22] and final positioning accuracy are still poor.

Detiček, E. – Župerl, U.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 394-404

Fig. 10. Ramp response of electro-hydraulic servo drive with hybrid fuzzy position controller

Fig. 11. Hybrid fuzzy control structure of electro-hydraulic servo drive position control 4 CONCLUSIONS Systems having structural uncertainties or a known complicated structure such as hydraulic systems are difficult to control.

The dynamic characteristics of such systems are usually very complex and highly nonlinear. For a practical control system, it is usually desired to have a fast accurate response with a small over-shoot.

In this paper, a possibility to apply a selforganising and self learning fuzzy algorithm for position control of hydraulic servo drive is represented. In addition to self-organisation it also possesses the ability of self learning. It is able of

adaptation to parameter changes and to therefore, deal with nonlinear dynamic behaviour associated with the drive motion.

Unfortunately, there is no systematic way to prove the convergence of a learning mechanism and overall stability of the control system. Trial and error method was used with these experiments.

Therefore, additional research will be required in the future.

5 REFERENCES [1] Shih, M.C., Tsai, C.P. (1995). Servohydraulic cylinder position control using a neuro-fuzzy

An Intelligent Electro-Hydraulic Servo Drive Positioning

403

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 394-404

controller. Mechatronics, vol. 5, no. 5, p. 497512.

[2] Backe, W. (1993). The present and future of fluid power. Proceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering, vol. 207, p. 193-212.

[3] Zadeh, L.A. (1965). Fuzzy Sets, Informat Control, vol. 4, no. 8, p. 126-174.

[4] Bouslama, F., Ichikawa, A. (1992). Fuzzy control rules and their natural control laws.

Fuzzy Sets Syst., vol. 48, p. 65-86.

[5] Shih, M.C., Chen, P.C. (1991). An experimental study on the position control of a hydraulic cylinder usinga fuzzy logic controller. JSME, Series III, vol. 34, no. 4, p.

481-489.

[6] Chen, C.Y., Liu, L.Q., Cheng, C.C., George, T. (2008). Fuzzy controller design for synchronous motion in a dualcylinder electro-hydraulic system. Control Engineering Practice, vol. 16, no. 6, p. 658673.

[7] Kastrevc, M., Pusenjak, R. (2005). Fuzzy pressure control of hydraulic system with gear pump driven by variable speed induction electro motor. Exp. Tech. (Westport Comm.), vol. 29, no. 3, p. 57-62.

[8] Berger, M. (1996). Self-tuning of a PI controller using fuzzy logic for a construction unit testing apparatus. Control Engineering Practice, vol. 4, no. 6, p. 785-790.

[9] Yamazaki, T. (1982). An improved algorithm for self-organising controller and its experimental analisys. John Hopkins, University Press, Baltimore.

[10] Mudi, R.K., Pal, N.K. (2000). A self-tuning fuzzy PI controller. Fuzzy Sets Systems, vol.

115, no. 2, p. 327-338.

[11] Liu, G.P., Daley, S. (2000). Optimal-tuning nonlinear PID control of hydraulic systems.

Control Engineering Practice, vol. 8, no. 9, p.

1045-1053.

[12] Kim, S.M., Han, W.Y. (2006). Induction motor servo drive using robust PID-like neuro-fuzzy controller. Control Engineering Practice, vol. 14, no. 5, p. 481-487.

[13] Branco, P.J.C., Dente, J.A. (2000). On using fuzzy logic to integrate learning

404

mechanisms in an electro-hydraulic system ‒ Part I: Actuator’s fuzzy modeling.

IEEE Transactions on Systems, Man, and Cybernetics ‒ Part C: Applications and Reviews, vol. 30, no. 3, p. 305-316.

[14] Jelali, M., Kroll, A. (2003). Hydraulic servosystem modeling, identification and control.

Springer, London, New York.

[15] Kim, M.Y., Lee, C.O. (2006). An experimental study on the optimization of controller gains for an electro-hydraulic servo system using evolution strategies. Control Engineering Practice, vol. 14, no. 2, p. 127-147.

[16] Xiang, G., Zheng-Jin, F. (2005). Design study of an adaptive Fuzzy-PD controller for pneumatic servo system. Control Engineering Practice, vol. 13, no. 1, p. 55-65.

[17] Čuš, F., Župerl, U. (2007). Adaptive selflearning controller design for feedrate maximization of machining process.

Advances in Production Engineering & Management, vol. 2, no. 1, p. 18-27.

[18] Milfelner, M., Župerl, U., Čuš, F. (2004).

Optimisation of cutting parameters in high speed milling process by GA. International Journal of Simulation Modelling, vol. 3, no.

4, p. 121-131.

[19] Lee, C.C. (1990). Fuzzy logic in control systems. IEEE Transaction on Systems, Man and Cybernetics, vol. 20, no. 2, p. 24-32.

[20] Herakovič, N. (1995). Piezoaktorbetätigung für ein einstufiges hochdynamisches Servoventil = [Piezoactuator for a singlestage servovalve with high dynamic response]. O+P, Ölhydraul. Pneum., vol. 39, no. 8, p. 601-605.

[21] Župerl, U., Čuš, F., Kiker, E., Milfelner, M. (2005). A combined system for off-line optimization and adaptive adjustment of the cutting parameters during a ball-end milling process. Strojniški vestnik – Journal of Mechanical Engineering, vol. 51, no. 9, p.

542-559.

[22] Roy, S.S. (2010). Modelling of tool life, torque and trust force in drilling: a neurofuzzy approach. International Journal of Simulation and Modelling, vol. 9, no. 2, p.

74-85.

Detiček, E. – Župerl, U.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 405-416 DOI:10.5545/sv-jme.2009.084

Paper received: 13.07.2009 Paper accepted: 12.01.2011

Determining the Capacity of Unloading Bulk Cargo Terminal Using Queuing Theory

Bugaric, U. ‒ Petrovic, D. ‒ Petrovic, Z. ‒ Pajcin, M. ‒ Markovic-Petrovic, G. Ugljesa Bugaric1,* ‒ Dusan Petrovic1 ‒ Zoran Petrovic2 ‒ Miroslav Pajcin3 ‒ Gordana Markovic-Petrovic4 1 University of Belgrade, Faculty of Mechanical Engineering, Serbia 2 Tecon Sistem d.o.o., Serbia 3 Jugoimport SDPR, Serbia 4 DZ-Zemun, Serbia Hierarchical structure of the system, river terminals for bulk cargo unloading, connected with queuing and servicing and stochastic character of the input/output values are underlined. The approach using the queuing theory is developed for engineering use as simpler, faster and more convenient than the approach using simulation. Results obtained using the queuing theory and previously obtained results using simulation modelling are shown alongside. The obtained results can be used at the beginning of the design process when rough estimations of the system behaviour are needed. Some of the obtained results are applied and verified on the existing system. © 2011 Journal of Mechanical Engineering. All rights reserved.

Keywords: bulk cargo, unloading, river terminal, queuing theory 0 INTRODUCTION Work of the ports with its optimal capacity assumes a prompt accommodation of vessels with minimal waiting time in the port and with maximal use of berth facilities. The capacity of a port generally depends on the number of berths available to ship traffic and cargo handling capacity. The terminal for bulk cargo unloading can stand alone as a specialized terminal or can be a part of a port. Specialized stand alone river terminal for unloading bulk cargo (i.e. river terminal) presents the organization of different activities, connected with the control and handling of material flow from the vessel to the transport or the storage system of the technological installations, which provides maximal servicing of vessels with minimum expenses. Those river terminals are mostly used in electrical plants for coal unloading, in steel mills for unloading of iron ore and coal and in chemical plants for unloading of raw materials etc. Terminals of this kind have a big unloading rate, which is their main characteristic.

Bulk cargoes, which are to be unloaded differ by granulation and density. Materials are relatively dry so that they do not compose compact mass and they can take shape of the cargo space of vessel. Very important characteristics of those

materials are the fact that the cost of transportation and manipulation is an important part of their final value. [3] This paper aims at determining the capacity of the unloading bulk cargo terminal using the queuing theory. The same topic with a different approach (simulation) was previously discussed and analysed in [3]. In addition to the new approach (queuing theory), the difference between obtained results using the queuing theory and previously obtained results using simulation are discussed and underlined.

The approach using the queuing theory is developed for engineering use as a simpler and faster and in this way more convenient for the use in the starting phases of the design process than the approach using simulation.

In this new analytical approach, two different models of the river terminal are still used. The first model is when unloading devices are working without strategy (i.e. model 1), while the second model is when unloading devices are working with strategy (i.e. model 2).

Therefore, the system modelling used in this paper is basically the same with system modelling in [3], with changes needed due to the queuing theory model limitations.

The examples of port modelling can be found in works of Agerschou et al. [1], Comer et

*Corr. Author’s Address: University of Belgrade, Faculty of Mechanical Engineering, Kraljice Marije 16, 11120 Belgrade 35, Serbia, ubugaric@mas.bg.ac.rs

405

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 405-416

al. [5], El Sheikh et al. [7], Kondratowicz [9], Park et al. [11], Sinowczik [12].

1 SYSTEM MODELLING In some areas of design practice there is a strong opinion that the design premises are known and unchangeable, such as electronics and mechanics where the characteristics and structure of materials and goods as well as the environment are known. While in the design of technical and technological systems of material flow (ports, specialized terminals, distribution centres, storages etc.) this is not the case. Design premises, such as: the number of barges in tows, different bulk cargoes in barges, the distance between anchorage and the berth, water level, meteorological conditions, are usually unknown and changeable, which means that the stochastic behaviour is present. This is also the case in the design of river terminals [3].

The above mentioned facts as well as the possibiliy of an occurrence of a natural phenomena, stochastic changes in water and land transport, failure in work of unloading mechanization and other involved equipment which are also stochastic indicate that the only way to design such systems is to use the stochastic approach.

The data needed for this paper were gained from an existing river terminal for bulk cargo

unloading on river Danube. A simplified layout of this terminal is shown in Fig. 1 [2] Modelling and design of the river terminal is not possible for the whole system at once, so first decomposition of the system must be done. Some sub-systems must be analyzed and designed separately and after this the modelling of the whole system can be done according to the parameters, which are gained through the behaviour of the sub-systems. There is a strong hierarchical structure of the terminal meaning that the output of the hierarchically lower sub-systems is usually input for the hierarchically upper subsystems. Hierarchical structure and levels of river terminal is shown in Fig. 2.

The river terminal usually consists of two sub-systems: »anchorage« and »vessel-operative coast«, due to the problem of the material flow.

The sub-system »vessel-operative coast« consists of three basic sub-systems: »unloading mechanization«, »conveyor« and »storage«. The basic sub-system »unloading mechanization« contains two elementary sub-systems »crane«. The »crane« elementary sub-system is hierarchically the lowest and represents knot point (bottleneck) for the material flow system.

The unloading bridge with a grab and hopper on it represents, in this case, the elementary sub-system »crane«.

Simplified servicing procedure, due to limitations of the queuing theory models, in the

ANCH

E ORAG

COAST ATIVE L-OPR CRANE II VESSE CRANE I UNLOADING MECHANIZATION STORAGE

CONVEYOR

Fig. 1. Terminal for bulk cargo unloading on river Danube - existing situation 406

Bugaric, U. ‒ Petrovic, D. ‒ Petrovic, Z. ‒ Pajcin, M. ‒ Markovic-Petrovic, G.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 405-416

I

TERMINAL

HIERARHICAL LEVEL

II

ANCHORAGE

III

System

VESSEL - OPERATIVE COAST

STORAGE

CONVEYOR

IV

Sub - system

UNLOADING MECHANIZATION CRANE I

CRANE II

Basic sub - systems Elementary sub - systems

Fig. 2. Hierarchical structure and levels of river terminal

λ = const.

Fig. 3. Servicing and material flow in river terminal Table 1. Data about bulk materials which are to be unloaded [2] Type of material

Absolute frequency of material appearance [t/year]

iron ore limestone coal

900000 200000 100000

Relative frequency of material appearance 0.75 0.167 0.083

Density of material ρm [t/m3]

Grab type

Coefficient of grab loading kg

2.2; 2.5; 2.7 1.43÷1.6 0.8

1 2 2

0.8; 0.75; 0.7 0.8 0.9

Mass [kg] 3600 3750

Dimensions L×B [m] 3.5×1.75 4.25×2

Table 2. Grabs - technical data [2] Grab type

Material type

1 2

iron ore limestone, coal

Volume Vg [m3] 3.2 5

river terminal and technological connections between sub-systems, due to material flow, are shown in Fig. 3.

The presented models cover a different kind of material in the vessels (Table 1, empirical distribution of material appearance - EMP), which are to be unloaded with different kind of grabs

Needed crane capacity [t] 12.5 12.5

(Table 2). Vessel type and number of vessels in the composition are given in Table 3.

Only one type of vessels and one composition size is used in modelling because of limitations of the basic queuing theory models.

Given quantities of material (Table 1) are planned for the period of one year of work and will

Determining the Capacity of Unloading Bulk Cargo Terminal Using Queuing Theory

407

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 405-416

be the starting values for the on-going experiment i.e. calculation of average arrival rate (λ), while other technical data are used for calculating different average servicing rates (μ).

Table 3. Vessel composition ‒ technical data [2] Vessel type 71701

Capacity (Qv) [t] / Volume (Vv) [m3] 1700/1700

No. of vessels in composition 6

1.1 Sub-System »Anchorage« The purpose of the “anchorage” subsystem is to accept and accommodate vessels.

Anchorage is the queue, where vessels are waiting for unloading. Probability that two vessel compositions arrive at the same time is extremely low and can be neglected, due to scheduling of arrivals of the vessel compositions. Inter-arrival time of the vessel compositions is randomly variable, considering weather conditions, river traffic etc. [2].

On the basis of the number of empty places at the anchorage, the decision of acceptance of the vessel composition is made, which means that when there is not enough space at the anchorage for the whole composition it can not be accepted to the system. Vessels are taken to servicing – unloading from the anchorage by the FIFO discipline [3].

1.2 The »Vessel - Operative Coast« Sub-System The “vessel - operative coast” sub-system, with its hierarchically lower systems, is in charge of vessel unloading. Unloading of the vessel consists of the two stages. The first stage is free digging up to approximately 80% (free digging rate μfd) of material to be unloaded from the vessel and the second stage is cleaning of the vessel until it is empty (cleaning rate μcl). After the 80% of the material is unloaded from the vessel, the thickness of the material in the vessel is too small, so there is the danger that the grab can damage the vessel.

Therefore, unloading must be changed to cleaning [2].

The model only identifies the existence of the »storage« basic sub-system and outlines the connections with the other two basic sub-systems in the environment of the »vessel - operative 408

coast« sub-system. The purpose of »storage« is to accept the material from the »unloading mechanization« when the means of land transport (highway and railway) are not available. Material from »storage« is brought out by means of land transport when they are available or by »unloading mechanization« using a »conveyor« when there is no vessel in the system and the means of land transport are available.

The »conveyor« basic sub-system consists of more mutually connected belt conveyors.

The purpose of the »conveyor« is to accept the material from »unloading mechanization« (through the hopper on the unloading bridge) and to pass it to the available means of land transport.

The »conveyor« has the priority due to material flow against »storage« i.e. if the means of land transport are available all unloaded material goes through it.

The Basic sub-system »unloading mechanization« contains two elementary subsystems »crane«. The “Crane” uses different grabs for unloading different types of materials from the vessel (Table 1 and 2). The purpose of »unloading mechanization« is to unload the material from a vessel to one of the two remaining basic subsystems the »conveyor« or the »storage«.

The main focus of this paper is in different modelling of the basic sub-system »unloading mechanization«. Two models are developed, which differ in manner of modelling »unloading mechanization«.

In the first manner (Model 1), two elementary sub-systems »crane« work independently, which means that the first “crane” unloads vessel on berth 1 and can not move to berth 2, while second “crane” unloads vessel on the berth 2 and can not move to berth 1. When unloading of the vessel on its berth is finished the “crane” is idle i.e. the “crane” is waiting for tugboat to drag empty vessel to anchorage and drag the loaded vessel from the anchorage to an empty berth.

In the second manner (Model 2), the two elementary sub-systems »crane«, work with a strategy. The strategy is developed and designed according to the criteria stipulating that at least one »crane« works with a free digging rate throughout the unloading process. The »unloading mechanization« strategy of work is as follows:

Bugaric, U. ‒ Petrovic, D. ‒ Petrovic, Z. ‒ Pajcin, M. ‒ Markovic-Petrovic, G.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 405-416

both »crane«-s are unloading the vessel together, for example, on berth 1 (Fig. 4a - Phase I) until approximately 80% of the material is unloaded (free digging rate μI = μII = μfd). After that the “crane” II goes to the second berth and starts unloading the vessel on berth 2 with free digging rate μII = μfd. During this “crane” I is cleaning the vessel docked on berth 1 with a cleaning rate μI = μcl (Fig. 4b - Phase II).

After cleaning the vessel docked on berth 1, »crane« I goes to berth 2 and starts unloading the vessel with a free digging rate μI = μfd, together with “crane” II (Fig. 4c - Phase III). At that moment a tugboat drags the empty vessel from berth 1 to the anchorage and drags the loaded vessel back from the anchorage to the empty berth 1, meaning that the time needed for vessel dragging is not wasted but overlapped with the engagement of “crane” I for unloading on berth 2 and does not need to be taken into consideration.

When approximately 80% of the material form vessel docked on berth 2 is unloaded, “crane” I goes to berth 1 and starts unloading the vessel with a free digging rate μfd. During this “crane” II is cleaning the vessel docked on berth 2 with a cleaning rate μcl (Fig. 4d - Phase IV). After cleaning the vessel docked on berth 2 is finished, the “crane” II joins the “crane” I on berth 1 and starts unloading the vessel with a free digging rate μfd. Also, at that moment the tugboat is dragging the empty and loaded vessel from/to berth 2, which means that this time overlapps with engagement of “crane” II for unloading on berth 1 and does not need to be taken into consideration. In this way, the unloading of the vessels is repeated cyclically.

The »crane« elementary sub-systems i.e. unloading bridges using a grab and hopper on it presents knot points of unloading terminals, and in most cases the »bottle necks«, so they are a basic prerequisite for optimal work of the whole terminal for bulk cargo unloading. Due to this fact the modelling of the “crane” must be done at the beginning of the modelling of the river terminal and must be performed with great detail [3].

1.3 The Elementary Sub-System »Crane« The unloading of the bulk cargo can be done with continuous unloading devices, or with

a)

b)

c)

d)

Fig. 4. The Sub-system »vessel - operative coast« work with a strategy; a) Phase I, b) Phase II c) Phase III, d) Phase VI grab crane devices. This paper considers only grab crane devices (Fig. 5).

The unloading cycle of grab crane devices consists of: material grabbing from the vessel, loaded grab transfer from the vessel to the receiving hopper, discharging and empty grab transfer from the hopper to the vessel. Full

Determining the Capacity of Unloading Bulk Cargo Terminal Using Queuing Theory

409

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 405-416

automation of the unloading process of the grab crane devices is possible but very expensive. On the other hand, the crane operator can not repeat the optimal unloading cycle in time. The only practical feasible solution is to introduce the halfautomatic unloading cycle i.e. unloading cycle.

This cycle consists of a manual part, where the crane operator controls the grab moving, and an automatic part in which the computer controls the grab moving according to the given algorithm.

Only the automatic part of the unloading cycle can be modelled and optimized [10].

Fig. 5. Elementary sub-system »crane« with grab The manual part of the unloading cycle consists of the empty grab lowering to material surface in the vessel, from one of the three points of the end of the automatic part of the unloading cycle (Fig. 5), material grabbing and loaded grab hoisting to one of the three points of the beginning of the automatic part of the unloading cycle. The automatic part of the unloading cycle consists of a grab transfer from one of the three points of the beginning of the automatic part of the unloading cycle to the hopper, grab discharging and empty grab return transfer from the hopper to the one of the three possible points of the end of the automatic part of the unloading cycle. The begin/ end point position of the automatic part of the unloading cycle depends on a given geometry of the system (Fig. 5), water level, material level in the vessel, etc., and therefore limits the set of possible grab hoisting/lowering and the trolley moving velocities for which it is possible to achieve the automatic part of the unloading cycle.

The main purpose of elementary subsystem crane modelling in this paper is to obtain boundaries for duration of half-automatic 410

unloading cycle. The duration of the unloading cycle is: tuc = 51.87 to 63.87 s [2] and [3].

2 QUEUING THEORY BASED MODEL OF TERMINAL FOR BULK CARGO UNLOADING The terminal for bulk cargo unloading will be modelled using multi channel queuing model with finite queue and bulk arrival of the units [4] and [6].

System characteristics: • there are c = 2 servicing channels i.e. elementary sub-systems “crane”, • queue size is m = 32 places i.e. anchorage capacity (maximal number of vessels that can wait for unloading), • units are arriving into system in groups of r = 6 i.e. the number of vessels in a composition (see Table 3), • arrival rate (λ) of the vessel compositions is constant i.e. Exponentially distributed interarrival times – E1 (see 2.1), i.e. λ k = 0,1,..., c + m − r , (1)

λk =  0 k > c + m − r • system servicing rate (μ) is constant if two elementary sub-systems »crane« are working independently (see 2.2.1), while if two elementary sub-systems »crane« are working with a strategy, then the servicing rate changes in time (see 2.2.2), • queue discipline is FIFO (First-In-First-Out), • Kendall notation of this system is M[6] /M/2/32.

System state is defined according to the number of vessels in the system. State-transitionrate diagram of described system is shown in Fig. 6.

If arriving vessel composition finds that all “crane”-s are idle, then whole composition will be accepted to the system in the following way: two vessels will be immediately taken to unloading while other vessels will be placed at the “anchorage”. In the case that one “crane” is busy and one is idle, then the entire composition will be accepted to the system i.e.: one vessel will be immediately taken to unloading while the other vessels will be placed at the “anchorage”.

If all “crane”-s are busy and the number of free places at the “anchorage” is greater or equal to

Bugaric, U. ‒ Petrovic, D. ‒ Petrovic, Z. ‒ Pajcin, M. ‒ Markovic-Petrovic, G.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 405-416

the number of vessels in the composition i.e. 6, then the whole composition will be accepted to the system and placed in the “anchorage”. When all “crane”-s are busy and the number of empty places at the “anchorage” is lower than the size of vessel composition, the whole composition will be cancelled.

Initial conditions for the system of differential Eq. (1) solving are: p0(0) = 1, pi(0) = 0 for i = 1,...,34, which means that the system is empty at the beginning of the unloading process (t = 0).

Modelling of the terminal for bulk cargo unloading will be analysed through the following system performances, according to vessel compositions arrival rate to the system, such as: • probability of servicing:

Pser =

p0' (t ) = −λ ⋅ p0 (t ) + µ1 (t ) ⋅ p1 (t ),

p1' (t ) = − [ λ + µ1 (t ) ] ⋅ p1 (t ) + 2 ⋅ µ2 (t ) ⋅ p2 (t ), pi' (t ) = − λ + µi (t ) ⋅ pi (t ) + µi +1 (t ) ⋅ pi +1 (t ),

1 Peq = ⋅ ted

pi' (t ) = − λ + µi (t ) ⋅ pi (t ) + + µi +1 (t ) ⋅ pi +1 (t ) + λ ⋅ pi − 6 (t ), for i = 6, ... , 28, ... pi' (t ) = − µi (t ) ⋅ pi (t ) + + µi +1 (t ) ⋅ pi +1 (t ) + λ ⋅ pi − 6 (t ), for i = 29, ... , 33, ... ' p34 (t ) = − µ34 (t ) ⋅ p34 (t ) + λ ⋅ p28 (t ).

ted

m

∫∑p

c+k

0 k =1

(t ) ⋅ dt , (4)

average number of vessels at the anchorage: 1 Nw = ⋅ ted

ted

m

∫ ∑k ⋅ p

c+k

0 k =1

(t ) ⋅ dt , (5)

•

average time that the vessel spends at the anchorage:

•

tw = N w / λ , (6) average number of vessels in the system:

•

for i = 2, ... , 5, ...

pk (t ) ⋅ dt , (3)

k =0

0

N ws =

1 ⋅ ted

ted c + m

∫ ∑ k ⋅ p (t ) ⋅ dt , (7) k

0 k =1

average time that the vessel spends in the system: tws = N ws / λ , (8)

where:

...

∫∑

probability of a queue:

On the basis of state-transition-rate diagram, (Fig. 6), the system of linear differential equations which describes system state probabilities changing in time can be written as:

ted c + m − r

•

•

Fig. 6. State-transition-rate diagram of unloading terminal

1 ⋅ ted

λ=

1 ⋅ ted

ted c + m − r

∫∑λ 0

k =1

k

⋅ pk (t ) ⋅ dt (9)

is average arrival rate [8], and ted is experiment duration time.

(2)

2.1 Average Arrival Rate of the Vessel Compositions On the basis of statistical data connected to material shipment for unloading terminal, the weather conditions in the terminal area, absolute frequency of material appearance, organization and other conditions and the average arrival rate of the vessel composition can be obtained.

A yearly navigation period (days) on the river in the area of unloading terminal, taking into

Determining the Capacity of Unloading Bulk Cargo Terminal Using Queuing Theory

411

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 405-416

consideration disturbances when navigation is not possible caused by weather conditions like ice (38 days), fog (6 days) and wind (3 days), is: tn = 365 – 38 – 6 – 3 = 318 days [2].

The average needed quantity of material that has to be unloaded in the period of one year is Qys = 1200000 t/year (see Table 1), meaning that the average daily quantity of shipment material is: Qds =

Qys ⋅ K1 tn

1200000 ⋅1.2 = = 4528.3 t / day 318

where K1 = 1.2 is coefficient of variable shipment [2].

The fact that material is shipped in compositions of six vessels (see Table 3) gives the total quantity of material shipped with one composition (Qc): Qc = 6×Qv = 6×1700 = 10200 t.

Finally, time interval between two successive arrivals of the vessel compositions is:

= tar

Qc 10200 = = 2.2525 days Qds 4528.3

or 54.04 hours, which gives the average arrival rate of λ = 0.0185 1/h.

2.2 Unloading Rates Average time needed for vessel unloading is directly connected to the quantity of the material which has to be unloaded. In order to obtain a free digging rate and cleaning rate it will be assumed that the whole quantity of the material from the vessel can be unloaded with either a free digging and cleaning rate.

The first step in obtaining the average free digging unloading rate is to calculate the number of unloading cycles needed for unloading a vessel with different types of material in it. The number of unloading cycles needed can be calculated using the following expression:

 Qv nuc =   Vg ⋅ ρ m ⋅ k g

  , (10) 

where: Qv = 1700 [t] – vessel capacity (Table 3), Vg [m3] – grab volume for specific type of material (Table 2), ρm [t/m3] – material density (Table 1), kg – coefficient of grab loading (Table 1).

412

In Table 4 the needed number of unloading cycles for vessel unloading depending on material type (see Table 1), using Eq. (10) is shown.

Table 4. Number of unloading cycles needed for vessel unloading CoeffiMaterial Grab cient density volume Material of grab ρm Vg loading [t/m3] [m3] kg iron ore 2.7 3.2 0.7 iron ore 2.5 3.2 0.75 iron ore 2.2 3.2 0.8 limestone ≈1.52 5 0.8 coal 0.8 5 0.9

No. of unload. cycles 281 283 302 280 378

The average time for unloading vessels with different material types, under the assumption that: the whole vessel can be unloaded with a free digging rate (average unloading cycle time tuc = 0.9645 min, see Chapter 1.3) and that relative frequencies of the appearance of iron ore of different density are equal and upon relative frequency of material appearance (Table 1), is: tvufd = (0.25 ⋅ 281 + 0.25 ⋅ 283 + 0.25 ⋅ 302 + +0.167 ⋅ 280 + 0.083 ⋅ 378) ⋅ 0.9645 = 4.7362 h , which leads to average free digging unloading rate μfd = 1/4.7362 = 0.21114 l/h.

During the vessel cleaning approximately 20% of material has to be unloaded (see Chapter 1.2) and this operation (cleaning), according to experience data [2], lasts 75 minutes. Therefore, virtual time needed for unloading all the material from the vessel using such unloading regime (vessel cleaning) is tvucl = 5∙75 = 375 minutes or 6.25 h, which leads to the average cleaning rate as μcl = 1 / 6.25 = 0.16 l/h.

2.2.1 Unloading Rate in the Case of Separate “Crane” Work (˝Work without Strategy – Model 1) The average time needed for vessel unloading, in the case of separate “crane”-s work - sw, can be obtained as a sum of the average free digging time and average cleaning vessel time as:

Bugaric, U. ‒ Petrovic, D. ‒ Petrovic, Z. ‒ Pajcin, M. ‒ Markovic-Petrovic, G.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 405-416

tvusw = 0.8 ⋅ tvufd + 0.2 ⋅ tvucl = = 0.8 ⋅ 4.7362 + 0.2 ⋅ 6.25 = 5.0390

•

The only way in which time needed for a tugboat to drag an empty vessel from berth to anchorage and drag a loaded vessel back from the anchorage to empty berth, can be incorporated in a mathematical model is through servicing (unloading) time i.e. unloading rate. According to experience, time needed for vessel dragging in both ways is approximately td = 60 minutes [2], which gives a fictive vessel unloading time of 6.0390 hours.

Finally, the fictive average vessel unloading rate is μav = 1/6.039 = 0.16559 l/h.

The unloading rate, in the case of separate “crane”-s work, depending on the system state has the following form:

 µ µi (t ) =  av 2 ⋅ µav

i =1 , i ∈ [2,..., 34]

•

26.75% of material with free digging rate μfd, which will last another 76 minutes. after that, “crane” II cleans the vessel, for 75 minutes on berth 2 with a cleaning rate μcl, while “crane” I goes to berth 1 and starts unloading the vessel with a free digging rate μfd. after the cleaning of the vessel on berth 2 is finished, “crane” II goes to berth 1 and starts unloading. At this moment each “crane” has to unload another (100 – 26.5 – 20)/2 = 26.75% of material with free digging rate μfd, which will last for another 76 minutes. Afterwards, the unloading rates change cyclically.

(11)

2.2.2 Unloading Rate in the Case of “Crane”-s Work with Strategy (Model 2) The procedure of vessel unloading in the case of “crane”-s work with strategy is presented in chapter 2.2. According to this, duration of periods at which “crane”-s are unloading vessels with free digging rare μfd or with cleaning rate μcl can be determined in following way (Fig. 7): • during the unloading of the first vessel (before this system was empty) both “crane”-s unload the vessel with a free digging rate μfd until each “crane” unloads 40% of material from the vessel on berth 1. The time needed for unloading 40% of material from vessel is approximately 114 minutes.

• after that, “crane” I cleans the vessel for 75 minutes with cleaning rate μcl, while “crane” II goes to berth 2 and starts to unload the vessel with free digging rate μfd. In 75 minutes (vessel cleaning time) ”crane” II on berth 2, with free digging rate μfd will unload 26.5% of material from vessel.

• after the cleaning of the vessel on berth 1 is finished, “crane” I goes to berth 2 and starts unloading. At this moment each “crane” has to unload another (100 – 26.5 – 20)/2 =

Fig. 7. Change of vessel unloading rate μ(t) In the described “crane”-s work with a strategy, three periods can be noticed: • I period: one “crane” alone unload the vessel with free digging rate μfd, • II period: both “crane”-s unload the vessel with free digging rate μfd, • III period: one “crane” alone cleans the vessel with a cleaning rate μcl.

Time interval from the beginning of unloading to 114 minute can be divided into two parts, i.e. the first part which lasts 38 minutes and the second which lasts 76 minutes. The first part (38 minutes) replaces “I period” in the work of each “crane”, because the system is empty at the beginning and both “crane”-s start to unload the full vessel, so the time needed is equal to one half of duration of “crane” work in “I period”.

On the basis of the above mentioned, it is possible to establish the change in the unloading rate as a function of time for each “crane”. The change of unloading rates as a function of time is a discontinuous function. The change of the unloading rate per time (minutes) for “crane” I, μI(t), takes the following form: (Fig. 8)

Determining the Capacity of Unloading Bulk Cargo Terminal Using Queuing Theory

413

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 405-416

 µ fd ; 0 < t ′ ≤ 76;  µ I (t ) =  µcl ; 76 < t ′ ≤ 151; (12)  µ ; 151 < t ′ ≤ 302;  fd

Fig. 8. Change of unloading rate of “crane” I per time The change of the unloading rate per time (minutes) for “crane” II μII(t), has the following form (Fig. 9):  µ ; 0 < t ′ ≤ 227; (13) µ II (t ) =  fd  µcl ; 227 < t ′ ≤ 302;

3 RESULTS OF TERMINAL MODELLING Initial conditions for system (unloading terminal) analysis are that the system (anchorage) is empty and all servicing facilities (crane-s) are idle. The results are obtained for basic arrival rate of the vessel compositions to the system (λ = 0.0185 1/h) and its increase by 1.5, 2, 2.5, 3, 3.5 and 4 times, for one year period (ted = 8760 hours).

The results of the terminal modelling, according to vessel compositions arrival rate to the system are: probability of servicing Pser (Fig. 10), probability of existing a queue Peq (Fig. 11), the average number of vessels at the “anchorage” Nw (Fig. 12), average time that the vessel spends at the anchorage tw (Fig. 13), the average number of vessels in the system Nws (Fig. 14), the average time that the vessel spends in the system tws (Fig. 15).

where: t’ [min] – is reduced time to “I period” of “crane”-s work and can be calculated as:  t − 38  t ′ = (t − 38) −   ⋅τ ; (14)  τ  where: t [min] current time in the system, and τ @ 302 min – duration of interval in “crane”-s work which it is successively repeated, in the case of “crane”-s work with strategy, (I period + 2×II period + III period).

Fig. 10. Probability of servicing

Fig. 9. Change of unloading rate of “crane” II per time At the end the change of vessel unloading rate, in the case of “crane”-s working with a strategy, can be obtained as (Fig. 7):

μi(t) = μI(t) + μII(t), i = 1, ..., 34.

(15) Fig. 11. Probability of a queue 414

Bugaric, U. ‒ Petrovic, D. ‒ Petrovic, Z. ‒ Pajcin, M. ‒ Markovic-Petrovic, G.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 405-416

in Figs. 10 to 15 are represented by “-*-“, while when work of “unloading mechanization” is modelled with a strategy (model 2), the results in Figs. 10 to 15 are represented by “-+-“.

Fig. 12. Average number of vessels at the anchorage

Fig. 15. Average time that vessel spends in the system

Fig. 13. Average time that vessel spends at the anchorage

Fig. 14. Average number of vessels in the system When work of “unloading mechanization” is modelled without strategy (model 1), the results

Figs. 10 to 15 alongside show the obtained results using the queuing theory and the results of simulation modelling given in [3].

It has been shown that the characteristics of the obtained results using the queuing theory and simulation are the same.

The results differ due to the limitations of the queuing theory as: • The servicing process is restricted to one phase; • The average vessel unloading time is fixed; • The influence of different type of materials; on vessel unloading time, is mathematically calculated as average; • The vessel composition is fixed to one size; • Only one vessel size is used.

The presented simulation results also differ because the servicing process is divided into three phases and all other parameters are presented by appropriate stochastic distributions [3].

All new results are on the safe side so that by using them the designer will deal with lover system capacity.

A high probability of servicing vessel compositions (0.95 and higher) is required in the work of terminals for bulk cargo unloading. If the system is modelled by the queuing theory, using model 1, then the maximal increase of the basic arrival rate of vessel compositions to the system

Determining the Capacity of Unloading Bulk Cargo Terminal Using Queuing Theory

415

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 405-416

can be 2.5 times, while if the system is modelled using simulation the maximal increase is 3 times.

When model 2 is used the basic arrival rate of vessel compositions to the system can be 3 times higher using the queuing theory modelling and 3.5 times using simulation modelling.

The above presented results, once again, show that the unloading capacity of the terminal for bulk cargo unloading is higher when “unloading mechanization” works with a strategy, for the same capital costs. The same capital costs means: the same number of unloading devices – “crane”-s which is reflected in the length of the operative coast, the same capacity of the anchorage etc.

4 CONCLUSION During the sailing period the river terminal for bulk cargo unloading works 24 hours, seven days a week. Even a very small cut-down of the duration time needed for vessel unloading, can save the energy needed for unloading and can minimize expenses which are caused by unsatisfactory servicing rate i.e. stay of the vessels at the system (terminal) longer than it is allowed [3].

The paper discusses the results of analytical modelling using the queuing theory and compares them with the simulation modelling result previously given in [3] and underlines and explains their differences.

The task of the engineers involved in design process is to define the capacity of the system, so that system input/output can be estimated.

Due to this, there is a need for developing a simpler and quicker design approach with appropriate accuracy which is the mean reason why the presented analytical model, using the queuing theory, has been developed. In other words, the newly developed model can replace the previous one, which was a more complicated simulation model at the beginning of the system design process.

The obtained results, using the queuing theory, can be used at the beginning of the design process for the first predictions of the system boundaries when rough estimations of the system behaviour is needed, while simulation modelling

416

results are to be used in the phase of detail design process.

5 REFERENCES [1] Agerschou, H., Lundgren, H., Sørensen, T. (1983). Planning and design of ports and marine terminals. John Wiley & Sons, New York.

[2] Bugaric, U. (1996). Contribution to optimization of bulk cargo unloading processes at river ports - M.Sc. Thesis, Belgrade: Faculty of Mechanical engineering.

[3] Bugaric, U., Petrovic, D. (2007). Increasing the capacity of terminal for bulk cargo unloading. Simulation Modelling Practice and Theory, vol. 15, p. 1366-1381.

[4] Clymer, J. (1990). System analysis using simulation and Markov models. Prentice-Hall International Inc., New Jersey.

[5] Comer, E., Taborga, P.N. (1987). Port simulation model (PORTSIM) - User’s manual. The World Bank, Washington.

[6] Cooper, R. (1981). Introduction to queuing theory. North Holland, New York.

[7] El Sheikh, A.A., Paul, J.R., Harding, S.A., Balmer, W.D. (1987). A MicrocomputerBased Simulation Study of a port. J. Opl. Res. Soc., vol. 38, p. 673-681.

[8] Hiller, F.S., Lieberman, G.J. (2001).

Introduction to operations research (7th edition). McGraw-Hill, New York.

[9] Kondratowicz, J.L. (1990). Simulation methodology for intermodail freight transportation terminals. Simulation, vol. 51, p. 49-57.

[10] Oyler, F. (1977). Handling of bulk solids in ocean ports. Wohlbier, R.H. (ed.), Stacking Blending Reclaiming. Trans Tech Publications, Clausthal.

[11] Park, S.C., Noh, D.Y. (1987). A port simulation model for bulk cargo operations.

Simulation, vol. 48, p. 236-246.

[12] Sinowczik, M. (1977). EUROPORT - A bulk handling plant for iron ores and pallets.

Wohlbier, R.H. (ed.), Stacking Blending Reclaiming. Trans Tech Publications, Clausthal.

Bugaric, U. ‒ Petrovic, D. ‒ Petrovic, Z. ‒ Pajcin, M. ‒ Markovic-Petrovic, G.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 417-424 DOI:10.5545/sv-jme.2010.059

Paper received: 15.03.2010 Paper accepted: 18.01.2011

Investigation of TiN Coated CBN and CBN Cutting Tool Performance in Hard Milling Application

Taylan, F. – Çolak, O. – Kayacan, M.C. Fatih Taylan1 – Oğuz Çolak1,* – Mehmet Cengiz Kayacan2 1 Suleyman Demirel University, Faculty of Technology, Department of Manufacturing Engineering, Turkey 2 Suleyman Demirel University, Faculty of Engineering, Department of Mechanical Engineering, Turkey Cubic Boron Nitride (CBN) and TiN Composite Coated CBN combines the thermal stability, super abrasiveness and cost effectiveness for hard machining applications. This paper reports the results of a study addressing wear performance of these CBN and TiN based coated CBN inserts (SNMN090308) for face milling of 61 HRC hardened 90MnCrV8 tool steel. Machining test conditions are obtained after dynamic stability simulation of cutting tools and machine tools. The tool wear and cutting forces are also analyzed and the results are presented. © 2011 Journal of Mechanical Engineering. All rights reserved.

Keywords: tool wear, hard milling, CBN tools, coated CBN 0 INTRODUCTION Hard milling has been recently employed to hardened steels (> 30 HRC) used for dyes and mould making industry [1].

Generally, in hard milling research fields have focused on tool life [2], surface integrity [3], white layer effects [4], chip formation [5], cutting force models [6], stability of hard milling [6] and optimal cutting parameters [7]. However, very few studies have dealt specially with the dynamic cutting mechanisms induced by hard milling such as the effects of machine tool stability. Better understanding of machining stability can lead to a better process economics, increased process stability, improved tool life, reduced tooling costs and enhanced surface integrity and component performance.

Many researchers have studied cubic boron nitride (CBN) tools in the hard milling of steel alloys [8] to [12]. Generally CBN cutting tools have greater wear resistance than other tool materials due to their high degree of hardness [8]. Raghavan [10] was used PCBN to face mill AIS1 HI3 (48 to 50 HRC) tool steel at cutting speeds of 100 to 200 m/min, feed of 0.1 mm/ tooth and 1.0 mm depth of cut. Heath et al. [11] recommended the use of PCBN at a cutting speed of 180 m/min, a feed of 0.2 mm/tooth and 1.0 mm depth of cut to face mill cold work tool steel (60 HRC). Nakagawa et al. [12] has reported the

effect of cutting fluid when high speed ball nose end milling 57 HRC tool steel AISI D2 using polycrystalline cubic boron nitride (PCBN) tools.

The application of a water-based cutting fluid led to catastrophic tool failure due to thermal shock, however, indirect supply of oil on the surface of the work material, rather than directly at the cutting edge, resulted in a longer tool life as compared to dry cutting. They have reported a feed/tooth of 0.05 mm, and axial/radial depths of cut of 0.05 and 0.2 mm, respectively, a length cut of 400 m for a flank wear of approximately 0.1 mm using oil as the cutting fluid at a cutting speed of 222 m/min.

58 HRC AISI D2 hardened tool steel is face milled by Koshy et al. [13] with a round CBN insert. They found acceptable tool life (43 cm3 of removed material) together with excellent surface finish in the range of 0.1 to 0.2 μm in Ra. This study points out that tool life would need to be extended to make the process economically viable.

CBN and coated CBN tools are still very expensive and it is difficult to find better cutting performance [14]. In our study the use of different CBN Grade is proposed.

In this study, cutting performance of the TiN coated CBN and CBN cutting tools for 61 HRC hardened 90MnCrV8 cold work tool steel face milling is investigated with considering hard milling stability analysis. Wear performance

*Corr. Author’s Address: Suleyman Demirel University, Faculty of Technology, 32260, Isparta, Turkey, ocolak@mmf.sdu.edu.tr

417

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 417-424

of CBN and TiN based coated CBN inserts (SNMN090308) for face milling of the hardened tool steel is examined. The tool wear formation and cutting forces are also analyzed and the results are presented.

1 EXPERIMENTAL METHODOLOGY Machining experiments are performed on a Harford VMC 1020 machining centre to ensure precisely computer controlled cutting conditions.

The depth of cut, the feed rate and cutting speed are selected by simulating analytical chatter stability analysis. Stability analysis procedure of hard face milling test is shown in Fig. 1.

X and Y direction of cutting tool dynamic characteristics is measured with tap testing. 2000 N Kistler 9722A Hammer and 100 g/V sensitivity accelerometer is used for tap testing. Frequency response function calculation (FRF) cutting tool is calculated in both directions using CutPRO MALTF software.

In Step-2 cutting coefficients are defined.

In this step, the slot milling test is done to define the average cutting coefficient. Table 1 shows the test conditions of hard milling.

Table 1. Average cutting coefficient test conditions [15] Test Feed rate Spindle Cutting Axial no [mm/ speed speed depth of tooth] [rpm] [m/min] cut [mm] 1 0.05 2 0.075 3 0.1 850 50.87 0.3 4 0.125 5 0.150 Average cutting forces are calculated by using measured cutting forces in three dimensions for each test condition. Average cutting forces are used for calculating average cutting coefficient model [16] which is given in Eq. (1).

Fig. 1. Procedures of face milling stability analysis

Fig. 2. Structural dynamic measurement of cutting tool From Fig. 1 it is seen that Step-1 machine tool-workpiece structural dynamic measurement is done using the tap testing method. Fig. 2 shows tap testing of cutting tool.

418

π Fye , K te = , Na Na −4 Fxc −π Fxe K rc = , K re = , (1) Na Na 2F π Fzc K ac = , K ae = ze .

Na Na K tc =

4 Fyc

In Eq. (1), N is the number of teeth, a is axial depth of cut [mm], Fxc , Fyc , Fzc are average cutting force in x, y and z directions respectively (Newton), Fxe , Fye , Fze are average edge cutting force in x, y and z directions (Newton), Krc, Ktc, Kac are cutting coefficients of tool-material interface in radial, tangential and axial directions [N/mm2], Kre, Kte, Kae are cutting tool edge cutting coefficients in radial, tangential and axial directions [N/mm].

After the slot milling test, the average cutting coefficient of 90MnCrV8 is found. Sub indices (c) and (e) represent shear and edge force

Taylan, F. – Çolak, O. – Kayacan, M.C.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 417-424

components, respectively. The edge cutting Kre, Kte and Kae are constants and related to the cutting edge length dS given in an infinitesimal length (ds) of a helical cutting edge. The shear force coefficients Krc, Ktc and Kac are identified either mechanistically from milling tests conducted at a range of feed rate. Table 2 shows average cutting coefficients.

In Step-3 stability of hard face milling experiment is simulated by using cutting tool frequency response function calculation (FRF) (Step-1) and average cutting coefficients (Step2). Stability simulation is calculated using Eq. (2) [16].

alim = −

2πΛ R (1 + κ 2 ) . (2) NK t

Stability lobe diagram is given Fig. 3.

In Step-4 the tool wear test condition which is given Table 3 is determined after stability simulation. Axial depth of cut is fixed as 0.6 mm related to stability simulation which is shown in Fig. 3.

The chemical properties of workpiece material are given in Table 4. Workpiece material, steel 90MnCrV8 (AISI – O2, EU – 90MnCrV8) is a cold work tool steel, with high dimensional stability at heat treatment, very high resistance to cracking, high machinability, medium toughness and resistance to wear. Hardness after annealing is max 229 HB. After quenching, the hardness achieved may be from 63 to 65 HRC. The field of application of 90MnCrV8 is compress measuring tools, machine knives for the wood, paper and metal industry, cold cutting shear blades, thread cutting tools [17].

Table 2. Average cutting coefficients of 62 HRC 90MnCrV8 tool steel [15]

Average Cutting Coefficients

Kte [N/mm] 581.35

Kre [N/mm] -724.47

Kae [N/mm] 158.96

Ktc [N/mm2] 3689.11

Krc [N/mm2] -5804.89

Kac [N/mm2] 2257.14

Fig. 3. Analytical Stability Lobes diagram of hard milling test Table 3. Stable tool wear test conditions for 90MnCrV8 Face milling [15] Spindle speed rpm Feed rate mm/tooth

2650 0.05

2280 0.075

1760 0.1

1405 0.15

1170

1005

875

Investigation of TiN Coated CBN and CBN Cutting Tool Performance in Hard Milling Application

780

419

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 417-424

The tests were discontinued when a flank wear criterion VB of 700 µm was reached. For cutting tool wear test chip volume is fixed 5896.8 mm3. Flank wear measurements were taken in accordance with the International Standard ISO 8688-1 [19] and using the Olympus TM optical microscope and SEM. Cutting forces were recorded for every test in Fx, Fy and Fz directions. For this purpose, a 9257B-type Kistler dynamometer was used.

Two type CBN tool is tested in a hard milling test. Properties of cutting tools are given in Table 5.

Hard milling of 90MnCrV8 cold work tool steel by CBN is characterized by the flow of chips at very high temperature and extremely deformed signs of intensive shearing at cutting edge (Fig. 4).

In this study a different type of chipping tool wears is investigated. This wear formation is given in Table 7. As seen from Table 7 TiN Composite Coated CBN grade cutting tools are four times braked than CBN grades. Milling operations are known as interrupt cutting so that TiN Composite Coated CBN grades is not useful for hard milling application.

Table 4. Chemical compositions of 90MnCrV8 [18] C 0.88

Chemical Composition (% weight) Si Mn Cr P S V 0.29 2.07 0.26 0.024 0.009 0.08 2 RESULTS AND DISCUSSION

2.1 Tool Wear Damage of the cutting tools over the entire range of cutting conditions was mainly in the form of chipping. Chipping tool wear measurement ISO 8688 standard of face milling conditions is given in Table 6.

Fig. 4. Heat dissipation through chip while machining, Vc = 278.1 m/min Fig. 5a shows the change in depth of the chipping wear with a cutting distance for the

Table 5. Cutting tools properties [20] Tool Properties ISO CODE Compositions

CBN300 Grade SNEN0903ENE-M06 CBN content approx. 90 vol % Average starting grain size (µm) – 22 Binder – Al ceramic Format – Solid

Physical Properties

Knoops hardness GPa – 30.4 Thermal conductivity [Wm-1K-1] (20 °C) – 130 Size L – 9.525 mm s – 3.18 mm B – 0.9 mm re – 0.8 mm Rake angle – 0°

Tool Geometry

420

CBN300P Grade SNMN090308 CBN content approx. 90 vol % Average starting grain size (µm) – 22 Binder – Al ceramic Format – Solid Coated Ti(C,N) + (Ti, Al) N + TiN with a total thickness of 2-4 µm Knoops hardness GPa – 30.4 Thermal conductivity [Wm-1K-1] (20 °C) – 130 Size L – 9.525 mm s – 3.18 mm B – 0.9 mm re – 0.8 mm Rake angle – 0°

Taylan, F. – Çolak, O. – Kayacan, M.C.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 417-424

Table 6. ISO 8688 Face milling chipping measurement standard

For y or z with corresponding length values

Illustration th Leng

A

Width

A-A

0.2 0.25 -

0.25 0.4 -

0.3 0.5 -

Depth

A B C

Chipping (breakage) Uniform Non-uniform Localized Length [mm] Micro-chipping <0.3 Macro-chipping 0.3-1 Breakage >1

Criteria [mm] N L (normal) (large)

z

CH 1 2 3

S (small)

y

Tool Deterioration Phenomena

A

Table 7. Tool wear type of TiN coated CBN and CBN tool in hard milling conditions CBN tool TiN Composite Coated CBN

Breakage ISO type C 2 tools 9 tools

Macro chipping ISO type B 30 tools (1 Small, 13 Normal, 16 Large) 23 tools (2 Small, 10 Normal, 11 Large)

a)

b) Fig. 5. Wear depth and length in the hard milling tests at various feed [mm/tooth] and cutting speeds [m/min]; a) CBN tools, b) coated Ti(C,N) + (Ti, Al) N + TiN CBN tools Investigation of TiN Coated CBN and CBN Cutting Tool Performance in Hard Milling Application

421

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 417-424

a)

b)

c) d) Fig. 6. Tool wear SEM photo images; a) Maximum tool wear for six insert CBN tools (vc = 154.4 m/min, f = 0.15 mm/teeth), b) Maximum tool wear for six insert Coated Ti(C,N) + (Ti, Al) N + TiN CBN tools (vc = 154.4 m/min, f = 0.1 mm/teeth), c) Minimum tool wear for six insert CBN tools (vc = 524.5 m/min, f = 0.05 mm/teeth), d) Minimum tool wear for six insert Coated Ti(C,N) + (Ti, Al) N + TiN CBN tools (vc = 524.5 m/min, f = 0.05 mm/teeth) tools grade CBN in the machining of 90MnCrV8 cold work tool steel at various depths of feed and cutting speeds. The change in the chipping wear depth with various feed and cutting speeds for coated Ti(C, N) + (Ti, Al) N + TiN can be seen in Fig. 5b. It was found that at low cutting speed and high feed rate, the chipping wear increased significantly and at high cutting speed and low feed rate tool wear is decreased. As seen in Fig. 3 CBN grade is more stable depth and length tools wear rates than TiN Composite Coated CBN due to more breakage of TiN Composite Coated CBN. 422

The chipping wear decreased significantly when the cutting speed was increased from 300 to 525 m/min in both tool types. SEM images of maximum tool wear rate for each inserts are given Fig. 6. Maximum tool wear is obtained at low cutting speed which is 154.4 m/min. For both cutting tool grade minimum tool wear is obtained at 524.5 cutting speed and 0.05 mm/teeth feed rate. Maximum tool wears and minimum tool wear SEM photographs are illustrated in Fig. 6.

Taylan, F. – Çolak, O. – Kayacan, M.C.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 417-424

b) a) Fig. 7. Average feed cutting forces (Fx), a) with different cutting speed and feed rate for test conditions CBN tool, b) coated Ti(C,N) + (Ti, Al) N + TiN CBN tool

2.2 Cutting Forces The average cutting forces (feed forces Fx) obtained with tool grades CBN and coated Ti(C,N) + (Ti, Al) N + TiN CBN are compared in Figs. 7a and b. It was found that when the chipping was increased (due to an increase feed rate or a reduction in the cutting speed), it also coincided with an increase in the cutting forces.

Optimum cutting speed and feed rate were found for both cutting tool grades between 450 to 550 m/min (means high speed) and 0.05 to 0.1 mm/teeth (means low feed rate).

Cutting forces were significantly increased with high speed and high feed rate.

In this study white layer effects and chip formations of hard milling operation is not considered.

3 CONCLUSIONS

4 REFERENCES

The following results can be extracted from the hard face milling of 90MnCrV8 tool steel by CBN and PCBN insert tools.

The face milling of 90MnCrV8 tool steel in the hardened state (61 HRC) were shown tool life values of <260 mm length of cut. Little difference was found in tool wear between the CBN insert tools and coated Ti(C,N) + (Ti, Al) N + TiN CBN insert.

An analysis of flank wear patterns indicated that macro chipping, governing mechanisms were responsible for tool wear. PCBN tools failed by fracture of the cutting edge.

In CBN cutting tools wear test 6.25% tool breakage is investigated. However, coated Ti(C,N) + (Ti, Al) N + TiN CBN cutting tools wear test 28.13% tool breakage was investigated.

[1] Zhang, S., Guo, Y.B. (2009). An experimental analytical analysis on chip morphology, phase transformation, oxidation, and their relationships in finish hard milling.

International Journal of Machine Tools & Manufacture, vol. 49, p. 805-813.

[2] Chandrasekaran, H., Saoubi, R.M. (2006).

Improved machinability in hard milling and strategies for steel development. Annals of CIRP, vol. 55, no.1, p. 93-96.

[3] Axinte, D.A., Dewes, R.C. (2002). Surface integrity of hot work tool steel after high speed milling-experimental data and empirical models. Journal of Materials Processing Technology, vol. 127, no. 3, p.

325-335.

[4] Pang, J.Z., Wang, M.J., Duan, C.Z. (2007).

White layer and surface roughness in

Investigation of TiN Coated CBN and CBN Cutting Tool Performance in Hard Milling Application

423

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 417-424

high speed milling of P20 steel. Advanced Materials Research, vol. 24-25, p. 45-54.

[5] Dolinsek, S., Ekinovic, S., Kopac, J. (2004).

A contribution to the understanding of chip formation mechanism in high-speed cutting of hardened steel. Journal of Materials Processing Technology, vol. 157-158, p. 485490.

[6] Altintas, Y., Weck, M. (2004). Chatter stability of metal cutting and grinding. CIRP Annals - Manufacturing Technology, vol. 53, no. 2, p. 619-642.

[7] Ghani, J.A., Choudhury, I.A., Hassan, H.H. (2004). Application of Taguchi method in the optimization of end milling parameters.

Journal of Materials Processing Technology, vol. 145, no. 1, p. 84-92.

[8] Camuscu, N., Aslan, E. (2005). A comparative study on cutting tool performance in end milling of AISI D3 tool steel. Journal of Materials Processing Technology, vol. 170, p.

121-126.

[9] Ateş, S., Er, A.O., Aslan, E., Camuşcu, N. (2006). Sertleştirilmiş P20 Kalip Çeliğinin Kübik Bor Nitrür Kesici Takimlarla Yüksek Hizlarda Frezelenmesi, 12th International Congress of Mechanical Design and Manufacturing, Kusadasi. (In Turkish) [10] Raghavan, K. (1988). Face milling of hardened die steel using CBN tooling. MSc. Project Report, Graduate School of Machine Tool and Manufacturing Technology, Department of Mechanical Engineering, University of Birmingham.

[11] Heath, P.J. (1987). Ultra hard tool materials, in machining, ASM Metals Handbook, 9th ed., vol. 16, p. 105-117.

424

[12] Nakagawa, T., Ikeda, T., Matsuoka, T. (1995). High speed milling of steel and tool life. Proceedings of the 8th International Conference of the Tool, Die and Mould Industry, Barcelona.

[13] Koshy, P., Dewes R.C., Aspinwall D.K. (2002). High speed end milling of hardened AISI D2 tool steel (similar to 58 HRC). J Mater Process Technology, vol. 127, p. 266273.

[14] Seco Tools (2007). Secomax PCBN technical information leaflet 00:I, available online at http://www.secotools.com (accessed on 10/2009) [15] Taylan, F. (2009). The Determination of Tool Wear Behavior in Milling of Hard Materials.

PhD Thesis, Graduate School of Natural and Applied Sciences, Suleyman Demirel University.

[16] Altıntaş, Y. (2000).

Manufacturing Automotion. Cambridge University Pres, Vancouver.

[17] Ekinovic, S., Dolinsek, S., Begovic, E. (2005). Machinability of 90MnCrV8 steel during high-speed machining. Journal of Materials Processing Technology, vol.

162.163, p.603-608.

[18] K720 DIN 1.2842 (90MnCrV8) Böhler tool steels www.osmanlicelik.com (accessed on 10/2009).

[19] International Standard ISO 8688-1, Tool life testing in milling-face milling, 1st ed. 198905-01.

[20] Seco Tools Milling Tool Catalogue. (2008).

Taylan, F. – Çolak, O. – Kayacan, M.C.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 425-439 DOI:10.5545/sv-jme.2009.055

Paper received: 24.09.2009 Paper accepted: 07.01.2011

Conservative-Force-Controlled Feed Drive System for Down Milling Tadic, B. – Vukelic, D. – Hodolic, J. – Mitrovic, S. – Eric, M. Branko – Djordje Vukelic2,* – Janko Hodolic2 – Slobodan Mitrovic1 – Milan Eric1 1 Faculty of Mechanical Engineering, University of Kragujevac, Serbia 2 Faculty of Technical Sciences, University of Novi Sad, Serbia Tadic1

Reviewed in this paper are the results of theoretical and experimental investigation of a novel down milling method. The method is based on controlling feed speed using conservative forces, where the active force which provides motion is the horizontal component of the cutting force. Feed motion is therefore realized without any active forces (active drive systems), while the feed speed is regulated by precision breaking of the workpiece, which is possible through hydraulic damping. In addition to the basic theoretical model of the feed drive system, the paper also presents the dynamic model of the drive system, taking into consideration fluid compressibility. Based on the model proposed, a physical instance of the feed drive system was designed, built and tested. The results of preliminary experimental investigation speak in favour of the proposed theoretical model, which enables practical application of this type of feed drive systems. © 2011 Journal of Mechanical Engineering. All rights reserved.

Keywords: down milling, special feed drive system, continuous feed speed control, damping 0 INTRODUCTION Milling is a basic machining process by which a surface is generated progressively by the removal of chips from a workpiece as it is fed to a rotary cutter in a direction perpendicular to the axis of the cutter [1]. Surfaces can be generated by two methods: up milling (the cutter rotates against the direction of the feed of the workpiece) and down milling (the cutter rotates in the same direction of the feed of the workpiece). Both methods have some advantages and disadvantages [2].

In the past few years major attention has been given to research of the milling process.

These research works refer to various aspects such as: influence of cutting parameters on surface roughness [3] to [5], dynamic problems in milling [6] to [8], tool wear in milling operations [9] and [10], stability of milling process [11] to [13], calculations of chip thickness in milling operations [14], optimal tool geometry selection [15], optimal tool path selection [16] to [18], optimization of machining parameters in milling using geometric programming [19], optimization of machining parameters in milling using genetic programming [20], optimization of machining parameters in milling operations using genetic

algorithm [21] to [23], optimization of machining parameters in milling operations using artificial neural networks [24] and [25], optimization of machining parameters in milling operations using a combination of the genetic algorithm and artificial neural networks [26] and [27], optimization of machining parameters in milling operations using a combination of artificial neural networks and particle swarm [28].

This extensive research volume testifies of the importance of milling as machining technology. In spite of this, some core principles underlying kinematics and statics of cutting, have remained unchanged. The shape realized in any milling process depends on three important factors; tool geometry, workpiece geometry and relative motion between tool and the workpiece.

The relative motion between tool and workpiece has two components. The first component of motion consists of the rotation of the tool around its own axis and the second is a feed motion of the workpiece relative to the tool. A combination of these two motions together with tool geometry and its interaction with the workpiece geometry defines the final shape of workpiece [1] and [2].

Having this in mind, a particularly significant machine tool component is the feed drive, which provides constant feed motion of the workpiece

*Corr. Author’s Address: University of Novi Sad, Faculty of Technical Sciences, Trg Dositeja Obradovica 6, 21000 Novi Sad, Republic of Serbia, vukelic@uns.ac.rs

425

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 425-439

during machining. Considering the influence of feed motion on the surface quality and tool life, an important question is whether feed speed should be kept constant during the pass of a cutter tooth through workpiece material.

Analysis of dynamic movement equations leads to a conclusion that down milling can be performed in a completely different manner. The fact that during down milling the feed speed vector remains co-linear with the horizontal component of the cutting force, leads to an idea that conservative force can be used to control feed motion. Once the motion is initiated, the active cutting force maintains feed motion.

Horizontal component of the cutting force can be used as an active force. Feed speed is controlled by an adequate conservative force. In that case, feed speed is not constant but depends on the magnitude of the horizontal component of the cutting force as well as on the characteristics of passive components, which serve to arrest the workpiece. Such a solution of the feed drive allows a braking system whose reaction-time is the magnitude of a hundredth of a second (under the influence of impulse forces), resulting in a very short braking path. With this type of feed speed control, the breaking path corresponds to the tooth pitch in conventional milling. Although it satisfies stringent demands, the technical solution of the conservative-force-controlled feed drive

system is much simpler than that of conventional, mechanical systems, since braking is performed by simple hydraulic damping.

1 MODEL OF FEED DRIVE SYSTEM With conventional milling machines, feed drives are designed as special transmission drives (gearbox, DC motor and recirculating-ball drive, etc.) which provide constant feed motion regardless of the magnitude of resisting forces which appear during cutting. At a constant feed speed, the workpiece is moving during cutting (i.e. during the passage of cutter tooth through workpiece material). In this case, during effective cutting time (tooth engagement) the relief surface of tooth cutting wedge is additionally loaded with an external force, which provides movement for the milling machine working table and workpiece.

In comparison to conventional milling, the proposed milling method has completely different basics. It uses active cutting forces to produce feed motion, while feed speed is controlled using passive hydraulic components. Conceptual design of the proposed machining method is illustrated in Fig. 1.

If the workpiece is set in motion with the initial speed ν0, at a given moment in time, the cutter, rotating at a constant cutting speed, is going to engage with a segment of workpiece material.

Fig. 1. Schema of the proposed feed drive system 426

Tadic, B. – Vukelic, D. – Hodolic, J. – Mitrovic, S. – Eric, M.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 425-439

Horizontal component of the cutting force (Fig. 1), can be expressed by formula:

Fh (ϕ ) = Ft (ϕ ) sin ϕ − Fr (ϕ ) cos ϕ , (1)

where Ft(φ) and Fr(φ) are tangential and radial components of the cutting force, respectively.

The radial component of the cutting force can be expressed as:

Fr (ϕ ) = K ⋅ Ft (ϕ ) , (2)

where K is a constant which defines the ratio between radial and tangential components.

Substituting Eq. (2) into (1) gives:

Fh (ϕ ) = Ft (ϕ ) ( sin ϕ − K ⋅ cos ϕ ) . (3)

Analysis of Eq. (3) leads to a condition of movement which is derived from the following Eqs.: Fh (ϕ ) > 0 (4)

Ft (ϕ ) > 0

(5) K < tg ϕ

where φ ranges between φ0 ≤ φ ≤ 90°. Bearing in mind the cutting depths which are most frequently used in milling, it can be concluded that the condition of movement has been satisfied in most cases.

According to Fig. 1 current chip thickness is expressed as:

δ (ϕ ) = f1 ⋅ cos ϕ , (7)

where f is the movement of the workpiece (feed per tooth). Using known expressions for specific cutting resistance and tangential force, the horizontal component of the cutting force can be expressed as follows:

Fh (ϕ ) = Bo ( sin ϕ − K cos ϕ ) cos ϕ (

1−1 ε k )

Bo = Ck ( b ⋅ f1 )

1−

1 εk

, (8)

∑

where vk denotes piston speed (i.e. workpiece speed), p1 is pressure in the piston front-end, v2 is speed of fluid in cross-section II-II, p2 is pressure in the piston aft-end, Σhm are total losses of mechanical energy on the streaming trajectory, from cross-section I-I to cross-section II-II, g is gravitational constant and ρ is fluid thickness. All mechanical energy losses along fluid streaming path can be disregarded as lower order values in comparison with the losses in the throttling valve. From the continuity equations for the crosssections:

ρ ⋅ vk ⋅ Ak = ρ ⋅ v2 ⋅ A2 , (11)

follows the speed of fluid through the throttling valve (cross-section II-II): v2 =

Ak vk , (12) A2

where Ak is the active surface area of piston, A2 is the active surface area at the point of damping.

Considering the surface areas ratio (Ak/A2) all the remaining losses in mechanical energy can be disregarded, compared to the losses in throttling valve, thus leading to:

II − II

∑ I −I

hm = ξ g

v22 ≈ 2g

∑h

m

, (13)

where ξg denotes mechanical energy loss coefficient. Pressure drop is determined as the function of external load, using:

F (φ ) p − p2 ∆p = 1 = h . (14) ρ⋅g ρ⋅g ρ ⋅ g ⋅ Ak

Substituting Eqs. (12), (13) and (14) into Bernoulli’s Eq. (10) leads to: 2

, (9)

where b is cutting width, while Bo, Ck and εk are constants depending on workpiece material.

Force Fh(φ) creates the difference in pressures (p1 – p2) within the cylinder, so that Bernoulli equations applied to cross-sections I-I and II-II give approximate equality of the following form:

II − II vk2 p v2 p + 1 = 2 + 2 + hm , (10) 2g ρ ⋅ g 2g ρ ⋅ g I − I

vk2 F (ϕ )  Ak  vk2 + h −  − 2 g ρ ⋅ g ⋅ Ak  A2  2 g 2

 A  v2 −ξ g  k  k = 0.  A2  2 g

(15)

Piston (workpiece) speed can be determined from Eq. (15), as the function of external load, fluid thickness and characteristic

Conservative-Force-Controlled Feed Drive System for Down Milling

427

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 425-439

dimensions of the hydraulic assembly, using the following expression:

vk =

2 Fh (ϕ )  A  2  ρ ⋅ Ak  k  1 + ξ g − 1  A2  

(

. (16)

)

There are two boundary states for feed motion.

1. The first boundary state pertains to a fully blocked system, when the throttling valve is closed (A2→0), turning Eq. (16) into: vk1 = lim vk = A2 → 0

=

2 Fh (ϕ )  A   ρ ⋅ Ak  k  (1 + ξ g ) − 1  ε   2

= 0 , (17)

vk 2 = lim vk = A2 = Ak Ak → 0

2 Fh (ϕ ) → ∞ . (18) ρ ⋅ε ⋅ξg

The first boundary state, with vk = 0, is practically achievable, and it can be proved that it is practically possible to create hydraulic design which provides continuous selection of feed speeds for all real cutting conditions. For down milling, it is the continuous selection of mean feed speed because feed speed is the function of horizontal cutting force component.

Eq. (16) can be expressed in following form: vk = v p = Rk Fh (ϕ ) , (19)

where Rk, the constant of feed speed regulation, is expressed as:

Rk =

2  A  2  ρ ⋅ Ak ⋅  k  1 + ξ g − 1  A2  

(

. (20)

)

This is called the constant of feed speed regulation as it can be set to any value from zeroup to values which fit current cutting conditions.

428

is derived:

Ω

dx = Rk Fh (ϕ ) , (23) dϕ

where Ω is the angular velocity of milling cutter.

Based on the expression derived, it is possible to calculate total displacement of workpiece during single tooth cutting: x2

∫

f1 = dx =

x1

where ε is an infinitely small value.

2. The second boundary state is a theoretical state in which A2 = Ak and Ak → 0, which means that:

Feed speed vp can be defined as the first derivative of traversed path, i.e. of coordinate x in time: dx v= = Rk Fh . (21)

p dt Using: dx dϕ dx dx

, (22) = =Ω dt dt dϕ dϕ

1 Rk Ω

π /2

∫

Fh (ϕ ) dϕ (24)

ϕo

that is, by substituting Fh(φ) from Eq. (8) into Eq. (24), there follows: R f1 = k ⋅ Ω

. (25)  1  π /2 ⋅∫

1−  εk 

Bo ( sin ϕ − K cos ϕ ) cos ϕ 

dϕ

ϕo

Should the value of this integral be denoted with I1, there follows: R

(26) f1 = k I 1 . Ω The value of integral I1 corresponds to the area shown in Fig. 2. The size of the area in Fig. 2, i.e. the value of integral I1 can be calculated as:

π  I1 = Fh (ϕ M )  − ϕo  = Fh (ϕ M )ϕ z , (27) 2 

where φz is the angle of tooth engagement, and Fh (ϕ M ) denotes mean value of function Fh (ϕ ) which defines cutting conditions and has a constant value.

By substituting Eqs. (27) to (26) the final expression, which defines displacement of workpiece during engagement of a single tooth is derived:

f1 =

Rk Fh (ϕ M )ϕ z Ω

. (28)

An analysis of this expression leads to the conclusion that workpiece displacement (the value of feed motion) also depends on the speed

Tadic, B. – Vukelic, D. – Hodolic, J. – Mitrovic, S. – Eric, M.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 425-439

of primary motion. With this method of milling, the feed speed, vp, and feed per tooth, f1, are not constant but vary as the function of the horizontal component of the cutting force. Fig. 3 shows theoretical dependence of Fh(t), vp(t), and L(t) on the time of machining.

can be achieved, while the conventional gearboxes on milling machines allow this change only stepwise, and • feed speed is variable and ranges from the selected value at point C, down to zero at point D (Fig. 1), which belongs to the machined surface.

However, it should be noted that both the conventional milling and the milling method which uses conservative forces to control feed speed, represent special cases of a more general theoretical milling method.

If, besides cutting forces, the workpiece is under the influence of an active force, Fak, which acts in the direction of feed motion, then feed speed can be calculated as:

Fig. 2. Graph of integral I1

v p = Rk Fak + Fh (t ) . (29)

For larger values of Fak, i.e. in cases when Fak >>Fh(t), feed speed is approximately constant, rendering this method equivalent to conventional down milling. When active forces are smaller (Fak<<Fh(t)) this milling method employs feed speed control as described above.

Previous theoretical analysis ignored the compressibility of fluid which fills the hydraulic cylinder. However, for the purpose of practical realization of conservative-force-controlled feed drive, a detailed analysis of dynamic equations was conducted, taking into account fluid (oil) compressibility.

2 DYNAMIC MODEL OF THE FEED DRIVE SYSTEM

Fig. 3. Dependence of horizontal component of cutting force Fh(t), feed speed vp(t), and traversed path L(t) on time t Control of feed motion using horizontal component of the cutting force differs from conventional methods for feed motion control in milling, in the following: • displacement of workpiece takes place only during engagement with tooth (teeth), while in the case of conventional milling the displacement takes place at a constant rate, • by regulating flow by throttling valve continuous control of the mean feed speed

As mentioned above, previous analysis revealed basic relationships between feed speed, displacement, and active force (horizontal component of the cutting force), but failed to give the real picture of dynamic behaviour of the feed speed drive. One of the basic problems related to modelling of dynamic behaviour of feed drives is the real compressibility of the fluid which fills up the cylinder of hydraulic assembly. Fig. 4, which shows dynamic model of the feed drive, can be used to analyze forces and displacement. The system is loaded in the direction of feed motion with following forces: horizontal component of cutting force (Fh), the resulting inertial force (Fin), friction force (FT), resistance of throttling valve

Conservative-Force-Controlled Feed Drive System for Down Milling

429

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 425-439

(Fp), and elastic force which is the result of fluid compressibility (Fe).

incompressible fluid. The resulting force shall influence displacement along ξ and x coordinates.

Force balance equation for this system is:      

F in + Fe + Fp + FT + Fh + Fak = 0 . (30) Inertial force Fin represents the total system inertial force and can be expressed as:

Fig. 4. Diagram of dynamic model of feed drive system Under the influence of the stated forces, during the finite time interval Δt, the piston shall traverse path from point P to point P2. The path PP1 traversed by the piston is proportional to the volume of fluid passing through the throttling valve during Δt interval, while the piston displacement P1 P2 is the result of fluid compressibility. A dynamic model of the feed drive system (Fig. 4) has two degrees of freedom.

Total displacement of workpiece comprises the controlled segment PP1 , which is regulated by the damping rate, and the segment P1 P2 which is not subject to control. The displacement P1 P2 depends on the coefficient of compressibility and the geometry of hydraulic cylinder. For the expected load level during machining, the flow speed of real fluid does not differ significantly from the flow speed of ideal fluid. Bearing this in mind, the dynamic model of feed drive can be represented by Fig. 5.

In cylinder 1 there is a real compressible fluid, while cylinder 2 contains the ideal, 430

d2x d 2ξ , (31) + m dt 2 dt 2

where m is the total movable mass (slider, piston rod, etc.), ξ denotes piston displacement due to compression of the real-compressible fluid, while x is piston displacement allowed by the selected damping coefficient.

From the equation: Ak ⋅ d ξ 1 dV

, (32) s= = V dp A ⋅ L 1 dF k e Ak follows that the elastic force Fe equals:

Fe =

Ak ξ , (33) s⋅L

where the stiffness is defined by the ratio (Ak/ s×L), and Ak is the active piston surface area, s is oil compressibility coefficient, and L is the length of cylinder at the front of the piston at the observed moment in time. Damping force Fp can be determined based on Eq. (19). The magnitude of this force retains its form of dependence, and can be calculated as function of piston speed:

Fig. 5. Dynamic model of feed drive system

F in = m

Fp =

1 Rk

2

 dx   dx  ⋅   ⋅ sgn   ,  dt   dt 

(34)

2

 dx   dx  Fp = Rr ⋅   ⋅ sgn   , (35)  dt   dt 

where Rr is reciprocal value of the regulation constant Rk given in Eq. (20), and dx/dt is the speed of piston due to flow of fluid through the throttling valve.

Damping force Fp is the conservative force used to control the down milling feed. In conventional milling this force does not exist, while feed is controlled in a completely different way.

Friction force FT represents the sum of all friction forces on movable feed drive components.

Bearing in mind the feed drive design (the built-in

Tadic, B. – Vukelic, D. – Hodolic, J. – Mitrovic, S. – Eric, M.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 425-439

linear roller guides) for the most part, the friction force can be neglected. The magnitude of active force Fak can be chosen arbitrarily, while the horizontal component of the cutting force, as a smooth function of time, can be defined as:

) (

(

)

Ft max Φ  − sin Ω t + β  Φ sin Ω t   ⋅ (36) Fh (t ) = sin β ⋅ sin Ω t ( cos Ω t + K sin Ω t )

where Φ is Heaviside function, and Ftmax is the maximum value of the tangential component of the cutting force.

Based on the forces defined, the differential equation of displacement can be set: d 2ξ d2x A m 2 +m 2 + k ξ + sL dt dt (37)

2  dx   dx  + Rr   ⋅ sgn   = Fh (t ) + Fak .  dt   dt  There is a following relationship between the elastic force Fe and damping force Fp: 2

Ak  dx   dx  ξ = Rr   ⋅ sgn   , (38) sL dt    dt 

from which the following is derived: 2 Ak d ξ  dx  d x = 2 Rr   2 , (39) s ⋅ L dt  dt  dt

while the acceleration follows from: 2

d x 1 = dt 2 2

Ak dξ . (40) s ⋅ L ⋅ Rr ⋅ ξ dt

For both senses of motion the same damping rate was adopted, i.e.:

Rr(x<0) = Rr(x > 0), (41)

although a solution can be found for other damping rates. By substituting both Eq. (40) and relationships between elastic and damping forces into Eq. (37), a nonlinear differential equation is derived:

m

d 2ξ m + dt 2 2

Ak dξ + s ⋅ L ⋅ Rr ⋅ ξ dt (42)

2 Ak ⋅ ξ = Fh (t ) + Fak s⋅L

The problem is now reduced to defining the law of motion ξ(t). Analysis of differential Eq. (37) leads to the conclusion that the magnitude of displacement ξ depends, among other factors, on the damping rate Rr. Observing the boundary condition of the system when the throttling valve is completely blocked (closed), it can be concluded that the displacement x equals zero, reducing Eq. (37) to the following form:

m

A d 2ξ + k ξ = Fh (t ) + Fak . (43) 2 s⋅L dt

In this boundary case, the total work of external forces Fak and Fh(t) shall be spent on fluid compression in cylinder 1. The magnitude of this work depends on the magnitude of damping rate Rr, so ξ - which is derived from Eq. (43) - has the largest possible value. Thus, the law by which ξ(t) changes in time can be expressed by:

m

A d 2ξ + k ξ = Z Fh (t ) + Fak  , (44) 2 s⋅L dt

where Z is an unknown constant. Eq. (44) has an analytical solution, which is given here in its final form:

ξ ( t ) = A1 ( t ) cos ω t + B1 ( t ) sin ω t +

(45)

+C1 ( t ) sin 2Ω t + D1 ( t ) sin 2 Ω t + E1 ( t ) ,

where Ω is forced frequency, ω is eigenfrequency.

Eigenfrequency is given as:

Ak , (46) s⋅L⋅m

ω=

while the remaining parameters follow from:

− Ft max ⋅ Z Φ ( sin Ω t ) m ⋅ sin β

, Φ  − sin ( Ω t + β )  h1 ( t ) =

h2 =

C1 ( t ) =

D1 ( t ) =

E1 ( t ) =

(47)

Fak , (48) m 0, 5h1 ( t )

ω 2 − 4Ω 2 h1 ( t ) K

ω 2 − 4Ω 2

, (49) , (50)

h2 − 2Ω 2 D1 ( t )

Conservative-Force-Controlled Feed Drive System for Down Milling

ω2

, (51) 431

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 425-439

B1 ( t ) =

A1 ( t ) =

−2Ω C1 ( t )

ω

, (52)

s⋅L Fak − E1 ( t ) . (53) Ak

The law of change ξ(t), defined by Eq. (45) and a constant Z = 0.5, identically satisfy Eq. (41) for any m, Ak, s, L, Rr and forces Fh(t), and Fak. The speed of fluid compression (v2) in the hydraulic cylinder is calculated as a derivative of displacement ξ(t) in time: dξ = dt = − A1 ( t ) ω sin ω t + B1 ( t ) ω cos ω t +

calculation of current feed speeds and workpiece displacements, for various input values.

Shown in Figs. 6 and 7 are diagrams of boundary feed speeds which are the result of two different damping rates Rr = 105 N1/2s/m (Fig. 6), and Rr = 1010 N1/2s/m (Fig. 7).

v2 ( t ) =

(54)

+2Ω C1 ( t ) cos 2Ω t + D1 ( t ) Ω sin 2Ω t , while the controlled segment of speed (v1) can be calculated as:

v1 =

dx = dt

Ak ξ . (55) s ⋅ L ⋅ Rr

Workpiece feed speed equals the sum of v1 and v2, that is:

v p = v1 + v2 , (56)

while the total displacement of workpiece equals: t

x p ( t ) = ξ ( t ) + x ( t ) = ξ ( t ) + v1 ( t )dt . (57)

∫ 0

Fig. 6. Feed speed resulting from dynamic model with following inputs Ftmax = 3000 N; Fr/Ft = 0.7; s = 3×10-10 m2/N; W = 31.4 s-1; m = 50 kg; L = 0.2 m; Ak = 0.0144 m2, Fak = 0 N; Rr = 105 N1/2s/m, δ = 5 mm; D = 100 mm Based on the equations given above, a software application was made which allows 432

Fig. 7. Feed speed resulting from dynamic model with following inputs Ftmax = 3000 N; Fr/Ft = 0.7; s = 3×10-10 m2/N; W = 31.4 s-1; m = 50 kg; L = 0.2 m; Ak = 0.0144 m2, Fak = 0 N; Rr = 1010 N1/2sm-1, δ = 5 mm; D = 100 mm It should be noted that higher damping rates Rr (Fig. 7) prohibit machining by the proposed technology. Namely, there is an area of damping rates Rr and stiffness (Ak/s×L) which provides sufficient displacement of workpiece.

For particular stiffness, the level of damping rate Rr can be selected such as to allow cutting with feed speeds common to milling, which is shown in Figs. 8 to 10.

Fig. 8. Feed speed resulting from dynamic model with following inputs Ftmax = 3000 N; Fr/Ft = 0.7; s = 3×10-10 m2/N; W = 31.4 s-1; m = 50 kg; L = 0.2 m; Ak = 0.0144 m2, Fak = 0 N; Rr = 5×107 N1/2sm-1, δ = 5 mm; D = 100 mm

Tadic, B. – Vukelic, D. – Hodolic, J. – Mitrovic, S. – Eric, M.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 425-439

Fig. 9. Feed speed resulting from dynamic model with following inputs Ftmax = 3000 N; Fr/Ft = 0.7; s = 3×10-10 m2/N; W = 31.4 s-1; m = 50 kg; L = 0.2 m; Ak = 0.0144 m2, Fak = 0 N; Rr = 5×108 N1/2sm-1, δ = 5 mm; D = 100 mm

Fig. 10. Feed speed resulting from dynamic model with following inputs Ftmax = 1000 N; Fr/Ft = 0.7; s = 3×10-10 m2/N; W = 31.4 s-1; m = 50 kg; L = 0.2 m; Ak = 0.0144 m2, Fak = 2000 N; Rr = 109 N1/2sm-1, δ = 5 mm; D = 100 mm

Fig. 11. Feed speed resulting from dynamic model with following inputs Ftmax = 3000 N; Fr/Ft = 0.7; s = 3×10-10 m2/N; W = 31.4 s-1; m = 50 kg; L = 0.2 m; Ak = 0.0144 m2, Fak = 2000 N; Rr = 109 N1/2sm-1, δ = 5 mm; D = 100 mm

If, in addition to the horizontal component of the cutting force, active force Fa is also acting upon piston rod, then the proposed feed drive dynamically behaves similarly to the conventional one. Fig. 11 shows the law of feed speed change in case of an active force of higher magnitude.

Total displacement of the workpiece, defined by Eq. (57), also comprises displacement component ξ which is not subject to control. The magnitude and law of change of displacement ξ(t) depend on the following parameters: the stiffness of hydraulic assembly (Ak/s×L), magnitude and ratio between the radial and tangential cutting force (Fr, Ft, K), mass of movable feed drive components (m), cutter diameter (δ), cutting depth (D), and angular speed of the cutter (Ω). For particular values of these parameters and within some period of cutting, the displacement ξ reaches its maximum. It is of utmost importance to choose such construction parameters Ak, s, L, and m, which would for the expected magnitudes of cutting force components allow the uncontrollable ξ to be kept at insignificant level. This can be realized by increasing the stiffness of the hydraulic assembly. Values of stiffness (Ak/s×L) and mass (m) should be chosen in such a mannerthat given real machining conditions system oscillations are kept out of the resonance area.

The magnitude of resonant frequency is defined by:

Ω=

1 2

Ak . (58) s⋅L⋅m

An analysis of the Eq. (58) shows that the resonance area is above the real frequency of impulse forces inherent to milling. This frequency rarely oversteps 400 Hz. Thus, the magnitude of hydraulic assembly stiffness and the mass of movable drive components, can be easily chosen so that the magnitude of eigenfrequency is 104 Hz and higher. Figs. 12 and 13 show a change of displacement ξ in time, for various angular speeds of the cutter. Simulation of displacement ξ, was performed with following parameter values: max. circumferential force Ftmax = 3000 N, ratio between radial and tangential cutting force K = 0.6, cutter diameter δ = 100 mm, cutting depth D = 20 mm, coefficient of fluid (oil) compressibility s = 3×10-10 m2/N, mass of movable components m = 50 kg, observed cylinder length

Conservative-Force-Controlled Feed Drive System for Down Milling

433

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 425-439

L = 0.2 m, and active piston area Ak = 0.014 m2.

For the given machining conditions, the maximum value of uncontrollable displacement ξ(t) does not overstep 3 mm (Figs. 12 and 13), which makes it practically possible to establish control from zero to the desired value by shifting x.

Fig. 12. Dependence of displacement ξ on cutting time interval, at angular speed Ω = 10 s-1

Fig. 13. Dependence of displacement x on cutting time interval, at angular speed Ω = 50 s-1 3 DESIGN OF FEED DRIVE SYSTEM The results of the theoretical analysis from previous sections were used to design a physical prototype of the feed drive system for down milling operations. This feed drive system establishes control over the feed motion based on the horizontal component of the cutting force.

Feed drive components are designed in such a way that they allow continuous shift of feed speed within the vp = 0 to 500 mm/min interval, at the maximum value of the horizontal component of the cutting force Fhmax < 10000 N. 434

For the sake of brevity, only a segment of design solution with some specific details is presented here. Based on the diagram in Fig. 14 characteristic components of the feed drive system are labeled and the functioning of particular subassemblies is given. The workpiece (1) is located and clamped in the fixture (2), which is screwed onto the movable slider (3) of the feed drive worktable. The worktable (3) consists of a movable slider, standard linear and roller guides, and table body with preloading screws. The linear roller guides are protected against chip by special metal sheet elements (4), while the entire assembly is tied to the base plate (5), which is screwed down to the worktable. The movable slide of the worktable is connected to the piston rod (6) through a special carrier. This connection works through a lever (7), piston rod extension (8), axle (9) and screws. Lever (7), piston rod extension (8) and axle (9) are connected without any clearance, which, during assembly, allows total alignment of kinematic axes of piston rod and worktable slider.

In addition, this connection does not allow torque transfer, which means that the piston rod is loaded with axial force only. Piston rod (6) is common to brake cylinder (10) and pneumatic cylinder (11), which are screwed down to base plate via holders (12) to (14). Left and right chamber of hydraulic cylinder are connected via a throttling valve (10a), which regulates feed speed. The same line also includes a unidirectional blocking valve (10b) in order to increase velocity of backstroke.

Motion is initiated via a pneumatic cylinder (11) which is fed by compressed air - distributed via air service unit (11a) and a sliding valve (11b) into the cylinder’s left chamber. At backstroke, the air is fed into the right chamber of the pneumatic cylinder, so the fluid (oil) from the left chamber of the hydraulic cylinder flows through the throttle valve and through unidirectional blocking valve, which allows a much faster execution of backstroke. Attached to the holder (12) of the hydraulic cylinder is support (15). The support has a groove which allows inductive sensors - (16) (start) and (17) (stop) - to be mounted at distance L and activated by platelet (18) which is mounted on support (7). These sensors are connected to a timer (19). Based on the known path L and elapsed time t, which is read on timer display, it is possible to calculate the mean feed speed - for the

Tadic, B. – Vukelic, D. – Hodolic, J. – Mitrovic, S. – Eric, M.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 425-439

chosen damping rate and machining parameter.

Hydraulic cylinder is equipped with a throttling valve (10c) which controls oil feed to the cylinder.

During feeding, the throttling valve (10a) is open due to alternate movements of piston rod and air let-off. In order to eliminate negative effects of trapped air in the hydraulic cylinder, oil is fed under pressure, which is controlled by the pressure gage (10d). The feed drive system described is a module independent from the original machine tool feed drive (milling feed drive system is shut off during entire machining process). The milling machine worktable, in this case, assumes the role of a stable platform to which a module is attached which assumes the function of the drive system.

The feed drive system is shown in Fig. 15.

Fig. 14. Schema of the proposed feed drive system

4 TESTING OF FEED DRIVE SYSTEM Prior to experimental investigation, the feed drive system was tested. In addition to testing the geometric accuracy of the moving components of the feed drive system, and checking the stiffness of joints, the relationship between feed speed and axial force was also determined. The axial force exerted by compressed air in the pneumatic cylinder simulates the horizontal component of the cutting force, while the table speed is calculated as the ratio between L and time of cutting t (Fig. 14). By varying the input air pressure one a range of values for axial force (Fai) and time (ti) for which the table travels along path (L) is derived.

The active force (Fa) is calculated based on the pressure which reads on the manometer of the air service unit and the diameter of the pneumatic cylinder. The described method was used with various damping rates, producing the experimental data shown in Table 1.

Statistical analysis of experimental data by regression analysis with the basic function of following form: (Fak – C) = Rr × v2 (59) results with the regression equations of high degree of correlation (Table 2). In Eq. (59) (Fak – C) represents effective force which performs the movement. Namely, the constant C includes friction forces and other opposing forces which are always present, while the member Rr × v2 defines quadratic dependency of feed speed on effective force. This relationship (Rr × v2) was derived from a set of Bernoulli equations within the theoretical investigation. Regression equations and correlation coefficients confirm this relationship (Table 2). Given in Table 2 are feed speeds and axial forces, derived by regression equations, as well as the percent deviations between the measured and calculated values.

The differences in constant C could be attributed to the influence of feed speed on the magnitude of friction force. The decrease of friction force is due to higher feed speeds, which is confirmed by statistical analysis.

5 PRELIMINARY EXPERIMENTS

Fig. 15. Feed drive system mounted on the milling machine worktable

In order to test the theoretical model proposed, as well as the physical prototype of

Conservative-Force-Controlled Feed Drive System for Down Milling

435

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 425-439

Table 1. The results of measurement for pressure p, time t, the calculated force Fa and feed speed vp p [bar] 1.5 2 2.5 3 3.5 4 4.4 -

Damping rate Rr1 Fak t [N] [s] 1062 140 1430 86 1788 65 2145 52 2502 44 2860 39 3146 35 -

vp [mm/min] 64.3 104.7 138.5 173 204.6 230.7 257.3 -

p [bar] 1 1.5 2 2.5 2.9 3.5 4 4.4

Damping rate Rr2 Fak t [N] [s] 715 50 1052 32 1430 24 1788 19 2073 17 2502 15 2860 13 3146 12

vp [mm/min] 180 281.3 375 473.8 529.5 600 692.5 750

Table 2. Results of statistical analysis Damping rate Rr1 vp Fak1 DF [mm/min] [N] [%] 64.3 1204 14.2 104.7 1430 8.5 138.5 1702 8.6 173 2058 8.7 204.6 2452 4.9 230.7 2828 3.1 257.3 3258 -11 Regression equation Fak-106.7=3.31×10-3×vp2 r = 0.993, s = 10.1 the feed drive, preliminary experiments were conducted. Extensive follow-up experimental investigation is planned in the future to determine the quality of a number of machining parameters derived by the novel method, as well as to compare them with those obtained by conventional methods.

Preliminary experimental investigation was conducted under following conditions: • The machining was performed using feed drive of a universal milling machine PGU–3.

• Workpiece material: C1530, hardness 190 to 205 HB, with 0.44% C; 0.18% Si; 0.27% Mn; 0.011% S; <0.010% P, • Cutter: standard end mill 80X10X27N, • Cutting regime parameters were as follows: cutting speed 22.6 m/min, feed speed 112.8 mm/min, cutting depth 1 mm, cutting width 10 mm.

The following parameters were monitored during the experiment: 436

Damping rate Rr2 vp Fak2 DF [mm/min] [N] [%] 180 872 -15.7 281.3 1083 -2.1 375 1361 6.8 473.8 1741 4.6 529.5 1994 7.8 600 2355 14.7 692.5 2896 -3.6 750 3272 -12.8 Regression equation Fak-72.5=4.53×10-4×vp2 r = 0.992, s = 11.3

Fig. 16. Images of worn cutter teeth after 347.6 minutes of cutting

Tadic, B. – Vukelic, D. – Hodolic, J. – Mitrovic, S. – Eric, M.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 425-439

•

curves is given in Fig. 17 for the characteristic cutter teeth, The profiles of the machined surface were taken after two cutting intervals, using Taylor Hobson – Talysurf 6 apparatus. The profiles are shown in Figs. 18 and 19.

6 CONCLUSION

Fig. 17. Wear curve for characteristic cutter teeth in dependence of cutting time

Fig. 18. Profile of machined surface after 31.8 minutes of cutting

Fig. 19. Profile of machined surface after 347.6 minutes of cutting •

Mean flank wear on the four characteristic cutter teeth. As an example, in Fig. 16 images of worn teeth generated on Meiji Techno MC50 microscope are shown. Based on the measured flank wear, a diagram of wear

The proposed theoretical model implicates the practical possibility of controlling feed motion in down milling by hydraulic damping, where the horizontal component of the cutting force is used to actuate motion. The theoretical model proposed should be taken as a fundamental model, which clearly implies the possibility of practical machining as described in this paper. Based on the proposed model, which can be conditionally treated as a static model, a dynamic testing of the feed drive system was also performed. The results of dynamic modelling show that the horizontal component of the cutting force can be successfully used as the active force in down milling feed motion. The results of dynamic modelling also show that the hydraulic damping allows a very precise control of the down milling feed motion.

Testing of the physical prototype of the feed drive system also allowed preliminary experimental investigation to be performed on a real machining system.

The results of preliminary experimental investigation allow the following conclusions: • Down milling is possible according to the proposed method, which completely verifies the proposed theoretical model.

• Tool life and surface quality obtained by the proposed method are of the same order of magnitude as those obtained by conventional machining methods.

Based on the investigation results it is the authors’ opinion that the feed drive systems of this type can have broad practical applications. Their advantage in comparison with conventional feed drive systems is their ability to shift feed speed continuously as well as the lower system cost. The results achieved so far provide a solid ground for continuation of both theoretical and experimental investigations.

Conservative-Force-Controlled Feed Drive System for Down Milling

437

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 425-439

7 REFERENCES [1] Stephenson, D.A., Agapiou, J.S. (2006).

Metal Cutting Theory and Practice, 2nd ed.

CRC Press, Boca Raton.

[2] Boothroyd, G., Knight, W.A. (2006).

Fundamentals of machining and machine tools, 3rd ed. CRC Press, Boca Raton.

[3] Sai, K., Bouzid, W. (2005). Roughness modeling in up-face milling. The International Journal of Advanced Manufacturing Technology, vol. 26, no. 4, p. 324-329.

[4] Davim, J.P., Clemente, V.C., Silva, S. (2009). Surface roughness aspects in milling MDF (medium density fibreboard).

The International Journal of Advanced Manufacturing Technology, vol. 40, no. 1-2, p. 49-55.

[5] Mansour, A., Abdalla, H. (2002). Surface roughness model for end milling: a semifree cutting carbon casehardening steel (EN 32) in dry condition. Journal of Materials Processing Technology, vol. 124, no. 1-2, p.

183-191.

[6] Miao, J.C., Chen, G.L., Lai, X.M., Li, H.T., Li, C.F. (2007). Review of dynamic issues in micro-end-milling. The International Journal of Advanced Manufacturing Technology, vol.

31, no. 9-10, p. 897-904.

[7] Liu, X., Cheng, K. (2005). Modelling the machining dynamics of peripheral milling.

International Journal of Machine Tools & Manufacture, vol. 45, no. 11, p. 1301-1320.

[8] Altintas, Y., Montgomery, D., Budak, E. (1992). Dynamic peripheral milling of flexible structure. Journal of Engineering for Industry, vol. 114, no. 2, p. 137-145.

[9] Zeng, W., Jiang, X., Blunt, L. (2009). Surface characterisation-based tool wear monitoring in peripheral milling. The International Journal of Advanced Manufacturing Technology, vol. 40, no. 3-4, p. 226-233.

[10] Xu, C., Chen, H., Liu, Z., Cheng, Z. (2009).

Condition monitoring of milling tool wear based on fractal dimension of vibration signals. Strojniški vestnik ‒ Journal of Mechanical Engineering, vol. 55, no. 1, p.

15-25.

[11] Mann, B.P., Insperger, T., Bayly, P.V., Stepan, G. (2003). Stability of up-milling and down438

milling, part 2: experimental verification.

International Journal of Machine Tools & Manufacture, vol. 43, no. 1, p. 35-40.

[12] Zhao, M.X., Balachandran, B. (2001).

Dynamics and stability of milling process.

International Journal of Solids and Structures, vol. 38, no. 10-13, p. 2233-2248.

[13] Altintas, Y., Budak, E. (1995). Analytical prediction of stability lobes in milling. CIRP Annals - Manufacturing Technology, vol. 44, no. 1, p. 357-362.

[14] Wan, M., Zhang, W.H. (2006). Calculations of chip thickness and cutting forces in flexible end milling. The International Journal of Advanced Manufacturing Technology, vol.

29, no. 7-8, p. 637-647.

[15] Kumar Reddy, N.S., Rao, P.V. (2005).

Selection of optimum tool geometry and cutting conditions using a surface roughness prediction model for end milling.

The International Journal of Advanced Manufacturing Technology, vol. 26, no. 1112, p. 1202-1210.

[16] Chu, C.H, Chen, J.T. (2006). Tool path planning for five-axis flank milling with developable surface approximation.

The International Journal of Advanced Manufacturing Technology, vol. 29, no. 7-8, p. 707-713.

[17] Park, S.C., Choi, B.K. (2000). Tool-path planning for direction-parallel area milling.

Computer-Aided Design, vol. 32, no. 1, p. 1725.

[18] Balič, J., Korošec, M. (2002). Intelligent tool path generation for milling of free surfaces using neural networks. International Journal of Machine Tools & Manufacture, vol. 42, no.

10, p. 1171-1179.

[19] Sonmez, A.I., Baykasoglu, A., Dereli, T., Filiz, I.H. (1999). Dynamic optimization of multi-pass milling operations via geometric programming. International Journal of Machine Tools & Manufacture, vol. 39, no. 2, p. 297-320.

[20] Čuš, F., Župerl, U., Kiker, E. (2007). A modelbased system for the dynamic adjustment of cutting parameters during a milling process.

Strojniški vestnik ‒ Journal of Mechanical Engineering, vol. 53, no. 9, p. 524-540.

Tadic, B. – Vukelic, D. – Hodolic, J. – Mitrovic, S. – Eric, M.

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 425-439

[21] Dereli, T., Filiz, I.H., Baykasoglu, A. (2001). Optimizing cutting parameters in process planning of prismatic parts by using genetic algorithms. International Journal of Production Research, vol. 39, no. 15, p.

3303-3328.

[22] Palanisamy, P., Rajendran, I., Shanmugasundaram, S. (2007). Optimization of machining parameters using genetic algorithm and experimental validation for end-milling operations. The International Journal of Advanced Manufacturing Technology, vol. 32, no. 7-8, p. 644-655.

[23] Čuš, F., Milfelner, M., Balič, J. (2006).

An intelligent system for monitoring and optimization of ball-end milling process.

Journal of Materials Processing Technology, vol. 175, no. 1-3, p. 90-97.

[24] Župerl, U., Čuš, F., Muršec, B., Ploj, T. (2006). A generalized neural network model of ball-end milling force system. Journal of Materials Processing Technology, vol. 175, no. 1-3, p. 98-108.

[25] Palanisamy, P., Rajendran, I., Shanmugasundaram, S. (2008). Prediction of tool wear using regression and ANN models in end-milling operation. The International Journal of Advanced Manufacturing Technology, vol. 37, no. 1-2, p. 29-41.

[26] Tansel, I.N., Ozcelik, B., Bao, W.Y., Chen, P., Rincon, D., Yang, S.Y., Yenilmez, A. (2006).

Selection of optimal cutting conditions by using GONNS. International Journal of Machine Tools & Manufacture, vol. 46, no. 1, p. 26-35.

[27] Ozcelik, B., Oktem, H., Kurtaran, H. (2005).

Optimum surface roughness in end milling Inconel 718 by coupling neural network model and genetic algorithm. International Journal Advanced Manufacturing Technology, vol. 27, no. 3-4, p. 234-241.

[28] Župerl, U., Čuš, F., Gečevska, V. (2007).

Optimization of the characteristic parameters in milling using the PSO evolution technique.

Strojniški vestnik ‒ Journal of Mechanical Engineering, vol. 53, no. 6, p. 354-368.

Conservative-Force-Controlled Feed Drive System for Down Milling

439

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 440-441 Instructions for Authors

Instructions for Authors All manuscripts must be in English. Pages should be numbered sequentially. The maximum length of contributions is 10 pages. Longer contributions will only be accepted if authors provide justification in a cover letter. Short manuscripts should be less than 4 pages. For full instructions see the Authors Guideline section on the journal’s website: http://en.sv-jme.eu/.

Send to: Strojniški vestnik Journal of Mechanical Engineering Aškerčeva 6, 1000 Ljubljana, Slovenia Phone: 00386 1 4771 137 Fax: 00386 1 2518 567 E-mail: info@sv-jme.eu strojniski.vestnik@fs.uni-lj.si Please add a cover letter stating the following information about the submitted paper: 1. Paper title, list of authors and affiliations.

2. The type of your paper: original scientific paper (1.01), review scientific paper (1.02) or short scientific paper (1.03).

3. A declaration that your paper is unpublished work, not considered elsewhere for publication.

4. State the value of the paper or its practical, theoretical and scientific implications. What is new in the paper with respect to the state-of-the-art in the published papers?

5. We kindly ask you to suggest at least two reviewers for your paper and give us their names and contact information (email).

Every manuscript submitted to the SV-JME undergoes the course of the peer-review process.

THE FORMAT OF THE MANUSCRIPT The manuscript should be written in the following format: - A Title, which adequately describes the content of the manuscript.

- An Abstract should not exceed 250 words. The Abstract should state the principal objectives and the scope of the investigation, as well as the methodology employed. It should summarize the results and state the principal conclusions.

- 6 significant key words should follow the abstract to aid indexing.

- An Introduction, which should provide a review of recent literature and sufficient background information to allow the results of the article to be understood and evaluated.

440

- -

A Theory or experimental methods used.

An Experimental section, which should provide details of the experimental set-up and the methods used for obtaining the results.

- A Results section, which should clearly and concisely present the data using figures and tables where appropriate.

- A Discussion section, which should describe the relationships and generalizations shown by the results and discuss the significance of the results making comparisons with previously published work. (It may be appropriate to combine the Results and Discussion sections into a single section to improve the clarity).

- Conclusions, which should present one or more conclusions that have been drawn from the results and subsequent discussion and do not duplicate the Abstract.

- References, which must be cited consecutively in the text using square brackets [1] and collected together in a reference list at the end of the manuscript.

Units - standard SI symbols and abbreviations should be used. Symbols for physical quantities in the text should be written in italics (e.g. v, T, n, etc.).

Symbols for units that consist of letters should be in plain text (e.g. ms-1, K, min, mm, etc.)

Abbreviations should be spelt out in full on first appearance, e.g., variable time geometry (VTG).

Meaning of symbols and units belonging to symbols should be explained in each case or quoted in a special table at the end of the manuscript before References.

Figures must be cited in a consecutive numerical order in the text and referred to in both the text and the caption as Fig. 1, Fig. 2, etc. Figures should be prepared without borders and on white grounding and should be sent separately in their original formats.

Pictures may be saved in resolution good enough for printing in any common format, e.g. BMP, GIF or JPG. However, graphs and line drawings should be prepared as vector images, e.g. CDR, AI. When labeling axes, physical quantities, e.g. t, v, m, etc. should be used whenever possible to minimize the need to label the axes in two languages. Multicurve graphs should have individual curves marked with a symbol. The meaning of the symbol should be explained in the figure caption.

Tables should carry separate titles and must be numbered in consecutive numerical order in the text and referred to in both the text and the caption as Table 1, Table 2, etc. In addition to the physical quantity, e.g. t (in italics), units (normal text), should be added in

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, 440-441

square brackets. The tables should each have a heading.

Tables should not duplicate data found elsewhere in the manuscript.

Acknowledgement of collaboration or preparation assistance may be included before References. Please note the source of funding for the research.

REFERENCES A reference list must be included using the following information as a guide. Only cited text references are included. Each reference is referred to in the text by a number enclosed in a square bracket (i.e., [3] or [2] to [6] for more references). No reference to the author is necessary.

References must be numbered and ordered according to where they are first mentioned in the paper, not alphabetically. All references must be complete and accurate. All non-English or. non-German titles must be translated into English with the added note (in language) at the end of reference. Examples follow.

Journal Papers: Surname 1, Initials, Surname 2, Initials (year). Title.

Journal, volume, number, pages.

[1] Zadnik, Ž., Karakašič, M., Kljajin, M., Duhovnik, J. (2009). Function and functionality in the conceptual design process. Strojniški vestnik Journal of Mechanical Engineering, vol. 55, no.

7-8, p. 455-471.

Journal titles should not be abbreviated. Note that journal title is set in italics.

Books: Surname 1, Initials, Surname 2, Initials (year). Title.

Publisher, place of publication.

[2] Groover, M.P. (2007). Fundamentals of Modern Manufacturing. John Wiley & Sons, Hoboken.

Note that the title of the book is italicized.

Chapters in Books: Surname 1, Initials, Surname 2, Initials (year). Chapter title. Editor(s) of book, book title. Publisher, place of publication, pages.

[3] Carbone, G., Ceccarelli, M. (2005). Legged robotic systems. Kordić, V., Lazinica, A., Merdan, M. (Eds.), Cutting Edge Robotics. Pro literatur Verlag, Mammendorf, p. 553-576.

Proceedings Papers: Surname 1, Initials, Surname 2, Initials (year). Paper title. Proceedings title, pages.

[4] Štefanić, N., Martinčević-Mikić, S., Tošanović, N. (2009). Applied Lean System in Process Industry.

MOTSP 2009 Conference Proceedings, p. 422427.

Standards: Standard-Code (year). Title. Organisation. Place.

[5] ISO/DIS 16000-6.2:2002. Indoor Air – Part 6: Determination of Volatile Organic Compounds in Indoor and Chamber Air by Active Sampling on TENAX TA Sorbent, Thermal Desorption and Gas Chromatography using MSD/FID. International Organization for Standardization. Geneva. www pages: Surname, Initials or Company name. Title, from http:// address, date of access.

[6] Rockwell Automation. Arena, from http://www. arenasimulation.com, accessed on 2009-09-07.

EXTENDED ABSTRACT By the time the paper is accepted for publishing, the authors are requested to send the extended abstract (approx. one A4 page or 3.000 to 3.500 characters).

The instructions for writing the extended abstract are published on the web page http://www.sv-jme.eu/ information-for-authors/.

COPYRIGHT Authors submitting a manuscript do so on the understanding that the work has not been published before, is not being considered for publication elsewhere and has been read and approved by all authors. The submission of the manuscript by the authors means that the authors automatically agree to transfer copyright to SV-JME and when the manuscript is accepted for publication. All accepted manuscripts must be accompanied by a Copyright Transfer Agreement, which should be sent to the editor. The work should be original by the authors and not be published elsewhere in any language without the written consent of the publisher.

The proof will be sent to the author showing the final layout of the article. Proof correction must be minimal and fast. Thus it is essential that manuscripts are accurate when submitted.

Authors can track the status of their accepted articles on http://en.sv-jme.eu/.

PUBLICATION FEE For all articles authors will be asked to pay a publication fee prior to the article appearing in the journal. However, this fee only needs to be paid after the article has been accepted for publishing. The fee is 180.00 EUR (for articles with maximum of 6 pages), 220.00 EUR (for articles with maximum of 10 pages), 20.00 EUR for each addition page. Additional costs for a color page is 90.00 EUR.

441

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5 Vsebina

Vsebina Strojniški vestnik - Journal of Mechanical Engineering letnik 57, (2011), številka 5 Ljubljana, maj 2011 ISSN 0039-2480 Izhaja mesečno

Povzetki člankov Domen Šeruga, Matija Fajdiga, Marko Nagode: Izračun poškodbe zaradi lezenja pri termomehanskem obremenjevanju Uroš Zupanc, Janez Grum: Integriteta površine po mehanskem utrjevanju aluminijeve zlitine 7075-T651 Uroš Trdan, José Luis Ocaña, Janez Grum: Modifikacija površinskega sloja aluminijevih zlitin po utrjevanju z laserskimi udarnimi valovi Edvard Detiček, Uroš Župerl: Inteligentni elektrohidravlični servopozicionirni sistem Ugljesa Bugaric, Dusan Petrovic, Zoran Petrovic, Miroslav Pajcin, Gordana Markovic-Petrovic: Določanje zmogljivosti terminala za razkladanje razsutega tovora z uporabo teorije vrst Fatih Taylan, Oğuz Çolak, Mehmet Cengiz Kayacan: Raziskava zmogljivosti rezalnih orodij iz materiala CBN s prevleko TiN in brez nje pri aplikacijah trdega rezkanja Branko Tadic, Djordje Vukelic, Janko Hodolic, Slobodan Mitrovic, Milan Eric: Sistem za pogon podajanja pri istosmernem rezkanju, krmiljen s konzervativno silo

SI 73 SI 74 SI 75 SI 76 SI 77 SI 78 SI 79

Navodila avtorjem

SI 80

Osebne vesti Doktorati, magisteriji, specializacije in diplome

SI 82

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, SI 73

Prejeto: 06.05.2010 Sprejeto: 07.03.2011

Izračun poškodbe zaradi lezenja pri termomehanskem obremenjevanju Šeruga, D. – Fajdiga, M. – Nagode, M. Domen Šeruga* – Matija Fajdiga – Marko Nagode Univerza v Ljubljani, Fakulteta za strojništvo, Slovenija

Namen prispevka: Konstrukcijsko-razvojni čas novega izdelka se zaradi uporabe naprednih numeričnih in eksperimentalnih metod vse bolj krči. Za termomehansko obremenjen izdelek je zanesljiva napoved termomehanske poškodbe eden izmed ključnih korakov v konstrukcijsko-razvojnem postopku.

Poškodbeni mehanizmi, ki se pojavijo pri termomehanskem obremenjevanju, so predvsem mehansko utrujanje, lezenje in oksidacija. Ker lezenju materiala pri termomehanskem obremenjevanju pripade manjši delež poškodbe kot utrujanju, se napovedovanje poškodbe zaradi lezenja ponavadi kar izpušča.

Razlogi so v manj dodelanih in zanesljivih postopkih določevanja poškodbe zaradi lezenja in velikem številu testov za določitev dobe trajanja do porušitve standardnih preizkušancev, kar pomeni visoke razvojne stroške, predvsem pa dolge čase testov.

Metodologija: Doba trajanja do porušitve zaradi lezenja standardnih preizkušancev pri nižjih napetostih in nižjih temperaturah se določa s časovno-temperaturnimi parametri, ki z enačbo in koeficienti povezujejo napetost, temperaturo in dobo trajanja do porušitve zaradi lezenja. Pregledali smo obstoječe časovno-temperaturne parametre. Ugotovili smo, kateri so najbolj uporabljani ter katere so njihove prednosti in pomanjkljivosti. Iz poznanih rezultatov testov smo izločili del podatkov. Predlagali smo hiter in robusten algoritem za določitev koeficientov obravnavanih časovno-temperaturnih parametrov iz izločenega dela podatkov. Napovedano dobo trajanja z različnimi časovno-temperaturnimi parametri smo primerjali in ovrednotili na preostalih podatkih. Razvili smo metodo za izračun poškodbe zaradi lezenja.

Za preprosto napetostno in temperaturno zgodovino obremenjevanja smo izračunali termomehansko poškodbo.

Ugotovitve: Ugotovili smo, da je uporaba Larson-Millerjevega parametra še vedno med najbolj uporabljanimi metodami za določevanje poškodbe izdelkov zaradi lezenja, čeprav ne napoveduje vedno optimalne dobe trajanja. Razvita metoda omogoča ločen in hiter izračun poškodbe zaradi lezenja. Na primeru izračuna termomehanske poškodbe se dobro vidi, kakšna zgodovina obremenjevanja vpliva na poškodbo zaradi lezenja in kakšna na poškodbo zaradi utrujanja.

Omejitve/uporabnost raziskave: Larson-Millerjev, Manson-Haferdov in Orr-Sherby-Dornov parameter imajo zaradi svoje preprostosti omejeno uporabnost. Niso uporabni za vse materiale. MansonBrownov parameter je po drugi strani bolj splošno uporaben, ima pa večje število koeficientov in algoritem za njihovo določitev je numerično izredno nestabilen. Namesto globalno optimalnih koeficientov so rešitev lahko zgolj lokalno optimalni koeficienti, odvisno od začetnih vrednosti v algoritmu. Posledica je nepravilen sklep o dobi trajanja do porušitve zaradi lezenja, kar je dobro vidno v pričujočem prispevku.

Nadaljnje raziskovanje bo v smeri posplošenega časovno-temperaturnega parametra, s katerim bo mogoče zanesljivo napovedati dobo trajanja do porušitve zaradi lezenja, brez omejene uporabnosti.

Izvirnost/pomembnost prispevka: S predlagano metodo je mogoče ločeno izračunati poškodbo zaradi lezenja pri termomehanskem obremenjevanju. Za določitev dobe trajanja do porušitve zaradi lezenja na celotnem obratovalnem napetostno-temperaturnem območju izdelka zadostuje le nekaj standardnih testov lezenja. Algoritem za določitev koeficientov je neodvisen od izbranega časovno-temperaturnega parametra. Predstavljena metoda je vgrajena tudi v komercialni programski paket LMS Virtual.Lab. ©2011 Strojniški vestnik. Vse pravice pridržane.

Ključne besede: lezenje, poškodbeni model, parametrične metode, akumulacija poškodbe, LarsonMiller, Manson-Brown, Manson-Haferd, Orr-Sherby-Dorn, porušitvene krivulje

*Naslov avtorja za dopisovanje: Univerza v Ljubljani, Fakulteta za strojništvo, Aškerčeva 6, 1000 Ljubljana, Slovenija, domen.seruga@fs.uni-lj.si

SI 73

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, SI 74

Prejeto: 02.06.2010 Sprejeto: 29.10.2010

Integriteta površine po mehanskem utrjevanju aluminijeve zlitine 7075-T651 Uroš Zupanc1 ‒ Janez Grum2,* 1 Slovenski institut za varilstvo, Slovenija 2 Univerza v Ljubljani, Fakulteta za strojništvo, Slovenija V prispevku so predstavljene raziskave učinkov hladnega mehanskega utrjevanja površine visokotrdnostne aluminijeve zlitine AlZn5,5MgCu (ENAW 7075). Obravnavane so korelacije med pogoji hladnega utrjevanja površine, integriteto površine materiala po utrjevanju ter uporabniškimi lastnostmi dinamično obremenjenih vzorcev iz izbrane aluminijeve zlitine. Namen prispevka je izbira tehnološko in ekonomično ustreznih parametrov utrjevanja površine, da bo dosežen ustrezen učinek na obratovalne lastnosti oz. povečanje odpornosti materiala na utrujanje pri dinamičnih obremenitvah.

Mikroplastično utrjevanje površinskega sloja poteka neenakomerno oz. stohastično zaradi neenakomernih udarnih pulzov kroglic na površino. Hladno utrjevanje materiala z mikroplastično deformacijo površinskega sloja kaže tipično in prepoznavno površinsko topografijo. Prispevek analizira topografijo površine, meritve hrapavosti, mikrostrukturne spremembe utrjenega površinskega sloja, meritve trdote, poteke zaostalih napetosti in teste materiala na utrujanje. Podatki o hrapavosti površine predstavljajo izhodišče za napovedovanje odpornosti materiala na utrujanje, saj lahko prevelika ali neenakomerna hrapavost utrjene površine zaradi lokalnih koncentracij napetosti na površini oz. zareznih učinkov pospeši nastanek utrujenostne razpoke. Meritve trdote omogočajo vrednotenje učinkov utrditve materiala pri različnih parametrih utrjevanja z mikroplastično deformacijo. Zaostale napetosti v materialu pomembno vplivajo na obratovalne lastnosti materiala pri dinamičnih obremenitvah oz. njegovo odpornost na utrujanje. Dinamično utrujanje je bilo izvedeno v upogibnem načinu obremenjevanja vzorcev. Wöhlerjeve krivulje odpornosti materiala na utrujanje so bile izdelane na področjih maksimalne upogibne obremenitve med 30 in 65% natezne trdnosti materiala (Rm).

Mikroplastično utrjevanje materiala pri trkih kroglic ob površino poveča trdoto tankega površinskega sloja materiala in v materialu inducira zaostale tlačne napetosti. Maksimalne izmerjene vrednosti mikrotrdote površinskega sloja znašajo do 220 HV0.3 in so za okoli 20 do 25% višje v primerjavi z materialom v dobavljenem stanju. Pri vseh izbranih intenzitetah utrjevanja so v površinskem sloju nastale zaostale tlačne napetosti. S povečanjem intenzitete obdelavo smo uspeli v materialu inducirati višje vrednosti zaostalih tlačnih napetosti, in tudi globino utrjenega sloja. Vrednosti zaostalih napetosti so v območju okoli 55% meje tečenja dobavljenega materiala (Rp02 = 532 MPa). Maksimalne globine utrjenega površinskega sloja znašajo do okoli 400 μm. Eksperimentalni rezultati kažejo povečanje časovne trdnosti utrjenih vzorcev z mikroplastično deformacijo površinskega sloja za faktor med 1,14 in 6,28 v primerjavi z materialom v dobavljenem stanju. Vrednosti maksimalnih upogibnih obremenitev trajne dinamične trdnosti utrjenih vzorcev pri številu ciklov 107 (σmax = 218 MPa) so za vsaj 15% višje v primerjavi z materialom v dobavljenem stanju (σmax = 189 MPa). Izboljšanje časovne trdnosti je odvisno od vrednosti maksimalnih upogibnih napetosti ter pogojev utrjevanja vzorcev. Pri nižjih upogibnih obremenitvah utrujanja vzorcev je vpliv zaostalih tlačnih napetosti na napetostni tenzor materiala vse večji.

Utrjevanje površinskega sloja je torej učinkovita metoda za izboljšanje obratovalne dobe pri strojnih delih iz izbrane aluminijeve zlitine, ki imajo zaradi oblike velike koncentracije notranjih napetosti. Enakomerno utrjen površinski sloj zavira nastanek in širjenje utrujenostne razpoke dinamično obremenjenega materiala. Zaostale tlačne napetosti znižujejo koncentracijo napetosti na površini dinamično obremenjenega materiala in tako vplivajo na podaljšanje časovne trdnosti materiala. Izvedene raziskave torej dokazujejo povečanje odpornosti na utrujanje materiala, hladno utrjenega z mikroplastično deformacijo površine. ©2010 Strojniški vestnik. Vse pravice pridržane.

Ključne besede: mehansko hladno utrjevanje, mikroplastično utrjevanje, aluminijeve zlitine, zaostale napetosti, integriteta površine, utrujanje materiala, trujenostna razpoka SI 74

*Naslov avtorja za dopisovanje: Univerza v Ljubljani, Fakulteta za strojništvo, Aškerčeva 6, 1000 Ljubljana, Slovenija, janez.grum@fs.uni-lj.si

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, SI 75

Prejeto: 25.05.2010 Sprejeto: 06.10.2010

Modifikacija površinskega sloja aluminijevih zlitin po utrjevanju z laserskimi udarnimi valovi Trdan, U. – Ocaña, J.L. – Grum, J. Uroš Trdan1 – José Luis Ocaña2 – Janez Grum1,* 1 Univerza of Ljubljani, Fakulteta za strojništvo, Slovenija 2 Centro Láser U.P.M., Politehnična Univerza, Madrid, Španija

Namen prispevka: Cilj opravljene raziskave je izvedba celovite analize utrjevanja tankega površinskega sloja z laserskimi udarnimi valovi na makro in mikro nivoju. Rezultati eksperimenta omogočajo določitev optimalnih parametrov laserskega utrjevanja za zagotovljanje izboljšane integritete površine. Zaradi vpliva plazme ter udarnih valov na stanje površine so bili izvedeni tudi anodni potenciodinamični korozijski testi v 3,5% NaCl vodni raztopini.

Metodologija: Lasersko udarno utrjevanje smo izvedli s Q-preklopnim Nd:YAG laserjem valovne dolžine 1064 nm in s trajanjem pulza 10 ns. Izvedena je bila podrobna analiza vpliva gostote pulzov na dveh vrstah aluminijevih zlitin, glede na hrapavost površine in podprto z mikrotopografsko analizo. Po globini vzorcev, t.j. v prečnem prerezu, smo določili še potek mikrotrdote in zaostalih napetosti. Naravo in velikost nastalih korozijskih poškodb na površini vzorcev po elektrokemičnem testu smo analizirali z elektronskim mikroskopom. Za statistično potrditev optimalnih parametrov obdelave z laserskimi udarnimi valovi smo uporabili faktorsko analizo, kjer prvi faktor eksperimenta predstavlja gostota pulzov 900 oz 2500 pulzi/cm2, drugi faktor pa vrsta materiala oz. dve aluminijevi zlitini AlMgSiPb ter AlSi1MgMn. Ugotovitve: Rezultati raziskave so potrdili, da z laserskim udarnim utrjevanjem izboljšamo tako mehanske lastnosti kot tudi korozijsko odpornost obeh aluminijevih zlitin.

Faktorska analiza je potrdila, da je zlitina AlMgSiPb bolj primeren material z manjšim porastom hrapavosti površine pri obeh gostotah pulzov laserskega utrjevanja. Analiza zaostalih napetosti je potrdila, da z višjo gostoto pulzov laserskega utrjevanja dosežemo večje tlačne zaostale napetosti. Utrjena zlitina AlMgSiPb je izkazala na globini 33 μm največje tlačne zaostale napetosti v višini -314 MPa pri manjši in -337 MPa pri večji gostoti pulzov. Potenciodinamični korozijski testi pa so potrdili, da lasersko udarno utrjevanje povzroča pomik porušitvenega potenciala v smeri zagotavljanja večje korozijske odpornosti.

Elektronska mikroskopija je pojasnila naravo značilnih jamičastih korozijskih razjed. Najhujši napad je bil zabeležen na materialu v dobavljenem stanju, po laserskem utrjevanju pa se s povečanjem gostote pulzov intenziteta jamičastega napada zmanjšuje. Zlitina AlMgSiPb se je izkazala s superiorno korozijsko odpornostjo oz. z manjšo gostoto korozijskih poškodb kot zlitina AlSi1MgMn. Omejitve/uporabnost raziskave: Rezultati raziskave so uporabni le za izbrano gostoto moči ter le za izbrano področje gostote pulzov laserskega snopa. Izvesti bo potrebno še raziskavo vpliva različnih gostot moči laserskega snopa, saj pričakujemo zanimive interakcije z integriteto površinskega sloja materiala.

Izvirnost/pomembnost prispevka: Izvedena raziskava je izredno perspektivna, saj se v industriji spodbujajo inovativni postopki za izboljšanje oz. oplemenitenje površin, predvsem zaradi finančnih prihrankov ob zagotavljanju podaljšanja življenjske dobe strojnih delov. V strokovni literaturi še ni bila opravljena raziskava vpliva laserskega udarnega utrjevanja obravnavanih aluminijevih zlitin. Rezultati raziskave so zato zanimivi in uporabni za navtično in letalsko industrijo, kjer so aluminijeve zlitine sistema Al-Mg-Si zaradi majhne gostote, dobrih mehanskih in odličnih korozijskih lastnosti že pogosto v uporabi. ©2011 Strojniški vestnik. Vse pravice pridržane.

Ključne besede: lasersko udarno utrjevanje, topografija površine, zaostale napetosti, mikrotrdota, potenciodinamični korozijski testi, jamičasta korozija, analiza variance

*Naslov avtorja za dopisovanje: Univerza v Ljubljani, Fakulteta za strojništvo, Aškerčeva 6, 1000 Ljubljana, Slovenija, janez.grum@fs.uni-lj.si

SI 75

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, SI 76

Prejeto: 14.03.2010 Sprejeto: 23.03.2011

Inteligentni elektrohidravlični servopozicionirni sistem Detiček, E. – Župerl, U. Edvard Detiček* - Uroš Župerl Univerza v Mariboru, Fakulteta za strojništvo, Slovenija

Ustaljena zahteva, ki jo sodobna strojegradnja postavlja proizvajalcem pogonskih sistemov, je doseganje velike točnosti pomikov ob velikih hitrostih in silah. Slednje se nanaša tudi na hidravlične pogone, ki se uporabljajo na strojih za preoblikovanje kovin in umetnih mas, na strojih za preizkušanje materialov in konstrukcij, pri montažnih in transportnih napravah ter na nekaterih vrstah obdelovalnih strojev.

Sodobni hidravlični pogon je razen iz napajalnega agregata sestavljen iz servoventila in hidravličnega valja, opremljenega z merilnikom pomika in sile. Omenjene komponente povezuje računalnik v učinkovit mehatronski sistem.

V prispevku so podani teoretični in eksperimentalni rezultati raziskave, ki se ukvarja z iskanjem učinkovitih strategij samodejnega računalniškega vodenja servohidravličnih pogonov.

Raziskava temelji na osnovah klasične teorije adaptivnega vodenja tehničnih sistemov s sklenjeno povratno zvezo, osrednja pozornost pa je namenjena uporabi mehke logike. Poleg osnovne izvedbe mehkega PD-regulatorja je še posebej raziskana možnost izvedbe digitalnega regulacijskega algoritma, ki omogoča učenje in s tem prilagajanje spremenljivim razmeram v realnem industrijskem okolju. Algoritem učenja je povzet po strokovni literaturi in omogoča tudi lastno organiziranje mehkega regulatorja v fazi prvega zagona. Na koncu je omenjeni mehki regulator kombiniran še s postopki iz klasične teorije adaptivnega vodenja.

Postopek učenja uporabimo najprej pri več zaporednih majhnih skokih želene vrednosti. Glavna baza pravil je na samem začetku prazna. Ob omenjenih skokih se začne polniti z ustreznimi pravili, kar predstavlja fazo organizacije.

Po nekaj opravljenih skokih sledi sprememba želene vrednosti v nagibno obliko (rampa). V normalnih delovnih razmerah je takšna sprememba na strojih običajna. V tej fazi eksperimentov je samodejno ustvarjena baza pravil podvržena še nadaljnjemu spreminjanju (učenju). Po določenem številu ponovitev se izkaže, da ustvarjena baza pravil ohranja konstantno obliko, če na sistemu ni prišlo do kakšne pomembne spremembe (npr. spremembe napajalnega tlaka ali zunanje obremenitve). Ker ima regulator še vedno značaj PD-regulatorja, ostajata pri nagibnem signalu želene vrednosti še vedno sledilni pogrešek ter statični pogrešek končnega doseženega položaja zaradi trenja. Sledilni pogrešek deloma odpravimo z dodanim predfiltrom v obliki inverznega modela sistema, medtem ko točnost doseženega položaja izboljšamo z dodatkom vključno-izključnega integratorja. Oba ukrepa sta sicer znana iz klasične regulacijske teorije.

Zaradi pomanjkanja dokazov o konvergentnosti predlaganega postopka učenja so potrebne še nadaljnje raziskave v tej smeri, predvsem pa še nadaljnje raziskave s področja adaptivnega vodenja nelinearnih sistemov.

V prispevku je nov predvsem postopek mehčanja (fuzifikacije) realnih signalov, ki uporablja cela števila namesto lingvističnih oznak mehkih množic. Slednje je omogočilo tudi izvirno izvedbo računalniškega algoritma regulatorja, predvsem pa postopka učenja. Posamezna pravila se tako modificirajo zgolj s prištevanjem oziroma odštevanjem ustreznih vrednosti. Postopek učenja je tako realiziran na izviren način.

Prikazani izsledki so uporabni pri nadaljnjem razvoju regulatorjev – ne samo v hidravliki, temveč na celotnem tehničnem področju. ©2011 Strojniški vestnik. Vse pravice pridržane.

Ključne besede: servohidravlika, premočrtni pogoni, regulacija pomika, nelinearni regulatorji, regulatorji z mehko logiko, samoučljivi regulatorji

SI 76

*Naslov avtorja za dopisovanje: Univerza v Mariboru, Fakulteta za strojništvo, Smetanova 17, 2000 Maribor, Slovenija, edvard.deticek@uni-mb.si

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, SI 77

Prejeto: 13.07.2009 Sprejeto: 12.01.2011

Določanje zmogljivosti terminala za razkladanje razsutega tovora z uporabo teorije vrst

Bugaric, U. ‒ Petrovic, D. ‒ Petrovic, Z. ‒ Pajcin, M. ‒ Markovic-Petrovic, G. Ugljesa Bugaric1,* ‒ Dusan Petrovic1 ‒ Zoran Petrovic2 ‒ Miroslav Pajcin3 ‒ Gordana Markovic-Petrovic4 1 Univerza v Beogradu, Fakulteta za strojništvo, Srbija 2 Tecon Sistem d.o.o., Srbija 3 Jugoimport SDPR, Srbija 4 DZ-Zemun, Srbija Glavni namen članka je razviti prikladnejši, enostavnejši in hitrejši način za analizo rečnih terminalov za razkladanje razsutega tovora, ki bi bil uporaben za inženirsko uporabo v začetnih fazah procesa projektiranja in bi namesto simulacije uporabljal pristop s teorijo vrst.

Specializiran rečni terminal za razkladanje razsutega tovora (rečni terminal) predstavlja sistem različnih dejavnosti, povezanih z upravljanjem in manipulacijo toka materiala iz ladje v transportni ali skladiščni sistem, ki mora zagotavljati maksimalno oskrbo ladij z minimalnimi stroški. Rečni terminali se najpogosteje uporabljajo pri termoelektrarnah za razkladanje premoga, pri jeklarnah za razkladanje železove rude in premoga, pri kemičnih tovarnah za razkladanje surovin itd. Glavna značilnost takšnih terminalov je visoka zmogljivost razkladanja.

Analitični pristop na podlagi teorije vrst analizira dva različna modela rečnih terminalov. Prvi model obravnava delovanje naprav za razkladanje brez strategije, drugi model pa delovanje naprav za razkladanje s strategijo. Pri modeliranju terminala za razkladanje brez strategije (1. model) deluje vsaka od dveh naprav za razkladanje ločeno na enem od dokov. Glavna zamisel članka je povečanje zmogljivosti razkladanja na terminalu z uvedbo strategije delovanja naprav za razkladanje, t.j. s skrajšanjem časa, potrebnega za razkladanje ladje. Zato je bil razvit 2. model, pri katerem naprave za razkladanje delujejo s strategijo. Validacija tega pristopa in modelov je bila izvedena na obstoječem rečnem terminalu za razkladanje razsutega tovora na reki Donavi v Srbiji.

Rezultati analize terminala z uporabo teorije vrst, predstavljeni v tem članku, so prikazani vzporedno z rezultati simulacije modela istega sistema, ki so ga avtorji izdelali pred tem. Vzporedna primerjava obeh skupin rezultatov kaže, da so rezultati simulacije in teorije vrst enake narave. Rezultati se razlikujejo zaradi omejitev teorije vrst: proces obdelave je omejen na eno fazo, povprečni čas za razkladanje ladje je nespremenljiv, vpliv različnih vrst materialov na čas razkladanja ladje je matematično povprečen, uporabljena je ena sama velikost ladje itd.

Predstavljeni rezultati kažejo, da je zmogljivost terminala za razkladanje razsutega tovora pri enakih investicijskih stroških večja, če naprave za razkladanje delajo s strategijo. Enaki investicijski stroški pomenijo enako število naprav za razkladanje, kar se odraža na dolžini operativne obale, enako zmogljivost sidrišča itd. Tudi zelo majhno skrajšanje časa, potrebnega za razkladanje ladje, lahko pomeni prihranek energije pri razkladanju in lahko zmanjša stroške zaradi nezadovoljive obdelave oz. predolgega zadrževanja ladje v sistemu (na terminalu).

Naloga inženirjev v procesu projektiranja je, da določijo zmogljivost sistema kot osnovo za oceno vhodov/izhodov sistema. Zato obstaja tudi potreba po razvoju enostavnejšega in hitrejšega pristopa k projektiranju z ustrezno natančnostjo, kar je bil tudi glavni razlog za razvoj predstavljenega analitičnega modela z uporabo teorije vrst. Z drugimi besedami, novi model lahko zamenja starejše, zahtevnejše simulacijske modele na začetku procesa projektiranja sistema.

Rezultate, dobljene s teorijo vrst, je mogoče uporabiti na začetku procesa projektiranja za prve ocene omejitev sistema, kadar so potrebne grobe ocene obnašanja sistema, rezultate simulacij pa je mogoče uporabiti v fazi podrobnega projektiranja. © 2011 Strojniški vestnik. Vse pravice pridržane.

Keywords: razsuti tovor, razkladanje, rečni terminal, teorija vrst *Naslov avtorja za dopisovanje: Univerza v Beogradu, Fakulteta za strojništvo, Kraljice Marije 16, 11120 Beograd, Srbija, ubugaric@mas.bg.ac.rs

SI 77

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, SI 78

Prejeto: 15.03.2010 Sprejeto: 18.01.2011

Raziskava zmogljivosti rezalnih orodij iz materiala CBN s prevleko TiN in brez nje pri aplikacijah trdega rezkanja Taylan, F. – Çolak, O. – Kayacan, M.C. Fatih Taylan1 – Oğuz Çolak1,* – Mehmet Cengiz Kayacan2 1 Suleyman Demirel Univerza, Tehnična fakulteta, Turčija 2 Suleyman Demirel Univerza, Fakulteta za strojništvo, Turčija

Predmet te študije je raziskava zmogljivosti oz. analiza stabilnosti pri čelnem rezkanju orodnega jekla za delo v hladnem 90MnCrV8, kaljenega na 61 HRC, z rezalnimi orodji iz materiala CBN s prevleko TiN in brez nje. Raziskana je bila obraba ploščic iz materiala CBN s prevleko TiN in brez nje (SNMN090308) pri čelnem rezkanju kaljenega orodnega jekla. Analizirani so bili tudi mehanizmi obrabe orodja in rezalne sile, rezultati pa so predstavljeni v članku.

Cilj študije je definicija analize stabilnosti pri trdem rezkanju za preizkušanje orodij iz materiala CBN s prevlekami in brez njih. Stabilnost eksperimenta z rezkanjem je pomembna za izbiro pogojev preizkušanja obrabe orodja.

Pri preizkušanju obrabe orodij za trdo rezkanje je bilo kot material obdelovanca uporabljeno jeklo 90MnCrV8 (AISI – O2, EU – 90MnCrV8) s trdoto 61 HRC. Pogoji preizkušanja so definirani pod četrto točko. V prvem koraku se merijo in izračunavajo funkcije frekvenčnega odgovora obdelovalnega stroja in rezalnega orodja z dinamično modalno analizo. V drugem koraku se merijo in izračunavajo povprečni rezalni koeficienti s preizkusom utornega rezkanja. V tretjem delu študije so bili analitično simulirani diagrami stabilnosti v odvisnosti od rezalne dinamike in rezalnih koeficientov. V zadnjem koraku študije so bili opravljeni preizkusi obrabe orodij iz materiala CBN s prevleko in brez nje.

Pri trdem čelnem rezkanju orodnega jekla 90MnCrV8 s ploščicami iz materiala CBN s prevleko oz. brez nje so bili dobljeni naslednji rezultati.

1. Življenjska doba orodja pri čelnem rezkanju orodnega jekla 90MnCrV8 v kaljenem stanju (61 HRC), izražena kot dolžina reza, je manjša od 260 mm. Obraba ploščic iz materiala CBN brez prevleke in s prevleko Ti(C,N) + (Ti, Al) N + TiN se razlikuje le malo.

2. Analiza vzorcev obrabe bokov je pokazala, da so za obrabo odgovorni mehanizmi makro krušenja. Orodja PCBN so odpovedala z lomom rezalnega roba.

3. Pri preizkusu obrabe orodij iz materiala CBN brez prevleke je bil ugotovljen 6,25-odstotni lom orodja. Pri rezalnih orodjih iz materiala CBN s prevleko Ti(C,N) + (Ti, Al) N + TiN pa je bil lom orodja 28,13-odstoten.

4. Optimalna rezalna oz. podajalna hitrost je bila za obe kvaliteti rezalnega orodja med 450 in 550 m/min (visoka rezalna hitrost) oz. med 0,05 in 0,1 mm/zob (nizka podajalna hitrost).

5. Rezalne sile so se signifikantno povečale pri večjih rezalnih hitrostih in hitrostih podajanja.

Pri tej študiji so bili izbrani omejeni pogoji rezalne hitrosti in podajanja. Prav tako je bila nespremenljiva globina reza. Pri študiji niso bili upoštevani učinki bele plasti in oblikovanja odrezkov pri trdem rezkanju.

Rezalni parametri za preizkuse obrabe orodja se izbirajo na osnovi ekspertnih ocen in priporočil proizvajalcev rezalnih orodij. V tej študiji pa so ti parametri izbrani tudi na podlagi dinamike rezanja in simulacije mehanike. Produktivnost operacij rezkanja je odvisna od dinamike in mehanike odrezavanja.

Novost v tej študiji je izbira parametrov obrabe orodja. Druga novost študije je obravnava zmogljivosti novih prevlečenih CBN-orodij pri trdem rezkanju. © 2011 Strojniški vestnik. Vse pravice pridržane.

Ključne besede: trdo rezkanje, stabilnost rezkanja, obraba orodja, orodja CBN s prevlekami in brez prevlek

SI 78

*Naslov avtorja za dopisovanje: Suleyman Demirel Univerza, Tehnična fakulteta, 32260, Isparta, Turčija, ocolak@mmf.sdu.edu.tr

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, SI 79

Prejeto: 24.09.2009 Sprejeto: 07.01.2011

Sistem za pogon podajanja pri istosmernem rezkanju, krmiljen s konzervativno silo Branko

Tadic, B. – Vukelic, D. – Hodolic, J. – Mitrovic, S. – Eric, M. – Djordje Vukelic2,* – Janko Hodolic2 – Slobodan Mitrovic1 – Milan Eric1 1 Fakulteta za strojništvo, Univerza v Kragujevcu, Srbija 2 Tehniška fakulteta, Univerza v Novem Sadu, Srbija

Tadic1

Osnovni cilj te raziskave je bil preučiti možnosti za krmiljenje podajanja pri istosmernem rezkanju s konzervativno silo, t.j. s silo hidravličnega dušenja. Zamisel sloni na dejstvu, da je horizontalno rezalno silo mogoče uporabiti kot aktivno silo za premikanje obdelovanca, medtem ko je podajanje možno krmiliti s hidravličnim dušenjem.

Praktična realizacija istosmernega rezkanja, pri katerem je podajanje krmiljeno s hidravličnim dušenjem. Podlaga za realizacijo so bile izčrpne teoretične analize, ki so bile uporabljene za snovanje in izgradnjo posebnega sistema za pogon podajanja.

V članku je predstavljen teoretični model novega sistema za pogon podajanja pri istosmernem rezkanju, na osnovi katerega je bil pogon zasnovan, zgrajen in preizkušen. Eksperimentalne raziskave so dokazale ne le njegovo praktično uporabnost, ampak tudi pomembne učinke pri njegovi uporabi. Takšna vrsta sistema za pogon podajanja ima v primerjavi s konvencionalnimi pogoni bolj kompleksno dinamiko, zato je tudi zelo zanimiv predmet teoretičnih in eksperimentalnih raziskav.

Rezultati potrjujejo osnovno hipotezo, da je hidravlično dušenje mogoče uporabiti za krmiljenje podajanja pri istosmernem rezkanju v širokem razponu vhodnih vrednosti. Poleg tega se življenjska doba in kakovost površine, ki jo daje predlagana metodologija, ujema z rezultati, ki jih pri identičnih pogojih obdelave dajejo sodobni CNC-obdelovalni stroji.

Takšno vrsto sistema za pogon podajanja je danes vsekakor mogoče uporabiti v konvencionalnih obdelovalnih strojih. Prihodnje raziskave bi bilo treba usmeriti v povečanje dinamične togosti, izboljšanje sistema dušenja in samodejno krmiljenje podajalne hitrosti. © 2011 Strojniški vestnik. Vse pravice pridržane.

Ključne besede: istosmerno rezkanje, poseben sistem za pogon podajanja, zvezno krmiljenje podajalne hitrosti, dušenje

*Naslov avtorja za dopisovanje: Tehniška fakulteta, Univerza v Novem Sadu, Trg Dositeja Obradovica 6, 21000 Novi Sad, Srbija, vukelic@uns.ac.rs

SI 79

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, SI 80-81 Navodila avtojem

Navodila avtorjem Članke pošljite na naslov: Strojniški vestnik Journal of Mechanical Engineering Aškerčeva 6, 1000 Ljubljana, Slovenija Tel.: 00386 1 4771 137 Faks: 00386 1 2518 567 E-mail: info@sv-jme.eu strojniski.vestnik@fs.uni-lj.si Članki morajo biti napisani v angleškem jeziku. Strani morajo biti zaporedno označene.

Prispevki so lahko dolgi največ 10 strani. Daljši članki so lahko v objavo sprejeti iz posebnih razlogov, katere morate navesti v spremnem dopisu.

Kratki članki naj ne bodo daljši od štirih strani.

Navodila so v celoti na voljo v rubriki “Informacija za avtorje” na spletni strani revije: http://en.sv-jme.eu/ Prosimo vas, da članku priložite spremno pismo, ki naj vsebuje: 1. naslov članka, seznam avtorjev ter podatke avtorjev; 2. opredelitev članka v eno izmed tipologij; izvirni znanstveni (1.01), pregledni znanstveni (1.02) ali kratki znanstveni članek (1.03); 3. opredelitev, da članek ni objavljen oziroma poslan v presojo za objavo drugam; 4. zaželeno je, da avtorji v spremnem pismu opredelijo ključni doprinos članka; 5. predlog dveh potencialnih recenzentov, ter kontaktne podatke recenzentov. Navedete lahko tudi razloge, zaradi katerih ne želite, da bi določen recenzent recenziral vaš članek.

OBLIKA ČLANKA Članek naj bo napisan v naslednji obliki: Naslov, ki primerno opisuje vsebino članka.

Povzetek, ki naj bo skrajšana oblika članka in naj ne presega 250 besed. Povzetek mora vsebovati osnove, jedro in cilje raziskave, uporabljeno metodologijo dela, povzetek rezultatov in osnovne sklepe.

- Uvod, v katerem naj bo pregled novejšega stanja in zadostne informacije za razumevanje ter pregled rezultatov dela, predstavljenih v članku.

- Teorija.

- -

SI 80

Eksperimentalni del, ki naj vsebuje podatke o postavitvi preskusa in metode, uporabljene pri pridobitvi rezultatov.

- Rezultati, ki naj bodo jasno prikazani, po potrebi v obliki slik in preglednic.

- Razprava, v kateri naj bodo prikazane povezave in posplošitve, uporabljene za pridobitev rezultatov. Prikazana naj bo tudi pomembnost rezultatov in primerjava s poprej objavljenimi deli. (Zaradi narave posameznih raziskav so lahko rezultati in razprava, za jasnost in preprostejše bralčevo razumevanje, združeni v eno poglavje.)

- Sklepi, v katerih naj bo prikazan en ali več sklepov, ki izhajajo iz rezultatov in razprave.

- Literatura, ki mora biti v besedilu oštevilčena zaporedno in označena z oglatimi oklepaji [1] ter na koncu članka zbrana v seznamu literature.

Enote - uporabljajte standardne SI simbole in okrajšave. Simboli za fizične veličine naj bodo v ležečem tisku (npr. v, T, n itd.). Simboli za enote, ki vsebujejo črke, naj bodo v navadnem tisku (npr. ms1, K, min, mm itd.)

Okrajšave naj bodo, ko se prvič pojavijo v besedilu, izpisane v celoti, npr. časovno spremenljiva geometrija (ČSG).

Pomen simbolov in pripadajočih enot mora biti vedno razložen ali naveden v posebni tabeli na koncu članka pred referencami.

Slike morajo biti zaporedno oštevilčene in označene, v besedilu in podnaslovu, kot sl. 1, sl. 2 itn. Posnete naj bodo v ločljivosti, primerni za tisk, v kateremkoli od razširjenih formatov, npr. BMP, JPG, GIF. Diagrami in risbe morajo biti pripravljeni v vektorskem formatu, npr. CDR, AI. Vse slike morajo biti pripravljene v črnobeli tehniki, brez obrob okoli slik in na beli podlagi.

Ločeno pošljite vse slike v izvirni obliki Pri označevanju osi v diagramih, kadar je le mogoče, uporabite označbe veličin (npr. t, v, m itn.). V diagramih z več krivuljami, mora biti vsaka krivulja označena. Pomen oznake mora biti pojasnjen v podnapisu slike.

Tabele naj imajo svoj naslov in naj bodo zaporedno oštevilčene in tudi v besedilu poimenovane kot Tabela 1, Tabela 2 itd.. Poleg fizikalne veličine, npr t (v ležečem tisku), mora biti v oglatih oklepajih navedena tudi enota. V tabelah naj se ne podvajajo podatki, ki se nahajajo v besedilu.

-

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, SI 80-81

Potrditev sodelovanja ali pomoči pri pripravi članka je lahko navedena pred referencami. Navedite vir finančne podpore za raziskavo.

REFERENCE Seznam referenc MORA biti vključen v članek, oblikovan pa mora biti v skladu s sledečimi navodili. Navedene reference morajo biti citirane v besedilu. Vsaka navedena referenca je v besedilu oštevilčena s številko v oglatem oklepaju (npr. [3] ali [2] do [6] za več referenc). Sklicevanje na avtorja ni potrebno.

Reference morajo biti oštevilčene in razvrščene glede na to, kdaj se prvič pojavijo v članku in ne po abecednem vrstnem redu. Reference morajo biti popolne in točne. Vse neangleške oz. nenemške naslove je potrebno prevesti v angleški jezik z dodano opombo (in Slovene) na koncu Navajamo primere: Članki iz revij: Priimek 1, začetnica imena, priimek 2, začetnica imena (leto). Naslov. Ime revije, letnik, številka, strani.

[1] Zadnik, Ž., Karakašič, M., Kljajin, M., Duhovnik, J. (2009). Function and Functionality in the Conceptual Design Process. Strojniški vestnik – Journal of Mechanical Engineering, vol. 55, no. 7-8, p. 455-471.

Ime revije ne sme biti okrajšano. Ime revije je zapisano v ležečem tisku.

Knjige: Priimek 1, začetnica imena, priimek 2, začetnica imena (leto). Naslov. Izdajatelj, kraj izdaje [2] Groover, M. P. (2007). Fundamentals of Modern Manufacturing. John Wiley & Sons, Hoboken.

Ime knjige je zapisano v ležečem tisku.

Poglavja iz knjig: Priimek 1, začetnica imena, priimek 2, začetnica imena (leto). Naslov poglavja. Urednik(i) knjige, naslov knjige. Izdajatelj, kraj izdaje, strani.

[3] Carbone, G., Ceccarelli, M. (2005). Legged robotic systems. Kordić, V., Lazinica, A., Merdan, M. (Eds.), Cutting Edge Robotics. Pro literatur Verlag, Mammendorf, p. 553-576. Članki s konferenc: Priimek 1, začetnica imena, priimek 2, začetnica imena (leto). Naslov. Naziv konference, strani.

[4] Štefanić, N., Martinčević-Mikić, S., Tošanović, N. (2009). Applied Lean System in Process Industry. MOTSP 2009 Conference Proceedings, p. 422-427.

Standardi: Standard (leto). Naslov. Ustanova. Kraj.

[5] ISO/DIS 16000-6.2:2002. Indoor Air – Part 6: Determination of Volatile Organic Compounds in Indoor and Chamber Air by Active Sampling on TENAX TA Sorbent, Thermal Desorption and Gas Chromatography using MSD/FID. International Organization for Standardization.

Geneva.

Spletne strani: Priimek, Začetnice imena podjetja. Naslov, z naslova http://naslov, datum dostopa.

[6] Rockwell Automation. Arena, from http://www. arenasimulation.com, accessed on 2009-09-27.

RAZŠIRJENI POVZETEK Ko je članek sprejet v objavo, avtorji pošljejo razširjeni povzetek na eni strani A4 (približno 3.000 - 3.500 znakov). Navodila za pripravo razširjenega povzetka so objavljeni na spletni strani http://sl.svjme.eu/informacije-za-avtorje/.

AVTORSKE PRAVICE Avtorji v uredništvo predložijo članek ob predpostavki, da članek prej ni bil nikjer objavljen, ni v postopku sprejema v objavo drugje in je bil prebran in potrjen s strani vseh avtorjev. Predložitev članka pomeni, da se avtorji avtomatično strinjajo s prenosom avtorskih pravic SV-JME, ko je članek sprejet v objavo. Vsem sprejetim člankom mora biti priloženo soglasje za prenos avtorskih pravic, katerega avtorji pošljejo uredniku. Članek mora biti izvirno delo avtorjev in brez pisnega dovoljenja izdajatelja ne sme biti v katerem koli jeziku objavljeno drugje.

Avtorju bo v potrditev poslana zadnja verzija članka. Morebitni popravki morajo biti minimalni in poslani v kratkem času. Zato je pomembno, da so članki že ob predložitvi napisani natančno.

Avtorji lahko stanje svojih sprejetih člankov spremljajo na http://en.sv-jme.eu/.

PLAČILO OBJAVE Domači avtorji vseh sprejetih prispevkov morajo za objavo plačati prispevek, le v primeru, da članek presega dovoljenih 10 strani oziroma za objavo barvnih strani v članku, in sicer za vsako dodatno stran 20 EUR ter dodatni strošek za barvni tisk, ki znaša 90,00 EUR na stran.

SI 81

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, SI 82-84 Osebne objave

Doktorati, magisteriji, specialistična dela in diplome

DOKTORAT ZNANOSTI Na Fakulteti za strojništvo Univerze v Ljubljani je z uspehom obranil svojo doktorsko disertacijo: dne 22. aprila 2011 Gregor ALIČ z naslovom: »Zaščitna mreža aksialnih turbinskih strojev v tokovnem polju« (mentor: prof. dr.

Branko Širok, somentor: izr. prof. dr. Marko Hočevar); To delo predstavlja vpliv oblikovanja konture zaščitne mreže aksialnega ventilatorja na njegovo aerodinamsko integralno karakteristiko, spekter ravni zvočnega tlaka in skupno raven zvočne moči.

Obratovalne karakteristike aksialnega ventilatorja brez mreže, aksialnega ventilatorja s standardno mrežo ter aksialnega ventilatorja z optimirano mrežo so bile analizirane v štirih karakterističnih obratovalnih točkah.

Rezultati lokalnih meritev izstopnih hitrosti aksialnega ventilatorja brez mreže pri optimalni obratovalni točki so bili uporabljeni za določitev konture optimirane zaščitne mreže. Zaščitna mreža aksialnega ventilatorja ima značilen vpliv na njegovo aerodinamsko obratovalno karakteristiko. Vpliv zaščitne mreže se še značilneje odraža v spektru ravni zvočnega tlaka preko frekvenčnega območja von Karmanove vrtinčne steze in posledično tudi na skupni ravni zvočne moči primerjanih aksialnih ventilatorjev.

Lastnosti tokovnega polja za zaščitno mreže so bile preučene z uporabo linijske cilindrične kaskade v prostem tokovnem polju pri primerljivih Reynoldsovih številih. Rezultati so predstavljeni kot močnostni spektri fluktuacije hitrosti ter kot spekter ravni zvočnega tlaka. Rezultati meritev, opravljenih na linijski cilindrični kaskadi, se dobro ujemajo z rezultati meritev, opravljenih na primerjanih aksialnih ventilatorjih. * Na Fakulteti za strojništvo Univerze v Mariboru je z uspehom obranil svojo doktorsko disertacijo: dne 15. aprila 2011 Niko ROZMAN z naslovom: »Razvoj visokotrdnostnih livnih aluminijevih zlitin s kvazikristali« (mentor: izr. prof. dr. Franc Zupanič); SI 82

V doktorski disertaciji je obravnavana problematika razvoja aluminijevih zlitin, ki vsebujejo nekatere izmed kompleksnih intermetalnih spojin, med drugim tudi kvazikristalne faze. Raziskovali smo vplive legirnih elementov in pogojev strjevanja na razvoj mikrostrukture ter posledičnega mehanskega obnašanja zlitin štirikomponentnega sistema Al-Mn-Be-Cu. Zato je bilo izdelanih več zlitin, ki so bile lite pri različnih pogojih.

Razvoj zlitin je potekal na podlagi znane konstitucije trikomponentnega sistema Al-MnCu in znanih vplivov kemijskega elementa berilija na aluminijeve zlitine sistema AlMn. S pomočjo raznih analiz, kot na primer mikrostrukturnih, kemijskih, termodinamskih in faznih, smo ugotovili fazno sestavo zlitin in razvoj mikrostrukture glede na kemijsko sestavo in ravnotežne razmere pri strjevanju. Med litjem smo zajemali ohlajevalne hitrosti in opredelili kritično ohlajevalno hitrost, potrebno za nastanek ikozaedrične kvazikristalne faze v sistemu AlMn-Be-Cu. Za odkrivanje vplivov fazne sestave na mehanske lastnosti zlitin so bili izvedeni tlačni in trdnostni preizkusi ter testi meritev mikrotrdote z uporabo naprave za zaznavanje globine vtiska.

Ugotovljeno je bilo, da kvazikristalna faza, ki nastopa v teh zlitinah, povišuje napetost tečenja in duktilnost zlitin, zmanjšuje pa utrjevalni efekt med plastično deformacijo; dne 19. aprila 2011 Simon BREZOVNIK z naslovom: »Optimizacija delovanja izdelovalnih strojev in sistemov z uporabo skupinske inteligence« (mentor: izr. prof. dr. Miran Brezočnik); Modernizacija sodobne proizvodnje vključuje nenehno posodabljanje in integracijo najnovejših tehnologij v proizvodne sisteme.

Vključevanje sodobnih tehnologij omogoča skrajševanje časa izdelave, povečanje zmogljivosti in zniževanje proizvodnih stroškov. Vzporedno z visoko stopnjo avtomatizacije sodobnih proizvodnih sistemov se povečuje tudi smotrnost individualizacije tržišča v smeri maloserijske proizvodnje. Zaradi dinamičnosti razvoja sodobnih tehnologij je učinkovito usklajevanje (t.j. optimiranje) materialnih, energetskih in informacijskih tokov še mnogo težje, kot je bilo v

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, SI 82-84

preteklosti. Znotraj množice vse bolj kompleksnih proizvodnih scenarijev optimalnega toka proizvodnje s klasičnimi metodami načrtovanja ni mogoče več doseči; Zaradi omenjenih razlogov je bil v doktorski disertaciji razvit optimizacijski sistem, ki ponuja inovativne rešitve optimizacije obdelovalnih, robotskih, nadzornih in montažnih sistemov z algoritmi umetne skupinske inteligence. S predlaganim pristopom je predstavljeno reševanje problemov izdelovalnih sistemov po zgledih iz narave. Algoritmi umetne skupinske inteligence omogočajo optimizacijo na samoorganizacijski način, kar daje pomembno prednost pred ostalimi optimizacijskimi metodami. V ta namen je bila opravljena preslikava naravnih zakonitosti kolonialno organiziranih bioloških organizmov v obliko matematičnih definicij in pravil, ki so bile uporabljene v optimizacijskih postopkih načrtovanja izdelovalnih strojev in sistemov; Optimizacijski sistem je sestavljen iz modula napovedovalnega sistema in modula sistema evalvacije. Proces optimizacije poteka na podlagi povratnozančnega izmenjevanja informacij med napovedjo in evalvacijo načrtovanja izdelovalnega sistema. Evalvacija napovedi načrtovanega izdelovalnega sistema se odvija v simulacijskem okolju računalniško podprtega konstruiranja, kar poveča uporabnost in prilagodljivost razvitega optimizacijskega sistema v praksi. Sistem evalvacije je neodvisen od napovedovalnega sistema, kar predstavlja univerzalen in fleksibilen pristop k inteligentnemu načrtovanju in modeliranju proizvodnih sistemov; Z uporabo razvitega univerzalnega optimizacijskega sistema predlagamo učinkovite rešitve inteligentnega načrtovanja in modeliranja naslednjih tehnoloških problemov: (i) optimizacija postavitve surovca v delovni prostor izdelovalnega sistema glede na gibljivost robotskega mehanizma, (ii) analiza izdelovalnosti obdelovanca glede na mesto vpetja, (iii) optimizacija simultanega obdelovalnega sistema z več robotskimi mehanizmi, (iv) tekmovanje robotskih mehanizmov za izvedbo tehnološkega procesa z oceno optimalne izdelovalnosti, (v) načrtovanje obdelovalnega sistema s hibridnim »Fuzzy-Swarm« optimizacijskim algoritmom, (vi) optimizacijski sistem za načrtovanje razmestitve robotskih obdelovalnih sistemov glede na minimalno pot obdelovanca in (vii) optimizacija regalnega skladiščnega sistema;

Za namen validacije rezultatov rešitev, ki jih predlaga optimizacijski sistem, je bila razvita projekcija hitrostne anizotropije delovnih prostorov robotskih mehanizmov z barvno interpolacijo. Predstavljena rešitev ponuja ključno orodje pri načrtovanju razmestitve robotiziranega tehnološkega postopka v področje delovnega prostora z optimalno gibljivostjo robotskega mehanizma; Za učinkovito delovanje optimizacijskega sistema evalvacije je bil razvit postopek dinamičnih meritev položaja in zaznavanje dotika (kolizije) med gibajočimi se deli v trirazsežnem prostoru. Omenjeni pristop omogoča dinamične meritve položaja objektov v razvojnem okolju pri načrtovanju optimalne razmestitve tehnološkega procesa s postopkom optimizacije umetne skupinske inteligence.

MAGISTERIJ ZNANOSTI Na Fakulteti za strojništvo Univerze v Ljubljani je z uspehom zagovarjal svoje magistrsko delo: dne 7. aprila 2011 Aljoša ROŽMAN z naslovom: »Integrirano osvajanje serijskega izdelka« (mentor: prof. dr. Marko Starbek, somentor: izr. prof. dr. Janez Kušar). * Na Fakulteti za strojništvo Univerze v Mariboru je z uspehom zagovarjala svoje magistrsko delo: dne 11. aprila 2011 Monika BRGLEZ z naslovom: »Možnost uporabe soje, oljnih hibridov koruze in rička v proizvodnji biodizla« (mentor: prof. dr. Željko Knez).

DIPLOMIRALI SO Na Fakulteti za strojništvo Univerze v Ljubljani so pridobili naziv univerzitetni diplomirani inženir strojništva: dne 26. aprila 2011: Matic KUMER z naslovom: »Energijska sanacija sakralnega spomenika Plečnikove kulturne dediščine« (mentor: prof. dr. Vincenc Butala); Alan LASIČ z naslovom: »Inovativnost pri vibracijskem vrtanju globokih lukenj« (mentor: SI 83

Strojniški vestnik - Journal of Mechanical Engineering 57(2011)5, SI 82-84

prof. dr. Janez Kopač, somentor: izr. prof. dr.

Borut Likar); Matej PERDEC z naslovom: »Koncept mreže za podporo hitri izdelavi prototipov in končnih izdelkov« (mentor: prof. dr. Janez Kopač, somentorica: izr. prof. dr. Slavko Dolinšek). * Na Fakulteti za strojništvo Univerze v Ljubljani so pridobili naziv diplomirani inženir strojništva: dne 13. aprila 2011: Dejan BATISTA z naslovom: »Smotrnost uporabe orodnega jekla Toolox in njegova obdelovalnost« (mentor: prof. dr. Janez Kopač, somentor: viš. pred. dr. Jože Jurkovič); Matija BERTONCELJ z naslovom: »Popravilo lopatic reaktivnega letalskega motorja« (mentor: prof. dr. Janez Kopač, somentor: doc. dr.

Damjan Klobčar); Andrej MAČEK z naslovom: »Načrtovanje preizkusnega orodja za brizgalno pihanje« (mentor: izr. prof. dr. Zlatko Kampuš); dne 15. aprila 2011: Peter FRANCA z naslovom: »Generator ozona« (mentor: doc. dr. Marjan Jenko); Borut NOVŠAK z naslovom: »Hidravlična naprava stiskalnice za vzdrževalno službo« (mentor: doc. dr. Jožef Pezdirnik); Tomo PEČAVER z naslovom: »Projektiranje tlačnega cevovoda za pretok 40 m3/s« (mentor: izr. prof. dr. Janez Kramar);

SI 84

Peter SELČAN z naslovom: »Membranske tehnologije čiščenja izcednih vod« (mentor: prof. dr. Iztok Golobič). * Na Fakulteti za strojništvo Univerze v Mariboru so pridobili naziv diplomirani inženir strojništva: dne 21. aprila 2011: Mateša BRIŠEVAC z naslovom: »Tehnološka zasnova in konstruiranje progresivnega orodja za preoblikovanje pločevine« (mentor: izr. prof. dr. Ivan Pahole); Primož KOPRIVNIKAR z naslovom: »Razvoj potrebnih znanj iz tehnike odrezavanja in proizvodnih meritev v podjetju Kopit d.o.o.« (mentor: prof. dr. Franci Čuš, somentor: izr. prof. dr. Bojan Ačko); Boštjan RADELJIĆ z naslovom: »Vpliv parametrov na karakteristike procesa vbrizgavanja« (mentorica: prof. dr. Breda Kegl, somentor: izr. prof. dr. Stanislav Pehan); Gregor RAKOVEC z naslovom: »Numerično kopiranje pri izdelavi glasbil« (mentor: izr. prof. dr. Ivan Pahole, somentor: prof. dr. Jože Balič); Žiga SKALE z naslovom: »Zasnova transfer orodja za preoblikovanje pločevine iz aluminijeve zlitine« (mentor: izr. prof. dr. Ivan Pahole, somentor: doc. dr. Janez Kramberger); Roman VALENČAK z naslovom: »Paletizacija vreč s pomočjo robota« (mentor: izr. prof. dr. Miran Brezočnik, somentor: Simon Brezovnik).

Platnica SV-JME 57(2011)5_kor1.ai 2 17.5.2011 7:52:14

Strojniški vestnik – Journal of Mechanical Engineering (SV-JME) Aim and Scope The international journal publishes original and (mini)review articles covering the concepts of materials science, mechanics, kinematics, thermodynamics, energy and environment, mechatronics and robotics, fluid mechanics, tribology, cybernetics, industrial engineering and structural analysis.

The journal follows new trends and progress proven practice in the mechanical engineering and also in the closely related sciences as are electrical, civil and process engineering, medicine, microbiology, ecology, agriculture, transport systems, aviation, and others, thus creating a unique forum for interdisciplinary or multidisciplinary dialogue.

The international conferences selected papers are welcome for publishing as a special issue of SV-JME with invited co-editor(s).

Editor in Chief Vincenc Butala University of Ljubljana Faculty of Mechanical Engineering, Slovenia Co-Editor Borut Buchmeister University of Maribor Faculty of Mechanical Engineering, Slovenia Technical Editor Pika Škraba University of Ljubljana Faculty of Mechanical Engineering, Slovenia

C

M

Y

CM

MY

CY

CMY

K

Editorial Office University of Ljubljana (UL) Faculty of Mechanical Engineering SV-JME Aškerčeva 6, SI-1000 Ljubljana, Slovenia Phone: 386-(0)1-4771 137 Fax: 386-(0)1-2518 567 E-mail: info@sv-jme.eu http://www.sv-jme.eu Founders and Publishers University of Ljubljana (UL) Faculty of Mechanical Engineering, Slovenia University of Maribor (UM) Faculty of Mechanical Engineering, Slovenia Association of Mechanical Engineers of Slovenia Chamber of Commerce and Industry of Slovenia Metal Processing Industry Association Cover: EOS Formiga P100 SLS machine (at the background), Selective Laser Sintering process (small photo above) and Additive Manufacturing products (small photo below).

Image courtesy: Intelligent Manufacturing Systems Laboratory, Production Engineering Institute, Faculty of Mechanical Engineering, University of Maribor

ISSN 0039-2480 © 2011 Strojniški vestnik - Journal of Mechanical Engineering. All rights reserved. SV-JME is indexed / abstracted in: SCI-Expanded, Compendex, Inspec, ProQuest-CSA, SCOPUS, TEMA. The list of the remaining bases, in which SV-JME is indexed, is available on the website. The journal is subsidized by Slovenian Book Agency.

International Editorial Board Koshi Adachi, Graduate School of Engineering,Tohoku University, Japan Bikramjit Basu, Indian Institute of Technology, Kanpur, India Anton Bergant, Litostroj Power, Slovenia Franci Čuš, UM, Faculty of Mech. Engineering, Slovenia Narendra B. Dahotre, University of Tennessee, Knoxville, USA Matija Fajdiga, UL, Faculty of Mech. Engineering, Slovenia Imre Felde, Bay Zoltan Inst. for Mater. Sci. and Techn., Hungary Jože Flašker, UM, Faculty of Mech. Engineering, Slovenia Bernard Franković, Faculty of Engineering Rijeka, Croatia Janez Grum, UL, Faculty of Mech. Engineering, Slovenia Imre Horvath, Delft University of Technology, Netherlands Julius Kaplunov, Brunel University, West London, UK Milan Kljajin, J.J. Strossmayer University of Osijek, Croatia Janez Kopač, UL, Faculty of Mech. Engineering, Slovenia Franc Kosel, UL, Faculty of Mech. Engineering, Slovenia Thomas Lübben, University of Bremen, Germany Janez Možina, UL, Faculty of Mech. Engineering, Slovenia Miroslav Plančak, University of Novi Sad, Serbia Brian Prasad, California Institute of Technology, Pasadena, USA Bernd Sauer, University of Kaiserlautern, Germany Brane Širok, UL, Faculty of Mech. Engineering, Slovenia Leopold Škerget, UM, Faculty of Mech. Engineering, Slovenia George E. Totten, Portland State University, USA Nikos C. Tsourveloudis, Technical University of Crete, Greece Toma Udiljak, University of Zagreb, Croatia Arkady Voloshin, Lehigh University, Bethlehem, USA President of Publishing Council Jože Duhovnik UL, Faculty of Mechanical Engineering, Slovenia Print Tiskarna Present d.o.o., Ižanska cesta 383, Ljubljana, Slovenia General information Strojniški vestnik – The Journal of Mechanical Engineering is published in 11 issues per year (July and August is a double issue).

Institutional prices include print & online access: institutional subscription price and foreign subscription €100,00 (the price of a single issue is €10,00); general public subscription and student subscription €50,00 (the price of a single issue is €5,00). Prices are exclusive of tax. Delivery is included in the price. The recipient is responsible for paying any import duties or taxes.

Legal title passes to the customer on dispatch by our distributor.

Single issues from current and recent volumes are available at the current single-issue price. To order the journal, please complete the form on our website. For submissions, subscriptions and all other information please visit: http://en.sv-jme.eu/.

You can advertise on the inner and outer side of the back cover of the magazine. The authors of the published papers are invited to send photos or pictures with short explanation for cover content.

We would like to thank the reviewers who have taken part in the peer-review process.

Strojniški vestnik - Journal of Mechanical Engineering is also available on http://www.sv-jme.eu, where you access also to papers’ supplements, such as simulations, etc.

57 (2011) 5 1

Platnica SV-JME 57(2011)5_kor1.ai 1 17.5.2011 7:52:01

Since 1955

C

M

Y

CM

MY

CY

CMY

Journal of Mechanical Engineering - Strojniški vestnik

K

5 year 2011 volume 57 no.