EnginSoft Newsletter 10-3 by EnginSoft

Newsletter EnginSoft Year 7 n°3 -

EnginSoft Flash

Creativity is one of the best ingredients for innovation – says Mr Nazario Bellato of Magneti Marelli Powertrain Italy in his recent interview with EnginSoft which we are proud to present to our readers in this latest edition of the Newsletter. Mr Bellato explains the principles of complex technological processes that lead to successful products: Each competence Ing. Stefano Odorizzi contributes to a unique EnginSoft CEO and President final result. Strong competitive advantage is based on many reliable technical results! At EnginSoft each engineer and professional contributes his/her knowledge to our Network of engineering expertise that has expanded from Italy to France, Spain, Sweden onto Germany, the UK, Greece and the USA since 2006. th

On 8 December 2010, Stanford University will host a unique Workshop on Optimization as part of its affiliation with Cascade Technologies and EnginSoft. We ask our readers to refer to the Event Calendar on page 66 for more information. Initiatives like these are the outcomes of years of knowledge exchange, personal efforts, trust, close collaboration or what we like to call: creative international networking ! Fall 2010 sees the EnginSoft teams working intensively on the preparations of the EnginSoft International Conference, 21st–22nd October 2010, Fiera Montichiara/Brescia, Italy. The event is a culmination of knowledge of engineering, CAE, simulation and Virtual Prototyping, brought together in Italy from Italy and around the world. An “excerpt” of this knowledge is presented to our readers on the following pages. Articles this time include among others: • Aerodynamic and acoustic optimization of radial fans by Technical Faculty Friedrich-Alexander-University Erlangen-Nuremberg; • Combustion noise prediction in a small diesel engine by Instituto Motori CNR and Università di Napoli;

• Illumination analysis and design optimization of an automotive speed meter by DENSO Corporation Japan; • Simple optimization of gradated biomaterial scaffolds..by Aalto University Foundation, Finland; • Reliability based structural optimization of an aircraft wing by Istanbul Technical University; • Model of a multimass hyperelastic system and its parametric identification by Tula State University, Russia; • An introduction of Feat Group, a world leader within the field of steel forging. Moreover, we wish to inform our readership about the latest software upgrades and applications with: • an outlook on ANSYS 13; • an example of how Ansoft Maxwell2D/3D e RMxprt and ANSYS are used to predict the functioning of an electrical engine; • a benchmark by Ansaldo Energia with ANSYS EKM (Engineering Knowledge Manager); • the ESAComp 4.1 Release Notes and an example for the analysis of composite materials; • an introduction of the software Coldform 2010; • FTI’s Forming Suite for cost optimization and forming simulation; • a presentation of Kraken, a reservoir simulation postprocessor developed by Engineering Simulation and Scientific Software (ESSS) Brazil; • a simple parallel implementation of a FEM solver in Scilab • a summary and emphasis on the importance of Simulation-Quality Material Data by DatapointLabs; • an article about CADdoctor for product data quality in PLM. Our corporate news feature, among others, EnginSoft’s new membership with E2BA, the Energy Efficient Buildings Association, and our certification with EMAS, the European Eco-Management and Audit Scheme that evaluates the environmental performance of enterprises. After two technical articles, the Japan Column tells us about the culture of wood and Wood MONODUKURI in the land of the rising sun. The Editorial Team and EnginSoft look forward to welcoming our readers to this year’s Conference and the beautiful Lake Garda region - Please meet us to discuss opportunities and let us share our knowledge to foster innovation! Stefano Odorizzi Editor in chief

- Newsletter EnginSoft Year 7 n°3

Sommario - Contents CASE STUDIES

Combustion Noise Prediction in a Small Diesel Engine Finalized to the Optimization of the Fuel Injection Strategy

13 16 19 22

Aerodynamic and Acoustic Optimization of Radial Fans Electromagnetic Interference: an Advanced FEM Calculation Approach Reliability Based Structural Optimization of an Aircraft Wing Simple Optimization of Gradated Biomaterial Scaffolds made of Calcium Phospates

SOFTWARE NEWS

24 26

ANSYS 13: Preview I prodotti ANSYS al servizio della progettazione e della simulazione dei motori elettrici: La verticalizzazione RMxprt-Maxwell-ANSYS Mechanical.

29 32 33 35 36 38

The New ANSYS Frontier Product: ANSYS EKM (Engineering Knowledge Manager) ESAComp Versione 4.1 – Strumento basilare per la progettazione delle strutture in composito ESAComp 4.1: New Features Peculiarità del software Coldform® Preventivazione e Valutazione di Formabilità Pushing Reservoir Data Handling to New Frontiers

IN DEPTH STUDIES

41 46 48

A Simple Parallel Implementation of a FEM Solver in Scilab The Need for “Simulation-Quality” Material Data Model of a Multimass Hyperelastic System and its Parametric Identification

INTERVIEWS

An Interview with Mr Nazario Bellato, Simulation Manager of Magneti Marelli Powertrain

The EnginSoft Newsletter editions contain references to the following products which are trademarks or registered trademarks of their respective owners:

MAGMASOFT is a trademark of MAGMA GmbH. (www.magmasoft.com)

ANSYS, ANSYS Workbench, AUTODYN, CFX, FLUENT and any and all ANSYS, Inc. brand, product, service and feature names, logos and slogans are registered trademarks or trademarks of ANSYS, Inc. or its subsidiaries in the United States or other countries. [ICEM CFD is a trademark used by ANSYS, Inc. under license]. (www.ANSYS.com)

Forge and Coldform are trademarks of Transvalor S.A. (www.transvalor.com)

modeFRONTIER is a trademark of ESTECO EnginSoft Tecnologie per l’Ottimizzazione srl. (www.esteco.com)

LS-DYNA® is a trademark of Livermore Software Technology Corporation. (www.lstc.com)

Flowmaster is a registered trademark of The Flowmaster Group BV in the USA and Korea. (www.flowmaster.com)

SCULPTOR is a trademark of Optimal Solutions Software, LLC (www.optimalsolutions.us)

ESAComp is a trademark of Componeering Inc. (www.componeering.com)

AdvantEdge is a trademark of Third Wave Systems (www.thirdwavesys.com)

Grapheur is a product of Reactive Search SrL, a partner of EnginSoft For more information, please contact the Editorial Team

Newsletter EnginSoft Year 7 n°3 -

JAPAN CAE COLUMN

Illumination Analysis and Design Optimization of an Automotive Speed Meter

The Culture of Wood

RESEARCH AND TECHNOLOGY TRANSFER

EnginSoft Joined the E2BA

TESTIMONIAL

FEAT Group: We forge all you need

EVENTS

Enginsoft ha partecipato allo Users’ Meeting Europeo di FORGE il 7 e 8 giugno 2010 a Sophia Antipolis, Francia

66 66

Newsletter EnginSoft Year 7 n°3 - Autumn 2010 To receive a free copy of the next EnginSoft Newsletters, please contact our Marketing office at: newsletter@enginsoft.it

Elysium’s CADdoctor Enriches Product Data Quality in PLM

EnginSoft contributes to the LION5 Conference EnginSoft Event Calendar

All pictures are protected by copyright. Any reproduction of these pictures in any media and by any means is forbidden unless written authorization by EnginSoft has been obtained beforehand. ©Copyright EnginSoft Newsletter.

Advertisement For advertising opportunities, please contact our Marketing office at: newsletter@enginsoft.it

EnginSoft S.p.A. 24124 BERGAMO Via Galimberti, 8/D Tel. +39 035 368711 • Fax +39 0461 979215 50127 FIRENZE Via Panciatichi, 40 Tel. +39 055 4376113 • Fax +39 0461 979216 35129 PADOVA Via Giambellino, 7 Tel. +39 49 7705311 • Fax 39 0461 979217 72023 MESAGNE (BRINDISI) Via A. Murri, 2 - Z.I. Tel. +39 0831 730194 • Fax +39 0461 979224 38123 TRENTO fraz. Mattarello - via della Stazione, 27 Tel. +39 0461 915391 • Fax +39 0461 979201 www.enginsoft.it - www.enginsoft.com e-mail: info@enginsoft.it

COMPANY INTERESTS

CONSORZIO TCN 38123 TRENTO Via della Stazione, 27 - fraz. Mattarello Tel. +39 0461 915391 • Fax +39 0461 979201 www.consorziotcn.it EnginSoft GmbH - Germany EnginSoft UK - United Kingdom EnginSoft France - France EnginSoft Nordic - Sweden Aperio Tecnologia en Ingenieria - Spain www.enginsoft.com

PAGE 24 ANSYS 13: NEW RELEASE

ASSOCIATION INTERESTS NAFEMS International www.nafems.it www.nafems.org

PAGE 50 AN INTERVIEW WITH NAZARIO BELLATO, SIMULATION MANAGER OF MAGNETI MARELLI POWERTRAIN

TechNet Alliance www.technet-alliance.com RESPONSIBLE DIRECTOR Stefano Odorizzi - newsletter@enginsoft.it PRINTING Grafiche Dal Piaz - Trento The EnginSoft NEWSLETTER is a quarterly magazine published by EnginSoft SpA

Autorizzazione del Tribunale di Trento n° 1353 RS di data 2/4/2008

PAGE 5 COMBUSTION NOISE PREDICTION IN A SMALL DIESEL ENGINE FINALIZED TO THE OPTIMIZATION OF THE FUEL INJECTION STRATEGY

ESTECO EnginSoft Tecnologie per l’Ottimizzazione 34016 TRIESTE Area Science Park • Padriciano 99 Tel. +39 040 3755548 • Fax +39 040 3755549 www.esteco.com

- Newsletter EnginSoft Year 7 n°3

Combustion Noise Prediction in a Small Diesel Engine Finalized to the Optimization of the Fuel Injection Strategy Paper published at SAE Noise and Vibration Conference and Exhibition, May 2009, St. Charles, IL, USA, SAE Paper 200901-2077. The worldwide demand for the engine optimization, in terms of power output, produced pollutants and fuel consumption, is continuously increasing [1]. A growing attention is also being devoted by automobile manufacturers to NVH characteristics of the whole vehicle, and, for this reason, the control of the noise produced by the combustion process in a diesel engine is being considered a very important topic [2,3]. For this reason, the combustion noise reduction is nowadays considered as an additional factor in engine development alongside performance, fuel consumption and emissions. The engine under investigation in the present work is a naturally aspirated, light-duty diesel engine, equipped with a mechanical Fuel Injection System (FIS) and utilized in non-road applications. It is currently under development; however, a new prototype equipped with a common-rail (CR) FIS, going to be installed within small city cars. It is well known that the use of a CR-FIS gives the possibility to respond to the noise emission legislation and market demand through modulation of the injection parameters. The above improvements are however more difficult to obtain on small displacement engines, because of the complexity and cost of the FIS itself [4,5]. In addition, a long development phase is usually required at the test bench in order to define the optimal injection strategies in different engine operating conditions. Based on the above considerations, the main scope of the present work is the development of an optimization procedure that is able to theoretically determine the best injection strategies compatible with high performance and low noise levels, while reducing the development phase and the time-tomarket of the engine. To fulfill the above goal, a multi-objective optimization tool was employed [6-8]. This tool was able to automatically vary the control parameters, and to compare the related performances. The optimization tool however required the development of proper simulation models of the engine. It is currently possible to simulate the physical and chemical processes occurring in the operation of internal combustion engines by using appropriate numerical codes (1D or 3D). 3D simulations can predict, for example, spray behavior, mixture formation, combustion process and toxic emissions [9]. However, they require high computational times, even when high-speed computers are employed. 1D models, on the contrary, are able to gain information on the overall engine behavior [10] and, due to the

reduced computational efforts, are better suited to be employed within an optimization procedure [11]. Concerning the prediction of noise emission, either detailed or simplified models can be utilized. Detailed approaches are usually based on the employment of FEM-BEM codes, which include the in-cylinder pressure cycle as an excitation on the engine structure [12,13]. A number of alternative and simplified procedures are also available in the current literature, based on a proper processing of the computed pressure cycle [14-16]. In this paper, a 1D model was chosen to predict the engine performance and a recent methodology [14], based on the decomposition of the 1D computed pressure signal, was utilized to estimate the combustion radiated noise. The whole activity was developed in four main steps: 1. an experimental campaign was initially carried out to gain information on performance and noise levels on the engine and to acquire the data required to validate the 1D and the combustion noise models; 2. the 1D simulation of the tested engine was combined with the GT-Power® code [17], for estimating the in-cylinder pressure cycles and the overall performances; 3. the methodology reported in [14] was included within a Matlab® routine to estimate the noise level. Some coefficients included in the above correlation were properly tuned to be in agreement with the experimental data; 4. an optimization process [18] was finally carried out with the modeFRONTIER® code to identify an optimal injection strategy of the prototype engine equipped with the CR-FIS. The objectives established were the maximization of the engine performance and the reduction of the noise emission, at a constant load and rotational speed. In the following, a description of each of the above steps is presented and some conclusions are finally drawn concerning, in particular, the trade-off between the two objectives, requiring the selection of a compromise solution. The latter was identified through the employment of a “Multi-Criteria Decision Making” (MCDM) tool, provided by the modeFRONTIER® code. Experimental Analysis In this study, a naturally aspirated, four stroke, two valve, single cylinder diesel engine (505 cm3 displacement) was experimentally investigated. The engine test bed included an electrical dynamometer, the data acquisition and control units, as well as emission and an acoustic measurement equipment. PERFORMANCE TESTS - A programmable electronic control unit (PECU), based on a dSpace processor, was used to manage the

Newsletter EnginSoft Year 7 n°3 -

Fig. 1 - Engine and test-bench laboratory

engine operating conditions. The in-cylinder pressure was detected by a piezoelectric pressure transducer, connected to the AVL IndiModul 621. The air flow rate was also estimated through the measurement of the fuel flow rate and Air/Fuel Ratio. The latter was derived from the analysis of the exhaust gas composition. Engine tests were carried out at full load, in a range of engine speed going from 1400 to 3000 rpm. ACOUSTIC TESTS - In order to measure the radiated noise with reasonable accuracy, the acoustic characteristics of the environment must be known. On an acoustic basis, the ideal environment is a space with no reflecting surfaces and no background noise. In practical terms the ‘best’ environment, for engine applications, is an open-air site with one hard reflecting surface (the ground) and no other obstructions for at least 50m from the noise source and microphone positions. Moreover, the background noise level should be at least 10 dB (preferably 20 dB) below the measured one. Engine noise measurements could be also made in ‘nonanechoic’ test cells with acoustic absorption. The latter set-up was followed for the present investigation, where certain important parameters (reverberation time of the room and

Fig. 2 - In-cylinder pressure and accelerometer signal

background noise) were taken into account, as prescribed by the applied ISO Standards [19]. For this aim, the engine was acoustically isolated from the electrical dynamometer, through properly designed acoustic absorption panels (see the white shields shown in figure 1). The radiated noise of the engine surfaces was measured by a free-field microphone located 1 m away from the engine block, to avoid near field effects, and in a position away from the intake and exhaust systems. The purpose was to minimize the effects of flow noise sources, so that the major contribution considered is the one coming from the engine block. The calibration of the microphone was performed before each test by means of a pistonphone. Simultaneously, an accelerometer was positioned on the engine head in order to correlate the vibration, microphone and pressure signals. The acquisition of the noise and vibration signals were performed at all

Fig. 3 - 1D scheme of the tested engine in GT-Power

investigated engine speeds, at sampling frequencies of 48 kHz. In this way a useful bandwidth wider than 20kHz, free from aliasing effects, was available. In order to synchronize the measurements, a pulse signal supplied by an optical encoder was used as a reference. Figure 2 reports an example of the acquired accelerometer signal, phased with the in-cylinder pressure, during both combustion and motored conditions, at 3000 rpm. In both cases, a strong vibration peak is well evident at the top dead center, due to the piston slap phenomenon. Some reduced spikes can also be identified, in motored conditions, when the incylinder pressure approximately crosses the crankcase pressure (around the atmospheric level). This is probably an indication of the occurrence of some piston movement around its pin, captured as a small vibration signal. The same does not happen when the piston is loaded by combustion pressure. The pressure profiles also show the presence of a strong disturbance at the inlet valve closure (IVC) event. The same spike is indeed absent on the accelerometer signal. Additional experimental data will be presented in the following sections in comparison with the numerical results.

- Newsletter EnginSoft Year 7 n째3

Fig. 4- Comparison on the pressure cycle at 1400 rpm

Fig. 5 - Comparison on the pressure cycle at 2200 rpm

Fig. 6 - Comparison on the pressure cycle at 3000 rpm

One-Dimensional Simulation The well-known GT-Power one-dimensional simulation code was employed to predict the performance of the investigated engine, schematized as shown in figure 3. The 1D code solves the mass, momentum and energy equations in the ducts constituting the intake and exhaust system, while the gas inside the cylinder is treated as a zero-dimensional system. Concerning the modeling of the combustion process, a classical Wiebe equation was utilized to compute the heat release rate in the base engine. Proper values of the combustion process duration during both premixed and diffusive phases were specified. Figures 4-6 show the comparison of the computed and experimental pressure cycles at three different engine speeds. The model was able to correctly reproduce the pressure evolution along both the compression, combustion and expansion phases. Figures 7 and 8 instead report the comparison of the inlet air and brake power through the whole range of investigated engine speeds. Once again a good agreement between experimental and numerical results was reached, especially concerning the air flow rate (fig. 7). The good matching obtained with the experimental data allowed the extension of the 1D analysis to the prototype engine, equipped with the CR-FIS. In this case, however, the employment of a Wiebe equation was no longer permitted. In order to reproduce the effects of the injection parameters modulation, a direct modeling of the spray behavior, fuel-air mixing and combustion process was required. For this reason, inside the optimization process, the GT-Power built-in DIJet (Direct-Injection-Jet) model was utilized, as explained in the next sections. Combustion Noise Estimation It is well known that the combustion noise generation mechanism is a complex phenomenon, including non-stationary and non-linear effects. In-cylinder pressure gradients during combustion process are considered the main excitation forces [20] on the cylinder liner and engine head. Additional contributions come, however, through the excitations exerted on the crankshaft by the inertia forces occurring as a consequence of the rotation and alternative motion of various engine components. While the first contribution depends on various operating parameters (engine speed, load, injection phasing and strategy, etc.), the second term is usually related solely to engine speed. Of course, along the sound propagation pattern from its generation inside the cylinder up to the noise acquisition location (usually at 1 meter from the engine block), the engine structure itself exerts a strong influence [21], in terms of natural vibration frequencies and vibration modes. The structure behavior is often synthesized in terms of a structural attenuation curve [16]. However, a more recent methodology was proposed [14] for the prediction of the overall combustion noise, which includes in the correlation a strict dependency on the engine operating conditions and injection strategy. The above described approach was used in this work. The main idea behind this technique was to decompose the total incylinder pressure signal (ptot) according to three main contributions: compression-expansion (pmot), combustion (pcomb) and resonance (pres) pressures:

Newsletter EnginSoft Year 7 nÂ°3 -

9 [3]

[4] n is the engine speed and nidle the idle rotational speed (fixed at 1000 rpm). The I1 index is a function of the maximum pressure gradient of the combustion contribution, occurring after the pilot (dpmax1/dt)comb and the main injection (dpmax2/dt)comb. The I1 index is also non-dimensionalized over the maximum pressure gradient of the pseudo-motored pressure (dpmax/dt)comb. In the case of a single-shot injection, a unique term is of course present in the eq. (3) numerator. The I2 index takes into account 2 the acoustic energies (â&#x2C6;Ťp dt) associated with resonance and motored pressure signals. An additional index In is finally defined in [14], accounting for mechanical noise contribution, related, as stated, to the sole engine speed:

Fig. 7 - Computed and experimental air flow rate

[5] Basing on the above definitions, the Overall Noise (ON) can be finally computed as: [6] Fig. 8 - Computed and experimental mechanical power

(1) The first contribution (also referred as pseudo-motored signal) is only related to volume variation, and was used as a reference signal. It was determined by a direct in-cylinder pressure acquisition during a fuel switch-off operation. The sum of combustion and resonance pressures is also referred to as excess pressure (pexcess) and was determined by the difference between the total and the pseudo-motored pressures: (2) This contribution is of course related to both fuel burning and to high-frequency resonant pressure fluctuations, induced by the pressure gradients during the combustion process [15]. To separate the above two terms, a high pass-band filter of the total pressure FFT was accomplished, as shown in figure 9. Above a proper cut-off frequency (about 4.5kHz) in fact, the pressure amplitudes (expressed in dB) tend to increase, thus indicating the occurrence of a resonance phenomenon [14]. An IFFT procedure was applied to the high pass-band filtered signal allowing to finally reconstruct the three contributions in eq. (1). They were compared together in figure 10. Despite the presence of the previously discussed high-frequency amplitudes, the resonant pressure was significantly lower than other contributions. Nevertheless, it may still exert a non-negligible effect on the overall noise. The three decomposed pressures were utilized to compute two characteristic indices I1 and I2 defined as:

Ci being proper tuning constants, depending on the engine architecture and size. Following the relations (3-6), a Matlab routine was developed to properly process the in-cylinder pressure cycle and compute the various noise indices and the overall noise. This routine was applied to both the experimental and GT-Power computed pressure cycles. In the experimental analysis, the motored signal was directly acquired by means of a sudden fuel-switch off maneuver. Similarly, to compute the motored pressure, the fuel injection was completely disabled in the numerical analyses. Figure 11 compares the ON computed levels with the ones experimentally measured at the test-bench. Some adjustment of the Ci constants was required with respect to the values proposed in [14]. The agreement obtained through the

Fig. 9 - Total pressure spectrum (in dB), at 1400 rpm

- Newsletter EnginSoft Year 7 n°3

Fig. 12 - Parametric Injection strategy

Fig. 10 - Decomposition of the total pressure in motored, combustion and resonance contributions, at 1400 rpm

tuning constants, acting as correction terms in the numerous correlations included in the model, were also assigned. Due to the absence of experimental data on the prototype engine, the tuning constants were identified in order to match the experimental pressure cycles measured on a similar engine, equipped with the same FIS, as reported in [26]. A pilot-plus-main strategy was specified for the CR-FIS at 3000 rpm, at full load conditions. Figure 12 shows the way the fuel injection strategy was schematized, based on the definition of three degrees of freedom, namely the start of pilot injection (SOIP), the dwell time between the pilot and main (DWELL) and the duration of the main injection (MDUR). In order to maintain the same injected fuel mass (22 mg), a constant overall duration (PDUR+MDUR=18.8°) was specified. Constant values of needlelift ramp-up (1.9°) and ramp-down (1.6°) were also assigned, depending on its dynamics. This allowed the computation of the crank angle position in all main points of the injection strategy, as a function of the 3 parameters SOIP, DWELL and MDUR:

Fig. 11 - Comparisons on the overall noise

employment of the experimental pressure cycles is satisfactory at each engine speed. A maximum absolute error of about 1.3 dB was found at a medium engine speed. As a consequence of the inaccuracies included in the engine simulation, the agreement at high speed slightly worsens when the predicted pressures are considered. Moreover, the employed zero dimensional model is unable to take into account the high frequency contributions of the computed pressure that are strictly related to the resonance phenomenon. Nevertheless, the satisfactory agreement shown authorizes the employment of the recalled methodology within the optimization procedure described in the next paragraph. Optimization Procedure The previously described 1D and combustion noise models constituted the basis for the optimization of the injection strategy of the prototype CR engine. However, as already stated, a more advanced combustion model was in this case required. The latter (DIJet model) follows the multi-zone Hiroyasu approach [22-25] and is able to describe the fuel injection, break-up, airentrainment, evaporation and combustion processes. Details on the employed model can be found in [18,26]. The injector characteristics (in terms of holes number and diameter) injection strategy profile and timing represents the input data. The value of a number of

[7]

The logical development of the optimization problem within the modeFRONTIER® environment is explained in figure 13. A number of Transfer Variables objects - together with the three Input Variables SOIP, DWELL and MDUR - were defined based on relations (7), and were written inside the GT-Power input file (LD500.dat). For each set of the above parameters, a proper script procedure runs the GT-Power code and extracts the incylinder pressure (pressure3000rpm), which is required by the

Fig. 13 - Logic scheme of the optimization process within modeFRONTIER®

Newsletter EnginSoft Year 7 n°3 -

Matlab routine computing the overall combustion noise. A multiobjective optimization was so defined to contemporarily search the maximum Indicated Mean Effective Pressure (IMEP) and the minimum of the overall noise. To solve the above problem, the MOGA-II algorithm was utilized, belonging to the category of genetic algorithms [27] and employing a range adaptation technique to carry out time-consuming evaluations. Figure 14 displays a scatter chart of the optimization procedure highlighting the Pareto Frontier occurring when the IMEP is plotted against the ON (a solution is said to be Pareto optimal if there is no other solution which is better in all objectives). A trade-off between the two objectives has clearly occurred. The position of the Base Engine, also shown in the same figure, was far away from the Pareto Frontier, thus indicating the possibility to attain a better level of both objectives. In order to select a single solution among the ones located on the Pareto frontiers, the “Multi Criteria Decision Making” tool (MCDM) provided in modeFRONTIER® was employed. It allows the definition of preferences expressed by the user through direct specification of attributes of importance (weights) among the various objectives. Depending on the above relations, the MCDM tool was able to classify all the Pareto Frontier solutions with a decreasing rank value.

the ones obtained in the CR engine in correspondence with the previously shown injection strategies. The retarded combustion process occurring in both the optimized cases contemporarily determined a lower pressure peak and a reduced pressure gradient (lower noise). Contemporary, a higher pressure level was found during the expansion stroke, which was mainly responsible of the small IMEP increase in the MCDM #1 solution.

Fig. 15 - Optimal injection strategies

Conclusion The paper described a methodology for the identification of optimal injection strategies of a CR light-duty diesel engine. The above objective was reached through the development of proper models for the prediction of engine performance and combustion noise. Both models were validated in reference to experimental data collected on a base, mechanical injection engine. Then a multi-objective optimization process was carried out with the aim of characterizing the trade-off between IMEP and ON on the

Fig. 14 - Scatter chart of the optimization process highlighting the Pareto Front

Two different specifications were attempted: in the first case, the IMEP was considered the most important objective and assumed a weight two times higher than ON. Under this hypothesis, the point gaining the highest rank was the “MCDM Solution #1”, whose position is highlighted in figure 14. In this way both a ON reduction and a small IMEP increase was obtained with respect to the Base Engine. As an alternative, the same weight was specified for both IMEP and ON. In this other case, the solution #2 was selected and a small IMEP reduction was accepted to obtain a more relevant ON drop. The position of the two solutions puts into evidence that the MCDM procedure effectively realized a compromise between the conflicting needs, quantified by the attributes of importance previously described. In addition, this procedure defined a standardized method for the selection of the “global” optimum. Figure 15 and table 1 compare the optimal injection strategies selected by the MCDM tool and synthesize the related outputs. High IMEP required an advanced start of both pilot and main injections, with a reduced dwell time. As expected, a lower ON is indeed found with a delayed SOI and a higher dwell time. Figure 16 finally compares the base engine pressure cycle with

Table 1 – Parameters of the injection strategies and related performance and overall noise at 3000 rpm

Fig. 16 - Comparisons of the pressure cycles obtained through the optimization process

- Newsletter EnginSoft Year 7 n°3

prototype CR engine. A standardized procedure was also defined in order to select a unique solution, on the base of the user preferences and weight of importance of the single objective. The optimization procedure was able to capture the expected effects of the injection strategy parameters on the overall performance and radiated noise. It represents a very useful tool to reduce the huge experimental activity usually required to develop the control logic of the FIS. The methodology can be easily extended to multiple operating conditions and can include additional constraints related, for example, to noxious species emission predicted through 3D-CFD analyses. Acknowledgements The authors would like to express their thanks to Dr. Gerardo Valentino for supporting the experimental activity carried out in the present study. References [1]

M. Stotz, J. Schommers, F. Duvinage, A. Petrs, S. Ellwanger, K. Koynagi, H. Gildein, “Potential of Common-Rail Injection System for Passenger Car Di Diesel Engines”, SAE Paper 2000-01-0944, 2000 [2] Zavala P.A.G., Pinto M.G, Pavanello R., Vaqueiro J., “Comprehensive Combustion Noise Optimization”, Sae Paper 2001-01-1509, 2001 [3] F. Mallamo, M. Badami, F. Millo, “Effect of Compression Ratio and Injection Pressure on Emissions and Fuel Consumption of a Small Displacement Common Rail Diesel Engine”, SAE Paper 2005-010379, 2005 [4] L. Allocca, S. Alfuso, A. Montanaro, G. Valentino, M. Lolli, “Innovative lift direct command to inner hydraulic circuit injector comparison for diesel engines”, ICEF2006-1518: 2006 Fall Conference of the ASME Internal Combustion Engine Division, November 5-8, 2006, Sacramento, USA [5] S. Alfuso, L. Allocca, G. Caputo, F.E. Corcione, A. Montanaro, G. Valentino, M. Lolli: Spray Analysis of an Innovative Direct Command Solenoid Injector for Common Rail Light Duty Diesel Engines, ICLASS-2006, Aug.27-Sept.1, 2006, Kyoto, Japan [6] Papalambros, P.V., and Wilde, D.J., “Principles of Optimal Design Modeling and Computation”, Cambribde University Press, Cambridge, 2000 [7] Assanis D.N., Polishak M.,“Valve event optimization in a sparkignition engine”, Journal of Engineering for Gas Turbines and Power; Vol/Issue: 112:3, 1990 [8] Stephenson P.W., “Multi-Objective Optimization of a Charge Air Cooler using modeFRONTIER® and Computational Fluid Dynamics”, SAE Paper 2008-01-0886, 2008 [9] C. Beatrice, P. Belardini, C. Bertoli, M. G. Lisbona, G. M. Rossi Sebastiano: Combustion process management in common-rail DI diesel engines by multiple injection, SAE Paper 2001-24-0007 [10] D. Siano, F. E. Corcione, F. Bozza, A. Gimelli, S. Manelli “Characterization of the Noise Emitted by a Single Cylinder Diesel Engine: Experimental Activities and 1D Simulation”, SAE Noise and Vibration 2005, Traverse City – 05NVC-133 [11] Siano D., Bozza F., Costa M., “Optimal Design of a Two-Stroke Diesel Engine for Aeronautical Applications Concerning both Thermofluidynamic and Acoustic Issues”, IMECE2008-68713, ASME IMECE Congress, November 2008, Boston.

[12] Bozza F., Costa M., Siano D., “Design Issues Concerning Thermofluidynamic and Acoustic Aspects in a Diesel Engine Suitable for Aeronautical Applications”, to be published in the International Journal of Vehicle Design, 2009. [13] Zienkiewicz, O. C., and Taylor, R. L., 2000, “The Finite Element Method”, Butterworth-Heinemann, ISBN 0750650494 [14] Torregrosa A.J., Broatch A., Martın J., Monelletta L., “Combustion noise level assessment in direct injection Diesel engines by means of in-cylinder pressure components”, Meas. Sci. Technol., 18 2131-2142 doi:10.1088/0957-0233/18/7/045, 2007 [15] F Payri, A Broatch, B Tormos and V Marant, “New methodology for in-cylinder pressure analysis in direct injection diesel engines— application to combustion noise”, Meas. Sci. Technol. 16 (2005) 540–547 doi:10.1088/0957-0233/16/2/029 [16] Corcione F. E., Siano D., Vaglieco B. M., Corcione G. E., Lavorgna M., Viscardi M., Iadevaia M. and Lecce L., “Analysis and Control of Noise Emissions of a Small Single Cylinder D.I. Diesel Engine”, SAE Paper 2003-01-1459, 2003 [17] GT-Power User’s Manual, 2005. [18] Bozza F., Siano D., Valentino G., “Integrated Numerical and Experimental Methodologies for Performance Optimization and Noise Reduction of a Light Duty Diesel Engine”, EnginSoft CAE Users’ Meeting 2007, Stezzano (BG), October 2007. [19] ISO 9614/1, ”Acoustics-Determination of sound power levels of noise sources using sound intensity-Part 1:Measurement at discrete points”, 1993. [20] Osawa H, Nakada T, “Pseudo cylinder pressure excitation for analysing the noise characteristics of the engine structure”, JSAE Rev. 20 67–72, 1999 [21] Corcione F.E., Siano D., Iadevaia M., Viscardi M., Corcione G.E., “Correlation between the acoustic intensity measurements with and without an electronically fuel injection system for a small single cylinder diesel engine”, Euronoise 2003, Naples, 334 [22] Hiroyasu H., Arai M., Tabata M. “Empirical Equations for the Sauter Mean Diameter of a Diesel Spray”, SAE Paper 890464. [23] Hiroyasu H., Arai M., “Fuel Spray Penetration and Spray Angle of Diesel Engines”, Trans. Of JSAE, Vol. 21, pp. 5-11, 1980 [24] Hiroyasu H., Kadota T., Arai M., “Development and Use of a Spray Combustion Modeling to Predict Diesel Engine Efficiency and Pollutant Emissions”, Bulletin of JSME, Vol. 26, N: 214, pp., 569575, 1983 [25] Young D., Assanis D.N., “Multi-Zone DI Diesel Spray Combustion Model for Cycle Simulation Studies of Engine Performance and Emissions”, SAE paper 2001-01-1246, 2001 [26] Siano D., Valentino G., Corcione F., Bozza F., Arnone L., Manelli S., “Experimental and Numerical Analyses of Performance and Noise Emission of a Common Rail Light Duty D.I. Diesel Engine”, SAE Paper 2007-24-0017, ICE 2007 Congress, Capri, Settembre 2007. [27] Sasaki, D., 2005, “ARMOGA, An efficient Multi-Objective Genetic Algorithm”, Technical Report, January 2005

Daniela Siano (Researcher) Istituto Motori CNR – Napoli - d.siano@im.cnr.it Fabio Bozza (Full Professor) (DIME). Università di Napoli “Federico II” - fabio.bozza@unina.it

Newsletter EnginSoft Year 7 n属3 -

Aerodynamic and Acoustic Optimization of Radial Fans In this work the multiple objective optimization of radial fans with respect to aerodynamic efficiency and noise generation has been performed. This has been achieved by coupling together an LSTM In-house Excel-VBA Impeller Design Tool (EVIDenT), the CAD program ProEngineer, the grid generator ANSYS ICEM and the CFD solver ANSYS CFX, as well as the LSTM In-house Acoustic Code SPySI (Sound Prediction by Surface Integration) within the optimization software modeFRONTIER速. From a technical point of view, the coupling of the different tools was one of the main challenges solved with modeFRONTIER速. The input variables for the optimization were the shape parameter, i.e. the wrap angle of the impeller and its number of blades. All simulations have been performed in a 2D scenario in order to capture primary fundamental aspects relevant to the impeller design. As a result of the optimization, the efficiency of the radial fans has been improved as well as the noise level reduced substantially. A set of non-dominated solutions (Pareto solutions) have been obtained which can be used according to the specific user requirements. The results show that the integration of acoustics and transient flow simulations within a multiple objective fully automated optimization process is feasible. Having established the fully integrated and automated process an extension also to 3D computations can be readily performed.

Fig. 1 - In-house Excel-VBA Impeller Design Tool (EVIDenT)

exported to the grid generator ANSYS ICEM where another script generates also automatically the grids, Figure 2. The grid is exported to CFX, the flow domain is automatically set up, Figure 4, and the solver starts to run to compute the CFD solution. The results of the CFD simulation before and after the optimization are shown in the stream line plots of Figure 3. One can clearly see that in the optimized design the flow velocities in the impeller where reduced keeping, however, the same pressure and flow rate, as well as reducing

Introduction The aim of this work is to analyze the possibility of optimization with modeFRONTIER速 [1] by integration of In-house and commercial tools in order to automate turbomachinery design with respect to efficiency and noise generation. The starting step in the optimization process is the design of impellers with the In-house ExcelVBA Impeller Design Tool (EVIDenT), which delivers high performance starting blade shapes Fig. 3 - Radial impeller for the fully integrated optimization process. The Fig. 2 - Wrap angle main optimization parameters in this work were the wrap angle, Figure 1, and the number of blades. Many other parameters can be included, e.g. the blade inlet and outlet angles, shroud shape, but the scope of this work was to establish the optimization work flow. Even so, with those two parameters already very good results were achieved. These geometries are then exported into the CAD program ProEngineer where the impeller, e.g. Figure 1, and the corresponding flow domains are generated. These geometries are then fully automatically Fig. 4 - Grid and fluid domain solver setup in ANSYS CFX Pre.

- Newsletter EnginSoft Year 7 n°3

modeFRONTIER® offers all features needed in order to integrate the automation processes of the different programs and to perform powerful optimizations.

Fig. 5 - CFD analysis of non optimized (left) and optimized impeller (right)

Fig. 6 - Sound pressure level of non optimized (left) and optimized (right) impeller

Multi-objective optimization In this case, as there were 2 objectives, a multiobjective algorithm had to be chosen. Therefore the MOGA II algorithm was selected with 5 generations and combined with a DOE Sobol of 8 designs. The work flow in modeFRONTIER® is shown in Figure 8. Here the different tools and scripts used are integrated [6] using the modeFRONTIER® workflow connectors. The input variable nodes are used for the optimization inputs (e.g. number of blades and shape parameter). These are then connected to the first node (1), the In-house Excel-VBA Impeller Design Tool (EVIDenT) through the scheduler (8), shown in Figure 8. This design tool EVIDenT generates the information about the number of blades and the data for the shape of the blades and writes them out as text files. These files are then passed to the python node (2), which passes the variables in the script to the CAD program ProEngineer (3). ProEngineer then creates the flow domain for the blades as a parasolid file, which is then transferred to the ICEM node (4). To this node (4) also the ICEM script and the parasolid files for the other parts of the geometry are transferred. Here in node (4) then the mesh is created and transferred as a CFX5 file to the CFX node (5). In this node (5) some additional CFX5 files and scripts for pre and post processing arrive also from the transfer and support file nodes. Node (5) runs then simulation, calculates the efficiency and writes out the result as a text file. From the CFX node (5) CSV (Comma separated Value) files are transferred to the next tool in node (5), which consist of the acoustic In-house tool SPySI. It runs the SPySI tool and writes out the results, i.e. the sound pressure level, as text file. These files are then transferred to the output nodes (6), which are then finally transferred to design

Fig. 7 - Prototypes

substantially the sound pressure level. In Figure 4 three prototypes are shown. The modeFRONTIER® optimization environment As described above, the work flow was carried out by integrating and automating with scripts a set of commercial (ProEngineer [2], ANSYS ICEM [3] and ANSYS CFX [4]) and In-house tools (Python based acoustic tool SPySI [5] and Inhouse Microsoft Excel-VBA Impeller Design Tool EVIDenT [6]). But how to integrate all these commercial and In-house tools in order to perform a multi-objective optimization? The answer was to use modeFRONTIER®. The multiobjective optimization environment tool

Fig. 8 - Work flow of the optimization process

Newsletter EnginSoft Year 7 n°3 -

[5] Scheit, C., Karic, B., Delgado, A., Epple, P. and Becker, S. (2009) Experimental and Computational Study of Radial Impellers With Respect to Efficiency and Noise Production. Conference on Modeling Fluid Flow (CMFF’09) The 14th International Conference on Fluid Flow Technologies Budapest, Hungary, September 9-12. [6] Masood, Rao M. A. (2010) Principle Study of Optimization of Radial Fans with respect to Aerodynamics and Aeroacoustics, Master Thesis, LSTM, University of Erlangen-Nürnberg.

Fig. 9 - The Pareto Front

objective nodes (7). These nodes make sure that efficiency is maximized and the noise level is minimized. Based on this information, the scheduler node (8) analyzes and generates a new design. Results obtained with modeFRONTIER® In this work a total of 96 possible designs were run, out of which 35 designs were evaluated. From those the final three optimized designs were selected and compared with the three original starting designs. A set of non-dominated solutions, as shown in Figure 9, have been found which showed substantial improvements in the efficiency and reduction in the noise level. In the case of a multi-objective optimization, there is no single best design but rather a set of non-dominated designs. The best design with respect to efficiency has an increase of 35%, while the best design with respect to noise level has a reduction of 3 dB as compared to the original design, which means a reduction of 50% in the sound power level Conclusions This work has shown, Figure 8, how it is possible to integrate and automate different codes, i.e. the In-house Excel EVIDenT code, ProEngineer, ANSYS ICEM, ANSYS CFX and the In-house Acoustic tool SPySI in modeFRONTIER® and finally how to establish and carry out an multi-objective optimization in this environment The efficiency of radial fans has been improved as well as the noise level reduced noticeably. A set of non-dominated solutions (Pareto solutions) have been obtained which can be used according to the user needs. References [1] modeFRONTIER®: http://www.esteco.com/products.jsp [2] Pro/Engineer Wildfire: http://ptc.com/products/proengineer/ [3] ANSYS ICEM CFD: http://www.ansys.com/products/icemcfd.asp [4] ANSYS CFX: http://www.ansys.com/products/fluid-dynamics/cfx/

Institute of Fluid Mechanics - Technical Faculty Friedrich-Alexander - University Erlangen-Nürnberg MSc. Engr. Rao Muhammad Atif Masood M.Sc Christoph Scheit Dr.-Ing. Philipp Epple Prof. Dr. A. Delgado - Professor and Head

The Institute of Fluid Mechanics (Lehrstuhl für Strömungsmechanik - LSTM) of the Friedrich-AlexanderUniversität Erlangen-Nürnberg has 8 departments working on a large variety of research topics: Aerodynamics, Turbulence, Aeroacoustics, chemical reacting flows, fluid flow process automatization, bio and medical technology, numerical flow simulation, process fluid dynamics and Turbomachinery, instationary fluid mechanics, Engineering of Advanced Materials and thermo-fluid-dynamics of bio technological processes. There are about 70 researchers working at the LSTM. The LSTM has many years of experience in the design, numerical computation and aerodynamic and acoustic optimization of turbomachines of all kinds – axial, diagonal and radial. The aerodynamic design and acoustic computation are done with Inhouse-codes as well as with commercial tools. www.lstm.uni-erlangen.de

Der Lehrstuhl für Strömungsmechanik (LSTM) der FriedrichAlexander-Universität Erlangen-Nürnberg setzt sich aus 8 Forschungsbereichen mit einer sehr breit angelegten thematischen Ausrichtung zusammen: Aerodynamik, Turbulenz und Aeroakustik, Strömungen mit chemischen Reaktionen, Prozessautomatisierung von Strömungen in Bio- und Medizintechnik, Numerische Strömungsmechanik, Prozessfluiddynamik und Strömungsmaschinen, Instationäre Strömungsmechanik, Engineering of Advanced Materials und Thermofluiddynamik biotechnischer Prozesse. Insgesamt arbeiten und forschen hier ca. 70 Mitarbeiter und Mitarbeiterinnen. Der LSTM besitzt langjährige Erfahrung in der Auslegung, numerischen Berechnung und strömungsmechanischen und akustische Optimierung von Turbomaschinen aller Bauformen (radial, diagonal und axial). Die strömungsmechanische Auslegung und akustische Berechnung erfolgen sowohl mit Inhouse-Codes wie auch über kommerziellen Tools.

- Newsletter EnginSoft Year 7 nÂ°3

Electromagnetic Interference: an Advanced FEM Calculation Approach Accurate analysis of the electromagnetic interference of complex systems is a fundamental requirement when two or more structures share the same working environment. The analytical and the empirical approaches are not

an extension of the methods available in the literature. In order to obtain fast and accurate results, the equivalent circuit has been analyzed by means of parametric software; ANSYS APDL and ANSYS Maxwell have been used to solve the finite element problem. Formulation By referring to the Figure 1, let us consider the pipeline (yellow line) and the three high voltage lines. In order to completely define the electromagnetic interference phenomena, both inductive and conductive coupling have been considered.

Fig. 1 - Geometry of the high voltage line â&#x20AC;&#x201C; pipeline system

The inductive coupling represents the main cause of electromagnetic interference between the pipeline and the high voltage line. This effect is due to the currents induced on the pipeline by a time-varying magnetic field, generated by a sinusoidal current flowing on the high voltage line. For accurate evaluation of the induced coupling, a typical approach consists of estimating the

suitable to this aim; in particular, in the former case the advantages of a rigorous method are offset by the limited availability of known solutions for few simple cases. In the latter case, post factum mitigation actions are completely based on operator experience, which is generally not sufficient to solve these problems due to the complexity of modern systems. The main drawback of the wide variety of empirical/numerical methodologies consists in the need for formulating a simplified hypothesis based on the electromagnetic material properties. Moreover extensive experience is needed in order to define the applicability field of the involved mathematical relationships. In recent years the improvements of computational capability and the memory availability of Fig. 2 - two-dimensional model of the pipeline-transmission line system: complete computational computing resources, allow the efficient domain (a), particular of the pipeline (b) and symmetric computational domain (c) solution of problems characterized by several unknowns. This work presents a calculation approach to the study of electromagnetic interference generated by high voltage lines and metallic pipelines, buried into the soil. In particular, a procedure for calculating the inductive and conductive coupling has been developed. The calculation of the equivalent generators of induced electromotive force has been performed by means of a finite element model. This approach produces a generalization and Fig. 3 - 2D computational domain mesh (a) and particular of the pipeline (b).

Newsletter EnginSoft Year 7 n°3 -

Fig. 4 - Current density on the pipeline section (a) and the flux lines (b)

Fig. 5 - elementary cell of the equivalent circuit

currents induced on the pipeline/high-voltage line system during both the working normal condition and in single phase breakdown case. The conductive coupling occurs in the single phase breakdown case in proximity to the inductive installations provided with good grounding, such as electrical substations and electricity network trellis. Under these conditions, the breakdown current flows into the soil increasing the electric potential in the local domain surrounding the pipeline. The evaluation of the generated voltage allows establishment of safety conditions for the operators, according to the limits imposed by the normative. Electromagnetic analysis has been performed with the following steps: 1. Generation of the 2D FEM model to get the equivalent generators of electromotive induced force. 2. Building and solution of the transmission line model using equivalent circuits. 3. Generation of the 3D FEM model to analyze the pipeline/high-voltage line system for the conductive disturb. 2D FEM model for analysis of inductive coupling In order to evaluate the inductive coupling between the pipeline and the high voltage line, the two-dimensional model of a transverse section of the geometry, represented in Figure 1, has been taken into account. In particular, it is composed by (Figure 2-a): • The soil; • The air; • The inducting line; • The pipeline.

By observing the symmetry of the problem, the computational domain has been reduced to half of the real one (Figure 2-c). The transmission line has been modeled by means of a current flowing orthogonally to the model plane, while the pipeline has been modeled by considering a metallic circular ring. The computational domain has been properly discretized by means of a triangular element mesh (Figure 3-a) with quadratic shape function. Moreover, in order to correctly evaluate the skin effect on the pipeline, a finer mesh has been obtained on the external pipeline surface (Figure 3-b). The flux lines of magnetic field have been imposed parallel in proximity of the boundary of the computational domain. Figure 4 shows some outputs of the electromagnetic analysis: the current density on the pipeline steel (a), and the flux lines on the computational domain (b). The plot, shown in Figure 4-a, highlights the skin effect on the pipeline and justifies the mesh model done on the external pipeline surface. The plot Figure 4-b depicts how the current induced on the pipeline affects the magnetic field. Mathematical model In order to evaluate the current induced on the pipeline, the characteristic impedance of the system has been evaluated in ANSYS APDL (ANSYS Parametric Design Language). This procedure allows the user to obtain an equivalent circuit model of the pipeline as a series of elementary equally spaced cells. Each cell (Figure 5) is constituted by: • An electromotive force that takes into account the effects of the inducting system on the induced one; • A longitudinal impedance that represents the pipeline impedance;

Fig. 6 - 3D model of the pipeline-transmission line system for the conductive problem

- Newsletter EnginSoft Year 7 n°3

the position along the pipeline and due to the three lines effect, has been shown in Figure 9. By referring to the previous figure, the voltage induced by each transmission line has been shown together with their quadratic mean square value. Fig. 7 - mesh of the computational domain (a) and of the dielectric cover (b)

Fig. 8 - induced current (a) and zero voltage condition (b)

• A trasversal impedance that accounts for the impedance between the pipeline and the transmission line. 3D FEM model for the analysis of the conductive coupling The analysis of the conductive coupling has been carried out by using a three-dimensional model (Figure 6) where a part of 30 meters of pipeline has been modeled buried into the soil, by performing a parameterization of both geometric dimensions and the material properties according to the inductive case. The model is constituted by the soil, the metallic pipeline and an insulating layer placed around the pipeline itself. The three-dimensional domain has been discretized by means of a tetrahedral element mesh with quadratic shape function (Figure 7-a). A finer mesh (Figure 7-b) has been obtained on the dielectric cover. In order to define the boundary conditions, a zero potential has been imposed on the lower part of the calculation domain (Figure 8-b), while the excitation has been modeled as a current orthogonally directed with respect to the surface soil (Figure 8-a). Results Inductive and conductive interferences have been analyzed on a pipeline placed in proximity to three highvoltage lines. The results relative to the inductive coupling obtained in normal working conditions have been shown in this article. By referring to the model shown in Figure 2-c, the inductive coupling has been evaluated; the induced voltage, calculated as function of

Conclusions A FE-based procedure for the analysis of electromagnetic interference, generated by a highvoltage line and a pipeline, has been presented in this article. In order to completely describe the interference system, both the inductive and the conductive coupling have been taken into account. The two-dimensional and the three-dimensional FEM models have been generated and analyzed, with proper boundary conditions, by using ANSYS Maxwell. The equivalent generators of electromotive force have been obtained from the FEM 2D model and given as input to the transmission line equivalent circuit. The 3D FEM model supplies the electric potential acting on the pipeline in case of single phase fault. The aim of the study is the generalization of semi-analytic approaches to electromagnetic interference problems: FEM models allow the user to obtain results even out of the validity domain of Carson Clem formulas.

For more information Alice Pellegrini - EnginSoft info@enginsoft.it

Dott. Giovanni Falcitelli - Enginsoft S.p.A. PhD Ing. Alice Pellegrini – Enginsoft S.p.A. Ing. Emiliano D’Alessandro – Enginsoft S.p.A.

Fig. 9 - Induced voltage along the pipeline

Newsletter EnginSoft Year 7 nÂ°3 -

Reliability Based Structural Optimization of an Aircraft Wing Today, in aircraft industry, there is a great competition to release new aircraft designs which are faster, more efficient, more economical, more reliable and even quieter than the former ones - both in military and civil applications. The challenging multi-disciplinary task of aircraft design can be realized by incorporation of numerical optimization techniques in the industrial design process. However, there are always uncertainties related to design parameters,

Fig. 1 - Computational Model of the Wing Structure

modelling, manufacturing process, operating conditions and human factors when designing a new aircraft. Conventional deterministic design and optimization processes may yield unreliable designs since the uncertainties are not accounted for in the design process. However, reliability-based design optimization (RBDO) is a methodology which can include probabilistic design criteria involving aleatory uncertainties in the optimization process. In this work, we propose an implementation of an RBDO algorithm into a structural optimization framework composed in the modeFRONTIERÂŽ optimization tool. Reliability analysis and optimization are two essential components of RBDO: (1) Reliability Analysis focuses on analyzing the probabilistic constraints to ensure the reliability levels are satisfied; (2) Optimization is seeking for the optimal performance subject to the probabilistic constraints. A simple aircraft wing which has a NACA0012 airfoil profile is modeled parametrically in Catia V5-R16. The wing's three dimensional geometric model consists of 90 skin panels, 10 ribs and 4 spars while some of the skin panels are stiffened by stringers along the wing span. The wing has a rectangular planform with 6m semi-span and 1.6m chord length. The finite element model of the wing is prepared for Abaqus 6.7.1 and is composed of linear shell and beam elements. The model is shown in Figure 1, and consists of

17,070 linear quadrilateral elements of shell type, 1264 linear line elements of beam type, for a total element number of 18,334 and 16,024 nodes, thus 96,144 degrees of freedom. In all members of the structure, aluminium is employed with Young's modulus E = 70000MPa, Poisson ratio v = 0.33, density Ď = 2700kg/m3, yield strength đ?&#x153;&#x17D;yield = 400MPa. As a cantilevered boundary condition, all of the degrees of freedom at the root of the wing are set to zero. The aerodynamic load that will be applied to the wing is supplied from a computational fluid dynamics (CFD) analysis performed for the initial design. An Euler inviscid flow analysis by using Fluent commercial software was performed for Mach = 0.3 at sea level. A structural optimization problem with two random variables which are Young's Modulus E and yield strength đ?&#x153;&#x17D;yield of the material will be solved. Thus, the allowable stress which is đ?&#x153;&#x17D;allowable = đ?&#x153;&#x17D;yield / 1.5 is calculated with a safety factor of 1.5 which counts for epistemic uncertainties. The constraints concerning stress (g1), displacement (g2) in those equations become probabilistic constraints due to their dependencies on the random variables vector X = [E đ?&#x153;&#x17D;allowable]. E and đ?&#x153;&#x17D;allowable are modeled with normal distributions using N(70000, 350) MPa and N(270, 20) MPa. On the other hand, there are two reliability subroutines, which one of them corresponds to reliability index (Î˛s) for stress constraint and the other one corresponds to reliability index (Î˛d) for displacement constraint, in main reliability code in the optimization process. The optimization variables are chosen as the thicknesses of skin panels, ribs, spars, stringers and location of first four ribs and two spars (Nikbay et al. [1]). Thus, the optimization problem can be formulated as;

In terms of reliability index, the probabilistic constraints of the above optimization problem can be expressed as;

- Newsletter EnginSoft Year 7 n°3

The nodes shown in Figure 2 can be explained briefly as follows. The Logic Nodes are the nodes used to define workflow actions like logic start, logic end and logic failed. The optimization algorithm and the way how the starting points for the process are distributed in the design space are indicated by this node. The Goal Nodes are the nodes used to define the user’s strategy for the optimization problem. The Variable Nodes are the nodes used to define the data definition for the optimization problem. On the other hand, the File Nodes identify text files that are sent to an application node or from which to extract values to be assigned to output variables. Dos Batch Node and Calculator Node stores and configures scripts in Windows ® DOS syntax and stores and configures scripts in JavaScript syntax, respectively. In this study Dos Batch Nodes are used to call ABAQUS and

Fig. 2 - modeFRONTIER® Workflow Nodes

Here, ( ) and ( ) are the target reliability indexes for stress and displacement constraints and chosen as to be 3.0 for a reliability of 0.99865 or a probability of failure of 0.00135. The actual reliability index values for the current design at each optimization iteration is calculated and passed to the outer optimization loop as a constraint.

Fig. 4 - Paretos of Aircraft Wing Structural Optimization With RBDO

Figures 2 and 3 show the workflow of the problem and its nodes constructed by using the modeFRONTIER® optimization tool.

MATLAB to run the processes whenever they are needed. Finally the CATIA V5 Node is used to wrap a CATIA V5 document, transferring data from and to it, and executing macro files on it. A CATIA document is either a CATPart, a CATProduct or a CATAnalysis. The optimization process is run with 52 design of experiments (DOE) with "Sobol sequence" where 300 maximum number of iterations per sub-iterations for the NLPQLP are defined. Finally, a total number of 140 designs are generated for the optimization problem. Solution of the problem took about 21 hours on a workstation with Intel(R) Core(TM)2 Quad CPU 6700@2.40 GHz processor, with 2 GB of RAM on Microsoft Windows XP operating system.

Fig. 3 - Workflow of the Reliability Based Structural Optimization Problem: Generic Wing

Finally, 40 designs were found to be feasible that satisfy the constraint conditions. Furthermore, there are 16 error designs. As a result, 4 designs are found in the pareto front set for this optimization problem. These Paretos are demonstrated in Figure 4. The design which corresponds to Pareto 2 in Figure 4 is chosen as

Newsletter EnginSoft Year 7 n°3 -

optimum design due to its minimum mass value while still satisfying the reliability index target constraint. Conclusion In this work, a reliability based design optimization methodology is proposed by implementation of a homemade RBDO code based on Reliability Index Approach (RIA) into a structural optimization framework composed of highfidelity commercial software. The presented work shows good results when compared to the deterministic optimization results of the same problem studied formerly (Nikbay et al. [1]). About The Istanbul Technical University (ITU) ITU was established in 1773, during the time of the Ottoman Sultan Mustafa III. With its original name "Muhendishane-i Bahr-i Humayun", The Royal School of Naval Engineering, its responsibility was to educate chart masters and ship builders. In 1795, the "Muhendishane-i Berr-i Humayun", The Royal School of Military Engineering, was established to educate the technical staff in the army. In 1847, education in the field of architecture was also introduced. Established in 1883, the School of Civil Engineering assumed the name "Engineering Academy", with the aim of teaching essentials skills needed in planning and implementing the country's new infrastructure projects. Gaining university status in 1928, the Engineering Academy continued to provide education in the fields of engineering and architecture until it was incorporated into ITU in 1944. Finally, in 1946, ITU became an autonomous university which included the Faculties of Architecture, Civil Engineering, Mechanical Engineering, and Electrical and Electronic Engineering. Of ITU's five campuses, the main campus is located at Ayazağa, a recently developed business area. The Rector's office and administrative units are situated on this campus. Faculties of: Civil Engineering, Electric and Electronic Engineering, Chemical and Metallurgical Engineering, Minig, Science and Letters, Aeronautics, and Naval Architecture and Ocean Engineering are all on this campus which extends over an area of 247 hectares. Of the five institutes ITU has, four are located on this campus comprising: the Institute of Earth Sciences and the Institute of Information Technology. The Faculty of Aeronautics and Astronautics was established on March 3rd, 1983 as 11th Faculty of Istanbul Technical University. The Faculty consists of three departments: Aeronautical Engineering, Astronautical Engineering, and Meteorological Engineering. Aeronautical Engineering Department was first established as a branch of the Mechanical Engineering Faculty in 1941, and then in 1944 became a department of the Mechanical Engineering Faculty. Meteorological Engineering Department was first established in 1953 within the Faculty of Electricity. In

1971, the department was transferred to the Faculty of Basic Sciences, in 1982 to the Mining Faculty, and in 1983 to the Faculty of Aeronautics and Astronautics. Astronautical Engineering Department was first established in 1983 together with the faculty and started accepting undergraduate students in 1986 [2]. The authors of this article are studying at the Faculty of Aeronautics and Astronautics, also in the frame of the National Scholarship Program for Graduate Students of The Scientific and Technological Research Council of Turkey (TUBITAK-BAYG) - under the special direction of Assistant Professor Melike NIKBAY. As they told EnginSoft: “We are now going on our graduate program in Aeronautical Engineering, this will see us working as researchers in the TUBITAK Project which is about “Analysis and Reliability Based Design Optimization of Fluid-Structure Interaction Problems Subject to Instability Phenomena”. We have been using the modeFRONTIER® optimization tool for our optimization problems for some time. Until now, we have presented our work at 4 International Conferences: American Institute of Aeronautics and Astronautics (AIAA), International Conference on Machine Design and Production (UMTIK), World Congress on Computational Mechanics and Asian Pacific Congress on Computational Mechanics (WCCM/APCOM), and 2 papers at the Third International Conference on Multidisciplinary Design Optimization and Applications by ASMDO Association for Simulation and Multidisciplinary Design Optimization”. References [1] Nikbay, M. and Ulucenk, C. and Yanangonul, A. and Aysan, A. Reliability-Based Multi-objective Optimization of an Aircraft Wing Structure with Abstract Optimization Variables In Proc. 5. Ankara International Aerospace Conference Metu, Ankara, Turkey, 2009. [2] ITU Official Website, http://www.itu.edu.tr/en.

Melike NIKBAY - Istanbul Technical University, Turkey Assistant Professor, Astronautical Engineering Department, Faculty of Aeronautics and Astronautics, Istanbul Technical University, Maslak, Istanbul, Turkey, 34469; AIAA Member (nikbay@itu.edu.tr) Necati FAKKUSOĞLU - Istanbul Technical University, Turkey Research Assistant, Faculty of Aeronautics and Astronautics, Istanbul Technical University, Maslak, Istanbul, Turkey, 34469 (necobjk@hotmail.com) Muhammet N. KURU - Istanbul Technical University, Turkey Graduate Student, Informatics Institute, Computational Science and Engineering, Istanbul Technical University, Maslak, Istanbul, Turkey, 34469 (muhammet_kuru@hotmail.com)

- Newsletter EnginSoft Year 7 n°3

Simple Optimization of Gradated Biomaterial Scaffolds made of Calcium Phospates One of the mainstreams of modern biomaterials is the application of calcium hydroxyapatite (HAP) [1,2]. Various combinations of HAP as scaffolds and coatings or as a component of bioceramics and composites are being tested. To enhance osteoconductivity and osteointegraton, additions of other calcium phosphates (CP) like beta-tricalcium phosphate Ca3(PO4)2 (ß-TCP), glass-ceramics etc. are being exploited. Different processing methods of these compounds result in a variety of their mechanical, chemical and biological properties, which is also affected by density (porosity) and possible cross-interactions. There are also differences between in vitro and in vivo conditions, and not all of them are precisely known. Earlier homogeneous compositions of HAP+(25…75%) ßTCP have been considered as the compromise solution for optimal osteointegration, since ß-TCP is known to dissolve faster in both simulated body fluid (SBF) and in vivo conditions [2], although this also depends on porosity and crystal size. Protein adsorption on HAP delays bone formation, so a compromise solution should be sought to balance all these factors to ensure proper osteointegration. It has been suggested that a functionally gradated material (FGM) with a smoothly changing concentration and porosity profile would provide a better solution for the implants and scaffolds [2]. This kind of controlled porous material can be manufactured by a powder metallurgy technique with mixing and sintering. The latter however leads to generation of thermal stresses due to the difference in thermal expansion and sintering rates. Thus in the case of a FGM disk, this will lead to bending and twisting, up to possible cracking. Uneven stresses and remained porosity will not guarantee proper bioresorbabilty of the material - degradation of CP occurs preferentially on grain boundaries when the soluble phases disappear and the grains of less-soluble CP phases are released into a body environment. Such particle release is a cause of concern due to osteolysis (bone loss). To define the most optimal FGM profile including thermal and sintering residual stresses over the whole processing range, kinetics of co-sintering of HAP and ß-TCP should be analysed and the experimental data fed into the sintering model. The developed generic model of sintering [3,4] is based

on visco-elasto-plastic behaviour of the material, when its properties depend on porosity and grain size, coupled with thermal expansion. For example, for pure HAP it was experimentally found that the sintering kinetics could be approximated (±10% error) with Avraami-Erofeev’s equation:

where α – degree of sintering as measured by dilatometry. Explicit solution of this differential equation for shrinkage is:

for any programmable heating rate. When sintering a mixture, the shrinkage is more complex and a numerical fitting is usually required to incorporate it into models [3,4]. Fig. 1 shows the measured and interpolated shrinkage of HAP + ßTCP mixtures. It could be seen that with <20% of HAP in the mixture, shrinkage remains low until high temperatures, and for mixtures with >60% HAP, it does not deviate too much from the pure HAP. Using experimental data and equations for thermal expansion and contraction, the MathCAD model for thermal stresses of FGM with arbitrary thickness and gradation profile has been set up. Properties of FGM were calculated using a

Fig. 1 - Interpolated global shrinkage of the HAP + ß-TCP composites from 700°C.

Newsletter EnginSoft Year 7 n°3 -

micromechanical model [4] and the simplified stress analysis was preformed using the linear plate theory. The gradation profile was assumed to follow a power function:

three S-criteria are plotted together (Fig. 2), the effect of gradation parameter (colour) is clearly more important than the thickness. Whereas mean values of P (green bubbles), corresponding to near linear gradation, provide minimum of S1, this does not automatically guarantee minimum of S2 values neither curvature (S3).

where the volume fraction of HAP depends on x (the thickness coordinate), HFGM (thickness of the FGM layer (1…5 mm)) and P – gradation parameter (0.01… 100). Because stress evolution is non-linear for processing temperature and composition, it is necessary to establish single integral variables, which represent some measure of these stresses and relate them to other compositions with different thickness and gradation. Global integrated parameters chosen for this analysis are stress derivatives differences (S1), combined averaged stress difference (S2) and combined curvature of the FGM plate (S3):

The choice of these parameters instead of traditional ones was dictated by the need of a single parameter, which is capable of integration of information about the whole FGM plate during the whole temperature range from beginning of sintering (T0) to sintering temperature (Tsin), without explicit analysis of stresses in every point at every time and temperature step. It is known from FGM barrier coatings optimization that stress differences and their derivative differences are also important for material performance besides the absolute stress magnitude. The criteria S1 and S2 do not distinguish between tensile and compressive stresses but consider only their absolute values and thus might be overcautious in indication of “optimal” gradation. Nevertheless, they are believed to represent the general trends and to define the area, which would be further analysed using numerical methods in more detail. The objective of finding this optimal gradation is in minimising of all three criteria S1-S3. The model was set up with modeFRONTIER® 4.1.2, using a single MathCAD node for model calculation. Stored results were exported to Grapheur 1.0 visualisation software, where these stress differential changes (S1), averaged stress difference (S2) and averaged curvature (S3) were analysed as functions of the input variables gradation parameter P (and log(P)) and thickness HFGM. It is expected that thicker layer will have much less curvature. FGM with a large value of P seems to have the least stress difference for all thicknesses analysed. When all

Fig. 2 - Mutual dependence of all criteria. The best solutions are located closest to the coordinates origin.

Fig. 2 shows that that thicker FGM plates (larger bubbles) with a larger gradation parameter (red colour), i.e. a thin graded layer (~20% of the total thickness of the ß-TCP layer) will lead to the lowest curvature, stresses and their derivatives during the whole sintering process. In the case of the optimised profile, this HAP-rich layer with lower porosity and dissolution rate might provide an interesting effect on osteointegration as well. This is expected to ensure better stability of the scaffold in the body after implantation due to less destructive acting of internal stresses. In the future, more complicated geometries and different sintering regimes might be also simulated to find out the optimal set of processing parameters.

References [1] Oonishi H., Biomater., 12 (1991) 3, 171-178. [2] Pompe W., Worch H., Epple M., Friess W. et al. Mater. Sci. Eng. A362 (2003) 40-60. [3] Gasik M., Zhang B. Comp. Mater. Sci., 18 (2000) 93-101. [4] Gasik M. Comp. Mater. Sci., 13 (1998) 42-55.

For more information: mgasik@cc.hut.fi

Michael Gasik Aalto University Foundation, Finland www.aalto.fi

- Newsletter EnginSoft Year 7 n°3

ANSYS 13: Preview Nei mesi conclusivi del 2010 è in previsione l’uscita della release 13 di ANSYS. In realtà, già da alcuni mesi, EnginSoft ha preso parte ai test della suddetta release, prendendo piena coscienza delle qualità e delle notevoli novità introdotte dalla nuova versione. In questa parte si vuole porre l’accento sugli importanti “improvements” che caratterizzano ANSYS 13, sottolineando tutti i potenziamenti che caratterizzano l’ambiente di meshatura, la modellazione geometrica e le varie tipologie di simulazione. Alcune novità importanti riguardano la modellazione delle beam e delle shell; si potrà scegliere di colorare gli spigoli delle shell in funzione del numero di elementi con i quali sono Figura 1 – Mesh in visualizzazione wireframe Uno degli aspetti innovativi riguarda la possibilità di a contatto; sarà possibile visualizzare la mesh in trasparenza selezionare geometrie in funzione delle coordinate, oppure e non solo il “wireframe” della geometria come nella collegando le facce agli spigoli selezionati e così via (Figura 3). precedente release (Figura 1). Inoltre si potrà definire per le shell uno spessore variabile in funzione di una coordinata (vedi Figura 2), mentre anche per I contatti, oltre a poterli rinominare in funzione della gli elementi beam si potrà inserire la pretensione. nomenclatura geometrica dei corpi, potranno essere raggruppati in maniera da renderli più accessibili e facilmente individuabili per assiemi complessi formati da molte parti in contatto fra loro (Figura 4).

Figura 2 – Spessore variabile per le shell

Figura 3 – Selezione in ambiente WB di differenti geometri

Per quanto riguarda nuove tipologie di analisi, è stata implementata all’interno di ANSYS WB l’analisi di creep, con il conseguente inserimento all’interno dell’”Engineering Data” dei materiali di creep. All’interno dell’Engineering Data si potranno anche inserire le proprietà dei materiali iperelastici in funzione della temperatura. Sarà possibile poi impostare un’analisi “Gasket”, introducendo per le parti di interesse un comportamento di rigidezza di tipo Gasket, oltre che semplicemente rigido o flessibile come nella release 12. Sempre nella finestra di dettaglio relativa ai settaggi dell’analisi è stata introdotta la gestione dei controlli di stabilizzazione per poter aiutare la convergenza di una soluzione. Per quanto riguarda il controllo del “run” di soluzione, ANSYS ha inserito la possibilità di poter ripartitre con l’analisi dall’ultimo istante converso; in questo modo si eviterà di dover

Newsletter EnginSoft Year 7 n°3 -

svolgere di nuovo completamente il calcolo, ripartendo invece dall’ultima soluzione raggiunta. ANSYS WB13 offrirà la possibilità di svolgere analisi armoniche ed analisi modali in presenza di coazioni, anche funzione di differenti load-step di analisi statiche precedentemente svolte potendo osservare come cambiano i modi propri di vibrare durante l’applicazione Figura 4 – Nuova organizzazione dei contatti del carico. Infine, altre novità importanti sono rappresentate dalla inserimento al suo interno di tutte le analisi a disposizione; facilità di importazione di risultati esterni, con la possibilità nella versione 13 sarà possibile utilizzare HFSS e Maxwell di trasferire carichi da analisi 2D ad analisi 3D; da CFD sarà direttamente in interfaccia Workbench, permettendone il possibile trasferire anche carichi termici volumetrici e non successivo collegamento ad analisi termiche e strutturali più solo superficiali (Figura 5). (Figura 6). Sempre all’inseguimento della creazione di un’unica piattaforma “multiphysics”, dalla quale accedere a tutte le diverse tipologie di analisi, ANSYS continua il processo di

Figura 5- Trasferimento carichi da 2D a 3D

Nel campo della dinamica esplicita è stato implementato anche l’ambiente euleriano all’interno di WB, con la possibilità di interazione eulerianalagrangiana tra le differenti parti del sistema in esame. In conclusione, è stato fatto un ulteriore passo avanti per quanto riguarda l’analisi cinematica con corpi rigidi; vi sarà la possibilità di introdurre contatti anche tra corpi rigidi: con la possibilità di mantenere comunque un solutore esplicito non ricorrendo al tradizionale solutore implicito di ANSYS con la possibilità di conseguenza, di risolvere analisi cinematiche in pochi minuti. Daniele Calsolaro - EnginSoft Per maggiori informazioni: Emiliano D’Alessandro - EnginSoft info@enginsoft.it

Figura 4 – Nuova organizzazione dei contatti

- Newsletter EnginSoft Year 7 n°3

I prodotti ANSYS al servizio della progettazione e della simulazione dei motori elettrici: La verticalizzazione RMxprt-Maxwell-ANSYS Mechanical. I prodotti Ansoft Da circa 2 anni il pacchetto di prodotti di casa ANSYS si è arricchito dei software Ansoft. I prodotti Ansoft sono programmi ad alte prestazioni per il design e l’automazione in ambito elettronico ed elettromeccanico (EDA -Electronic Design Automation software). Con all’attivo ormai oltre 25 anni di sviluppo, i software di casa Ansoft rappresentano lo stato dell’arte nella simulazione elettromagnetica.

riportati i principali software di Ansoft per l’analisi di componenti e di sistemi. La tecnica della cosimulazione, implementata nei prodotti Ansoft, consente in particolare di simulare all’interno dello stesso ambiente di lavoro schemi circuitali accoppiati a modelli agli elementi finiti (Figura 2). Il fulcro di questo tipo di applicazioni è la tecnologia Simplorer, che verrà trattata in

Tra i prodotti Ansoft più diffusi citiamo HFSS, software che implementa il metodo agli elementi finiti per l’analisi “full wave” di problemi elettromagnetici in alta frequenza, Simplorer, simulatore di circuiti complessi per la modellazione e l’analisi di sistemi meccatronici, e Maxwell, software agli elementi finiti per l’analisi di dispositivi operanti in bassa frequenza, di cui accenneremo in questa nota. Una delle principali linee di sviluppo dei prodotti Ansoft è la possibilità di fornire da un lato un insieme di tools per la simulazione e la Figura 2; Ansoft Simplorer: cosimulazione del modello agli elementi finiti, dello schema verifica di singoli componenti elettromagnetici, circuitale e dei controlli. dall’altro strumenti e metodologie che maniera approfondita nei prossimi numeri della della permettano l’analisi di sistemi complessi, meccatronici ed newsletter. ibridi in generale, implementando tecniche di cosimulazione e di estrazione di modelli a parametri concentrati. Tale approccio viene sintetizzato in Figura 1, dove sono Le nuove soluzioni software di Ansoft affiancano EMAG, il tradizionale prodotto per le analisi elettromagnetiche in ANSYS, rafforzandone le potenzialità in ambito multifisico e introducendo una serie di verticalizzazioni e customizzazioni su specifiche tipologie di prodotti. La simulazione multifisica si realizza nell’interfaccia ANSYS Workbench che consente l’integrazione tra i solutori elettromagnetici di Ansoft e quelli CFD, termici e strutturali di ANSYS.

Figura 1: Component design e System design in Ansoft

Nel presente articolo è presentata la soluzione relativa alla progettazione e alla simuolazione dei motori elettrici. La verticalizzazione proposta utilizza i software RMxprt e Maxwell 2D/3D di Ansoft, oltre ad i solutori termici e strutturali di ANSYS.

Newsletter EnginSoft Year 7 n°3 -

laddove il modello presenta delle zone critiche assicurando la convergenza (b) della soluzione. In Figura 4 il metodo di mesh auto adattivo viene sintetizzato con un diagramma a blocchi. È mostrato inoltre, su un modello 3D, la mesh prodotta al primo ed all’ultimo passo di convergenza.(b). Maxwell permette, in funzione del solutore utilizzato, la determinazione accurata delle forze Figura 3; Alcune applicazioni in Maxwell 3D: campo di induzione magnetica valutato mediante e delle coppie, dei valori di capacità, di analisi transient su un motore elettrico (sinistra); perdite ohmiche valutate sul primario e sul induttanza, di resistenza e di impedenza. il secondario di un trasformatore elettrico (destra) software consente altresì di valutare agevolmente I software Maxwell2D/3D e RMxprt alcune grandezze di indubbia importanza in ambito Maxwell è un software efficace ed efficiente per la industriale quali correnti parassite, perdite ohmiche e nel simulazione dei campi elettrici e magnetici. Basato sul ferro. metodo agli elementi finiti, Maxwell consente di analizzare il comportamento elettromagnetico di strutture e componenti Nell’ambito della progettazione di macchine rotanti, a quali motori elettrici, attuatori, trasformatori, converters ed Maxwell si affianca, condividendone l’interfaccia, uno altri congegni elettrici ed elettro-meccanici comuni ai settori strumento dedicato: RMxprt. automotive, aereospace, della difesa e dell’industria in generale. Maxwell consente inoltre di effettuare diverse tipologie di analisi statiche, armoniche e transient operando su geometrie complesse in domini bidimensionali e tridimensionali. In Figura 3 vengono mostrati alcuni esempi di applicazioni in Maxwell3D. Un’ analisi transient di un motore elettrico brushless a magneti permanenti (a) e un’analisi armonica di un trasformatore elettrico (b). Fin dalla prima release, Maxwell utilizza tecniche di meshatura auto-adattive che Figura 5; Alcune applicazioni dell’interfaccia di RMxprt consentono di ottenere un ottimo compromesso tra onere computazionale della simulazione, Utilizzando la teoria classica delle macchine elettriche ed il dovuto al numero di elementi necessari a discretizzare una concetto di circuito equivalente RMxprt calcola struttura, e accuratezza della soluzione proposta. Basati su istantaneamente, fornendo le principali caratteristiche, il criteri energetici, tali algoritmi creano e raffinano la mesh comportamento della macchina nelle diverse alternative progettuali. RMxprt ha una semplice interfaccia grafica per inserire i parametri progettuali relativi alla geometria di rotore e statore, ai settaggi degli avvolgimenti e alle caratteristiche dei materiali. In Figura 5 è mostrata l’interfaccia di RMxprt. Il motore rappresentato è un asincrono trifase. RMxprt consente di analizzare in modo esaustivo una vasta gamma di motori elettrici: macchine sincrone ed asincrone, macchine con commutazione a spazzola ed elettronica, alternatori, ecc.

Figura 4; Come lavora l’altgoritmo di mesh auto-adattivo: diagramma a blocchi (a); mesh al primo ed all’ultimo passo di convergenza (b)

La release 13 di RMxprt, ultima nata in casa Ansoft, presenta inoltre un nuovo slot editor che permette al progettista la modellazione della geometria di qualsiasi tipo di slot. RMxprt considera gli effetti pelle ed alcuni effetti 3D, come le geometrie delle teste di matassa e l’inclinazione delle cave.

- Newsletter EnginSoft Year 7 n°3

Figura 6; L’integrazione RMxprt Maxwell2D/3D: RMxprt automaticamente genera i modelli per Maxwell2D e Maxwell3D

Vengono inoltre considerati nel calcolo gli effetti della saturazione del ferro. Le elevate potenzialità del software sono completate con un ottimizzatore integrato: Optimetrics, che consente la

L’accoppiamento Maxewll-ANSYS per le analisi multifisiche La soluzione calcolata in Maxwell2D e 3D può essere utilizzata come boundary condition per il solutore termico e strutturale di ANSYS. In maniera completamente automatizzata, e senza bisogno di implementare script esterni, è infatti possibile eseguire analisi termiche 2-way fra Maxwell e ANSYS thermal. Maxwell rende disponibili le perdite di potenza per i modelli di ANSYS e legge le temperature output dell’analisi termica, per ricalcolare la soluzione elettromagnetica. L’esportazione delle perdite di potenza avviene da modelli 2D e 3D di Maxwell verso i solutori statici e transient di ANSYS. In Figura7 viene riportato un esempio di quanto detto: le perdite di potenza nel ferro calcolate da Maxwell2D sono mappate sulla geometria 3D in ANSYS. Per quanto riguarda l’analisi transient, possono essere trasferite le perdite valutate su singoli istanti o come media calcolata su un intervallo temporale opportuno. Sia il modello 2D di Maxwell che la geometria 3D utilizzati in figura 7 sono stati creati a partire dal modello parametrico di RMxprt.

Figura 7; Le perdite di potenza nel ferro calcolate da Maxwell 2D (sinistra) sono mappate sul modello termico di ANSYS (centro) per valutare le temperature (destra)

valutazione istantanea di diverse configurazioni di parametri, per individuare quella che permette il raggiungimento degli obbiettivi progettuali imposti, nel rispetto dei vincoli.

Per quanto riguarda le analisi strutturali sia la densità di forza volumetrica che le forze superficiali magnetiche calcolate da Maxwell possono essere importate dal modello strutturale di ANSYS. In figura 8 la densità di forza volumetrica calcolata da Maxwell 3D è mappate sul modello strutturale di ANSYS, come condizioni al contorno per una successiva analisi strutturale.

La verticalizzazione RMxprt-Maxwell per l’analisi dei motori elettrici Una volta definito il modello analitico attraverso l’interfaccia, RMxprt crea automaticamente il modello agli Come accennato precedentemente, l’accoppiamento elementi finiti 2D e 3D per Maxwell, trasferendo: la multifisico descritto non richiede l’utilizzo di script esterni o geometria, le caratteristiche di moto e le proprietà macro, poichè la procedura è implementata completamente meccaniche (inerzia e coppia resistente all’albero), i dati dei materiali (curva BH e di perdita del ferro), il setup degli avvolgimenti e l’alimentazione. In definitiva il modello così creato è pronto ad essere lanciato con i solutori transient di Maxwell. La soluzione agli elementi finiti in Maxwell permette oltre alla determinazione dei transitori, i valori puntuali di campo ed una soluzione più raffinata ed accurata di quella analitica ottenuta con RMxprt. La Figura 6 mostra lo stesso modello in RMxprt ed Figura 8; La densità di forza volumetrica calcolata da Maxwell 3D (a sinistra la mesh in in Maxwell 2D e 3D. Maxwell) è mappata sul modello strutturale di ANSYS (destra)

Newsletter EnginSoft Year 7 n°3 -

condividere le informazioni con altri blocchi di analisi attraverso le procedure tipiche dell’interfaccia WB. In Figura 9 come appare l’applicativo Maxwell in interfaccia ANSYS Workbench 13.

Figura 9; Maxwell ed altri prodotti Ansoft sono presenti in interfaccia AnsysWB.

dalle interfacce di Maxwell e ANSYSWB. Inoltre le mesh di ANSYS e Maxwell sono indipendenti. L’interfaccia ANSYS13-Maxwell14 Con le prossime release di Maxwell14 ed ANSYS13 a breve in uscita, l’integrazione fra i moduli elettromagnetici di Ansoft e l’ambiente di simulazione di ANSYSWB verrà ulteriormente migliorato. Maxwell potrà infatti essere lanciato direttamente dall’interfaccia di ANSYSWB, l’icona presente nella finestra dell’Analysis System. In questo modo Maxwell2D/3D potrà essere utilizzato all’interno del Project Schematic di ANSYS Workbench e

Conclusioni Nella presente nota è stata presentata la soluzione ANSYS per la progettazione e la verifica dei motori elettrici. La procedura proposta non si limita al dominio elettromagnetico, ma grazie alle tecnologie implementate in ANSYS, permette l’analisi anche in ambito termico e strutturale, in maniera semplice ed efficace. La piattaforma nella quale si realizza questa tecnologia è ANSYSWB, che nella release 13, segna un significativo avanzamento nella completa integrazione fra le principali tecnologie Ansoft ed i solutori di ANSYS. Per maggiori informazioni: Emiliano D’Alessandro - EnginSoft info@enginsoft.it

The New ANSYS Frontier Product: ANSYS EKM (Engineering Knowledge Manager) An interesting benchmark experience with Ansaldo Energia ANSYS Engineering Knowledge Manager (ANSYS EKM) is a simulation process and data management (SPDM) software product that provides solutions for engineers who are challenged with managing the vast amounts of data and best practices that are generated in simulation activities. ANSYS EKM is a web-based solution with an easy-to-use and intuitive user interface. The technology’s capabilities range from simple archival and management of simulation data to process automation and capture and deployment of best practices, version control and branching, audit traceability and dependency mapping, advanced search and retrieval, report generation and simulation comparison and extensive customization. It is worth noting that ANSYS EKM is a different solution compared to existing PLM/PDM

Fig. 1 - Web access interface

- Newsletter EnginSoft Year 7 n°3

systems. ANSYS EKM is focused on the CAE simulation aspect of product engineering. This ANSYS product allows engineers and other users to index, archive, search and retrieve simulation and supporting files. ANSYS EKM supports automated meta-data extraction and report generation for all ANSYS simulation products. It also supports other commercial CAE solution formats and can be configured to support internal/legacy codes, if necessary. ANSYS EKM is presented as a “complementary” solution to PLM/PDM systems. The ANSYS EKM Datalink capability allows direct bi-directional interface between ANSYS EKM and commercial PLM/PDM systems.

Typically, a new analysis project for a turbine blade begins by searching existing CAE analysis files. The pre-analyzed CAE database is a good starting point for a new product, since typical CAE database includes a wealth of information like simulation intent, materials, solver settings, results etc that can be leveraged in the analysis work at hand.

By using a web browser and a password-protected internet/network connection, ANSYS EKM can be accessed from any geographical location, even when the user is travelling.

At Ansaldo Energia, ANSYS EKM was used for managing the simulation data and defining a simulation process that led to an efficient and collaborative method for new turbine blade design and analysis. This method established a benchmark case study at Ansaldo Energia.

EnginSoft, an ANSYS channel partner in Italy, is committed to developing benchmark evaluations and case studies that showcase how CAE can save companies time and money in their product development activities.

Fig. 2 - EnginSoft Compute Cluster Configuration

Recently, EnginSoft was offered an opportunity to present the ANSYS EKM solution to Ansaldo Energia for their Simulation Data and Process Management needs. The Benchmark at Ansaldo Energia Ansaldo Energia is a long-time user of software from ANSYS. The energy company leverages multiple ANSYS simulation tools to analyze different physical phenomenon in the process of typical turbine blade analysis.

The indexing and archival capability of ANSYS EKM is very useful in searching and retrieving existing simulation files. These files can then be efficiently leveraged for the current simulation work at hand.

Using ANSYS EKM Studio, a simple simulation workflow was built for this benchmark case study (Figure 2). The workflow consisted of multiple nodes or tasks, connected

by transitions, or actions. For example, geometry /CAD design creation (using Pro/ENGINEER software), a CFD analysis (using ANSYS CFX) and a mechanical analysis (using ANSYS static structural code) are the three tasks involved in this workflow. Each task can have a staff member assigned who is expected to complete the work – or a machine assigned (a queue system such as RSF/SGE and compute cluster, etc.) This workflow involved four users: a lead analyst, a CAD expert, a CFD expert and an FEM expert. These staff members each work in different departments, sometimes at geographically different locations. They usually communicate with each other using conventional communication methods, such as telephones, emails, etc.

Newsletter EnginSoft Year 7 n°3 -

Using ANSYS EKM Studio, the lead analyst creates the workflow and sets up the project structure and access permissions for the other assignees. This helps ensure that appropriate data access is granted to the users and they can work in a collaborative way, without affecting each others’ work. When the lead analyst starts a simulation process using this workflow, the next-in-line assignee automatically receives an email notification via ANSYS EKM about the work item or task that is assigned to him. This work item includes all necessary information about the task. The user can then complete the work and mark it as done in ANSYS EKM. This generates an email to the next assignee, and the automated communication goes on.

Fig. 3 - Total deformation and equivalent stress results for the new turbine blade

The simulation process was made very efficient and productive by this automated communication. The actual actions of the user were captured in the ANSYS EKM process status, a tool helpful in reviewing or auditing the process at a later dated, if needed. Through such coordination, the software can reduce downtime associated with communication and data transfer between different users. Furthermore, it was possible to create an interface between ANSYS EKM and the company’s server and cluster; the benchmark development was organized between the different EnginSoft headquarters staff, depending by their singular capabilities (Figure2). By using automated or batch processing nodes in the workflow, it was possible to execute the simulation in batch mode on EnginSoft server and cluster. At the end of the batch execution, simulation result files were automatically uploaded back to ANSYS EKM. This helped in using these files in subsequent nodes/actions in the workflow. Based on the server cluster configuration and available simulation software, the ANSYS EKM benchmark workflow was executed in batch mode to complete the remote simulation runs At the end of the benchmark evaluation, Ansaldo Energia concluded that • Prior analysis files were easily searched and retrieved using the advanced search capability in ANSYS EKM. The turn-around time was quick and the search process was efficient. • Since it was easy to find pre-existing solutions and reuse them, the need to perform a repeat analysis was

minimized. This resulted in better utilization of resources. • The overall design and analysis process was well coordinated. The various team members rated the ability to collaborate in the simulation process as excellent.

Ansaldo Energia staff expressed an interest in improving ANSYS EKM so it could execute multi-objective optimization, leveraging the parametric analysis capability of the software used in this workflow.

Andrea Pancrazzi - EnginSoft info@enginsoft.it Ing. Daniele Calsolaro - EnginSoft info@enginsoft.it

Comment by Shantanu Bhide (Product Manager, ANSYS EKM), ANSYS, Inc. as inset: The ANSYS Engineering Knowledge Manager (EKM) is a comprehensive solution for simulation-based process and data management challenges. ANSYS EKM provides solutions and benefits to all levels of the enterprise, from the individual engineer interested in spending less time handling data and more time focusing on true engineering efforts to the entire organization looking for increased productivity in all aspects of its simulation activities. It enables the enterprise to address the many critical issues associated with simulation data including backup and archival, traceability and audit trail, process automation, collaboration and capture of engineering expertise, and IP protection.

- Newsletter EnginSoft Year 7 n°3

ESAComp Versione 4.1 – Strumento basilare per la progettazione delle strutture in composito ESAComp è un software sviluppato espressamente per l’analisi ed il design delle strutture in materiale composito. Esso è capace inoltre di supportare efficacemente lo sviluppo della fase concettuale e preliminare nel rispetto dei molteplici requisiti caratteristici di un progetto nell’ambito di strutture complesse in materiale composito. ESAComp dispone di strumenti e capacità che possono essere sfruttate sia in maniera autonoma sia come supporto ed integrazione delle funzioni dei più diffusi pacchetti di software agli elementi finiti. ESAComp comprende inoltre un database costantemente aggiornato e basato su informazioni provenienti da fonti complementari, tra cui anche fornitori industriali, che racchiude

ed integrati ed ESAComp costituisce uno degli elementi principali e specifici del processo logico (cfr. Figura 2); così come ANSYS ACP permette di verificare la necessaria fattibilità produttività di componenti geometrici in composito tramite l’analisi di drappabilità e di risolvere puntualmente il problema strutturale di dettaglio, ad ESAComp è demandata la risoluzione

Figura 2 – Processo logico progettuale di strutture in materiale composito

Figura 1 – Modulo specifico per il dimensionamento e l’analisi dei rinforzi strutturali in composito

tutte le proprietà dei materiali necessarie alla soluzione di problemi ingegneristici. La nuova versione 4.1 di ESAComp estende e migliora tutte le precedenti capacità, rendendolo lo strumento ideale e necessario per la progettazione delle strutture in composito. In particolare la nuova versione introduce strumenti e processi che permettono di incrementare la velocità di sviluppo di un progetto e raggiungere soluzioni maggiormente performanti, come ad esempio il modulo per il predimensionamento dei rinforzi strutturali su piastre piane o curve (cfr. Figura 1). In un processo progettuale che miri a risolvere efficacemente ed esaustivamente ogni specifica fase per una struttura in materiale composito, integrando strumenti CAD e codici numerici agli elementi finiti, come ad esempio ANSYS ACP – ANSYS Composite PrePost, è necessario disporre di strumenti verticalizzati

delle seguenti fasi: progetto concettuale, selezione dei materiali, tramite logiche specifiche e di confronto, e progettazione preliminare dei singoli piani di laminazione (cfr. Figura 3). Attraverso la completa integrazione e le iterazioni progettuali, i dati possono essere interscambiati tra ANSYS ACP ed ESAComp. Infine una sintesi delle principali novità introdotte nella nuova versione di ESAComp è riportata nelle pagine seguenti.

Figura 3 – Funzioni specifiche per la soluzioni di problemi complessi e soddisfazione di requisiti differenti

Newsletter EnginSoft Year 7 n°3 -

ESAComp 4.1: New Features Version 4.1.086 for MS Windows NEW ANALYSIS CAPABILITIES • Curved plates extend the old analysis capability for rectangular plates to singly curved plates defined by the plate dimensions and the radius of curvature. Stiffeners can be placed in the axial direction. The boundary conditions can be independently defined for each edge of the plate. Linear-static load response and failure analysis can be performed under pressure and point loads. Buckling and natural frequency analyses are introduced in future ESAComp releases. As a special

case of curved plates, flat plates can also be defined. The new implementation will fully replace the old plate analysis in the future. • Hat stiffeners can be defined for curved plates besides the beam type stiffeners supported by earlier ESAComp versions. The hat stiffener types include bonded and integral stiffeners. The capabilities for defining both stiffener types are extensive. For instance, besides the hat laminate there are possibilities to define additional reinforcing layers for the sides and top part of the hat. In the analyses hat stiffeners are modeled using shell elements. • The Cylindrical shell add-on module allows analyses of cylinder and tube-like structures. The cylinder may have a constant diameter or it may be conical. The laminate lay-up may vary in the axial direction by assigning different laminates for ring type cylinder segments. The boundary conditions at the ends of the structure are defined using an innovative and simple-to-use approach. Forces and moments can be applied at the ends of the

cylinder. In addition, a pressure load or inertial loads due to linear acceleration or rotation may be applied. The analysis types include static load response and failure, as well as buckling and natural frequency analyses based on linear eigenvalue approach. • The Stiffened cylindrical shell add-on module is a further extension of the cylindrical shell module. Beam type stiffeners can be placed in the axial and circumferential directions. The locations are specified independently for each stiffener. The stiffeners may be on the inside or outside of the cylinder. The analysis possibilities are identical to the standard cylindrical shell module. • The Elmer FE solver by CSC, The Finnish IT Center for Science (www.csc.fi/elmer) is now included as a standard module in the ESAComp distribution. It is used for realizing the curved plate and cylindrical shell analyses and it also provides a basis for introducing advanced nonlinear analyses in the future ESAComp releases. • A new 3D result viewer is introduced for viewing the results of curved plate and cylindrical shell analyses. Versatile capabilities for model rotation and zooming are included. The selected result item can be viewed as a contour plot with optional annotations for failure modes and critical layers. The features include also deformed plots and animation of eigenmodes. For a selected element, layer level stresses, strains and reserve factors can be viewed as layer charts and in numeric format. NEW DATA EXCHANGE CAPABILITIES • FE export to ANSYS Composite PrepPost (ACP). ACP supports data exchange with ESAComp using the

•

- Newsletter EnginSoft Year 7 n°3

ESAComp XML format. The new export capability improves the possibilities by writing an ACP specific Python script which can be simply copied and pasted to the ACP command prompt. Both ply materials and laminate lay-ups can be exported. Laminate lay-ups from ESAComp can be interpreted as Sub Laminates or Stack-ups in ACP. FE export to ANSYS Workbench allows creation of an ANSYS specific XML file that can be read in by the WB Engineering Data module. Isotropic and orthotropic ply materials can be exported. FE export to ComPoLyX allows transfer of ESAComp ply material data in the form of a ComPoLyX Python script. The typically incomplete material data from FE models can be completed with the material description from ESAComp before performing advanced failure analyses in ComPoLyX. ABAQUS export has been improved for ABAQUS SHELL elements through the use of ESAComp extension variables (Edit -> Extension variables...). For each ply of the active case, an FEA related material ID can be specified. Similarly, for each laminate an FEA related section type ID and reference plane data can be set. Consequently, these ID's are used when export is made. Support for unit systems in ESAComp XML. In the earlier versions all ESAComp XML data exchange was done in basic SI units. Now, the FE import/export unit options can be used for selecting the unit system. If imported XML includes a header indicating the unit system, this information is used instead of the selected unit options.

DATA BANK UPDATE • The ESAComp Data Bank has been updated extensively. The update covers the following material types: foam cores, honeycomb cores, other cores, carbon fibers, glass fibers, typical aramid fibers, polyester resins, vinylester resins, some epoxy resins including typical classes, homogeneous plies, typical FRP, CSM, Spray up rovings, MMC, and plywood. LICENSING AND INSTALLATION • RLM license manager by Reprise Software, Inc. has replaced the earlier-used FLEXlm in ESAComp licensing. To the IT administrator and ESAComp end-user, RLM provides a userfriendly web browser interface for configuring the license server and for monitoring license usage. Node-locked

licenses are handled with a very simple license file based approach. No license server is needed for nodelocked licenses. Old license files are not compatible with the new licensing system. ESAComp users that are eligible for the version upgrade will receive new license files. • Along with the new licensing, version numbers are now based on the release date – OR “A release-date-based version number is now used with the new licensing”. In the license file, the highest supported version number is shown, for example, as “2010.12”. This indicates that the license is valid for all versions released in December 2010 or before that. The release date based version number is shown on the ESAComp start-up screen besides the “normal version number”, e.g. “4.1.086 (2010.06)”. When a customer renews maintenance, a new license file with the updated version number (maintenance end date) is provided. This approach increases transparency of the licensing and makes it easy to take in use new software upgrades when available. • The new installation system allows flexible installation of ESAComp and RLM license server from the same installation package. The new installation procedure supports multiple users on the same PC. The user specific ESAComp files are by default placed under each user’s home directory (“$USERPROFILE\ESAComp\ …”). • In addition, many smaller enhancements have been made.

For more information: Marco Perillo - EnginSoft info@enginsoft.it

Newsletter EnginSoft Year 7 n°3 -

Peculiarità del software Coldform

Transvalor ha rilasciato a Giugno la versione 2010 di Coldform, il software dedicato alla simulazione dello stampaggio a freddo di viteria e minuteria metallica. 1. OTTIMIZZATORE multi-parametro e multi-obiettivo integrato: impostata una sequenza campione, Coldform è in grado di individuare, il miglior compromesso dei parametri (dimensioni e posizione filo) per ottenere il miglior risultato (assenza di ripieghe, completo riempimento, minor carico pressa,...). Per configurazioni più complesse, possibilità di interfacciamento con il nostro software di ottimizzazione modeFRONTIER®, in grado di guidare più software e collegarli tra di loro (es: entrare nel CAD e modificare dettagli parametrici quali lunghezze e curvature, trasferendo poi le geometrie a Coldform per il calcolo). 2. DATABASE DEI MATERIALI con circa 900 leghe (acciai, inox, alluminio, rame-ottone, titanio, nickel, …), con curve di deformazione ottenute da prove sperimentali. 3. DATABASE DI CINEMATICHE estremamente completo, con tutte le presse standard (meccaniche, idrauliche, a doppia ginocchiera, link-drive, ad energia, …), con la possibilità di impostare leggi di moto a piacere (rotazioni singole e multiple, traslazioni, combinazioni a piacere) per ogni utensile. Ogni stazione può essere legata alle precedenti, considerando tutti i transfer intermedi. 4. LICENZA multi-utente: nella configurazione standard, una macchina di calcolo e un numero a piacere, all'interno dello stesso stabilimento, di stazioni per la preparazione dei calcoli e

l'analisi dei risultati. Possibilità di licenze floating o configurazioni ad-hoc. 5. MULTI-PROCESSORE fin dalla nascita, nel 1994: possibilità di ridurre significativamente i tempi di calcolo usando tutti i "core" a disposizione. Possibilità di installare la licenza su pc standard e quindi estenderla a pc multi-core \ multi-processore (es: dualquadcore) o a cluster windows o linux fino a 32 core. Tipologia di analisi impostabili: • analisi 2D molto rapide per configurazioni assialsimmetriche (fasi di estrusione diretta ed inversa o combinata, ricalcatura della testa); • analisi 3D per configurazioni più complesse (ricalcatura di testa esagonale, creazione della chiave e di dettagli sottotesta, …); • possibilità di trasferire i risultati da 2D a 3D mantenendo la "storia di deformazione" del prodotto; • possibilità di concatenare le operazioni all'interno di una singola simulazione, con trasferimento automatico dei risultati tra le fasi. Livelli di accuratezza impostabili: 1. analisi limitata al pezzo, con ipotesi di stampi «rigidi»; 2. analisi "non accoppiata" degli stampi: calcolo del pezzo e valutazione della deformazione elastica degli stampi sottoposti al carico trasmesso dal pezzo in deformazione; 3. analisi "accoppiata" con calcolo congiunto di pezzo e stampi. Per i casi 2. e 3., possibilità di impostare configurazioni precaricate (blindaggio) tramite definizione dell'interferenza, anche per più anelli consecutivi (tool-stack). Inoltre: • analisi di trafilatura; • analisi di tranciatura di bave o di punzonatura, con configurazioni eventualmente flottanti su molle o su cuscini; • analisi di rollatura del filetto; • analisi del trattamento termico di tempra; • analisi di comportamento in esercizio: prove di serraggio vite-bullone (o fasteners in senso lato) e calcolo degli sforzi e delle deformazioni su vite, bullone e sugli oggetti da fissare.

- Newsletter EnginSoft Year 7 n°3

I principali risultati ottenibili dalla simulazione sono: Per il pezzo: • analisi dei contatti ed indviduazione della mancanze; • analisi delle ripieghe, con valutazione della genesi ed evoluzione nel pezzo; • analisi delle cricche generate da eccessivo stiro del materiale; • analisi termica dei riscaldamenti dovuti a scorrimenti interni, attrito con gli stampi, picchi di pressione; • analisi del ritorno elastico all'apertura degli stampi o all'estrazione dalle matrici. Per gli stampi: • analisi delle zone maggiormente sollecitate a rischio di rottura; • calcolo dell'usura termo-meccanica degli stampi; • analisi della deflessione dello stampo sotto il carico trasmesso dal pezzo in deformazione. Per la pressa di stampaggio: • calcolo del carico pressa e della sua distribuzione su ogni utensile (stampi, punzoni, spine, matrici); • calcolo del centro di carico dello stampo, per valutare sbilanciamenti rispetto al baricentro di stampaggio; • calibrazione della forza assorbita in ogni stazione di stampaggio, in modo da evitare sbilanciamenti della pressa, tenendo conto eventualmente anche della rigidità a flessione della macchina.

Preventivazione e Valutazione di Formabilità Partendo dal pezzo da realizzare, in pochi minuti si ottengono forma e dimensioni dello sviluppo, con un calcolo degli spessori e dei costi. Le versioni più complete consentono una valutazione di stampabilità, evidenziando rotture o grinzature, che possono essere eliminate intervenendo sui parametri di stampaggio (premilamiera, fori di centraggio, superfici di appoggio curve, …).

Identificare in pochi minuti le modifiche nel design che consentano una riduzione del costo del componente dal 10 al 15%!

Per maggiori informazioni: Marcello Gabrielli - EnginSoft info@enginsoft.it

Per tutti i prodotti • possibilità di intervenire sulle geometrie importate (.iges, .vda, .step) • meshatura automatica in pochi istanti • completo database di materiali con curve di deformazione e FLD

Newsletter EnginSoft Year 7 n°3 -

• report generato in automatico in formato .html e .xls • interfaccia comune nella suite • analisi rapidissime: in pochi minuti si ottiene il risultato • analisi facilitata dei risultati Per CATIA e SolidWorks • le versioni specifiche per CATIA e SolidWorks sono associative e rigenerative: ogni modifica introdotta nel modello si trasporta istantaneamente nella simulazione di formabilità.

Chiedi subito una prova gratuita del Prodotto! www.enginsoft.it/link/testfti

- Newsletter EnginSoft Year 7 n°3

Pushing Reservoir Data Handling to New Frontiers promote the integration of Kraken and the reservoir simulators GPAS (General Purpose Adaptive Simulator) and UTCHEM (University of Texas Chemical Compositional Simulator).

Developed by Engineering Simulation and Scientific Software (ESSS), Kraken is a reservoir simulation postprocessor that features a modern and powerful user interface designed for the visualization and handling of multiple scenarios and data sets. This functional post-processor natively reads and interprets data from ECLIPSE, IMEX, UTCHEM and several other simulators. By properly integrating numerical solutions, times and units, Kraken provides unique means of analyzing and comparing multiple simulations. Additionally, it is capable of handling grid and well data, while supporting structured, unstructured, and reservoir type grids. Kraken’s power and functionality is built upon ESSS’ expertise in developing and implementing numerical solutions for companies such as Petrobras, Shell, ExxonMobil, Total, Statoil, Chevron and Maersk. A few notable examples of these customized applications are: • SourSimRL: a simulator currently in use by most of the major oil and gas companies to avoid the formation of sulfide gas (H2S) during the secondary oil recovery process. • Cyclope: a powerful volumetric mesh converter for characterization and simulation of reservoirs. • SCBR 2.0: a complete simulation tool for analysis of chronic risk to human health, considering the multiple transportation routes of contaminants: air, ground, underground aquifers, and surface water. • PWDa: Pressure While Drilling (PWD) simulation, manipulation and data analysis tool. It identifies the key phenomena which impacts annular pressure during well drilling of petroleum reservoirs. It is also important to mention the collaborative development between ESSS and the Center for Petroleum and Geosystems Engineering (CPGE), in an effort to

“We have been working with ESSS during two years in the development of pre- and post-processing software for our reservoir simulators”, said Kamy Sepehrnoori, professor of Petroleum Engineering at CPGE. “We feel that ESSS has done an excellent job in working with us and performing the various tasks for this project”, he added. Kraken environment 1 - Visualization 3D Visualization of any type of grid, from traditional reservoir simulation to complex unstructured grids Kraken features a complete set of visualization processes for extracting, cutting and plotting the information obtained from the grid solution. It allows the user to select a region of interest within the grid, by using an IJK block or a set of cells with predefined property values.

Fig. 1 - This figure shows three examples of visualization processes to aid fluid behavior analysis. From top to bottom: streamlines from the injector wells, a plane cut between two wells and an iso-surface tracing the water saturation front.

Overall information such as average pressure, minimum and maximum fluid saturation plots, water saturation front along the transient solution, among others, can be obtained for the entire grid or specific regions of interest. 2 - Plotting XY Plots are fully interactive and linked to 3D visualization for easy data transition from one visualization context to another. A particular well can be selected and its

Newsletter EnginSoft Year 7 n°3 -

Fig. 2 - This figure illustrates the link between the 3D objects selection and the well’s production curves visualized using an XY plot and a spreadsheet.

production curves evaluated within an XY Plot window. To inspect numerical values and create additional information, formulas may be inserted using any data set from the model.

5 - Report generation Kraken’s features allow image and data exchange with other applications. Simple operations such as Copy and Paste can be used to transfer images from Kraken visualization windows to other applications, allowing the user to quickly insert a visual analysis into documents and reports. Data values may also be copied from XY Plot curves and pasted directly into spreadsheet applications such as Microsoft Excel and open Office Calc. Exporting document files such as PDF, HTML or RTF format are part of Kraken’s framework and additional features allow the creation of customized reports by appending images from the available views.

3 - Workflow automation Easily record a sequence of steps using a macro tool for reproducing daily tasks. Extend the behavior and data computations using a high-level Python API. Users can create additional grid properties and curves using their own routines. Fig. 5 - Kraken panel for creating customized reports directly from the visualization windows.

Simulation solutions for the subsurface oil and gas sector

Fig. 3 - Kraken interface for recording and playing back recorded macros.

4 - Smart properties management Kraken assembles a complete description of the simulation data, providing information such as components, phases, condition of all properties. Units are handled seamlessly, which allows changing or reconciling different metrics at no additional computational cost.

Fig. 4 - Kraken editor for listing the grid solution array information and the direct selection of units of a given XY plot axis.

ESSS has been working closely with major oil and gas companies to develop technologies for subsurface applications for over 15 years. As the leading CAE solution provider in South America, ESSS has developed a wide range of computational solutions to attend to the needs for reliable software applications in several areas, including: • Reservoir Modeling and Simulation • Basin Modeling and Simulation • Well Data Interpretation • Microstructural Characterization Cor Kuijvenhoven, Sr. Production Chemist at Shell International Exploration and Production says: “ESSS has a complete package of technology and expertise to build applications for the Oil and Gas Industry, from chemical and microbiological modeling to numerical simulation and postprocessing.” Likewise, in the field of reservoir simulation, ESSS’ expertise allows them to build applications for upscaling and geological uncertainty analyses, automatic history match, and product scheduling and steam injection optimization. "The integration environment for engineering technologies developed by ESSS has greatly simplified the simultaneous use of various commercial and in-house tools for petroleum reservoir analysis", states Régis K. Romeu, Reservoir Engineer Consultant at Petrobras.

- Newsletter EnginSoft Year 7 n°3

6 - Rich data comparison A case comparison can be created to display the differences between two models. Kraken handles merging time-steps, units and property identification, despite the type of simulator and the unit system utilized in each model. As a result of the operation, the new simulation model contains all of the features of a standard model.

ESSS and EnginSoft broaden CAE portfolios through partnership ESSS and EnginSoft are highly-innovative and wellestablished Computer Aided Engineering (CAE) solution leaders in South America and Europe, respectively. In light of their shared technical expertise, objectives and engineering services, both companies have recently created a corporate partnership to complement and expand their respective CAE solutions portfolios. Such partnership entails a diverse range of activities from project collaborations, engineer exchange programs and a joint office in Houston, Texas.

Fig. 6 - Simulation comparison example: The water saturation profile from models A and B are displayed as a side by side difference of the two grids (A-B), with the production curves and its differences in a XY plot.

7 - Powerful inspection of the available data Local inspection of grid information is available by interacting with the 3D visualization window. This feature is composed of a visual representation of the selected data and a floating panel with detailed topological and geometric information, solution arrays and non neighboring connections, which provides a straightforward way of evaluating the grid solution arrays around a selected cell.

Fig. 7 - Grid data inspection composed of a 3D visual representation of the selected cells and a list of detailed information in a floating panel.

XY curve plots can be obtained from the selection, and utilized for tracing the transient behavior of a specified grid property. Changes in the selected cell properly are updated in the XY curves plotted, thus creating an interactive environment for local transient analysis. For more information please contact: kraken@esss.com.br Contact in Italy: Livio Furlan - EnginSoft info@enginsoft.it

A technical personnel exchange program was carried out at the end of 2009 as the initial step of the ESSS/EnginSoft partnership. The primary objective of this program was to foster a collaborative exchange of expertise between ESSS and EnginSoft personnel across the various areas of their CAE portfolios. ESSS and EnginSoft engineers relocated to the EnginSoft-Padova and ESSS-Rio de Janeiro sites, respectively, for a period of three months. Through project collaboration and interactions with other engineers and technical managers, a very promising and warm collaborative climate was developed. Moreover, the successful outcome of this program serves as further incentive for future personnel exchanges. A natural step forward in the ESSS/EnginSoft collaboration is the establishment of a joint ESSS/EnginSoft office in Houston, geared toward providing CAE consulting services and software sales to the Houston area, with a specific focus on the Oil & Gas and Off-Shore industry sectors. A synergistic effort between ESSS and EnginSoft will lead to a combined portfolio of expertise which includes Finite Element Analysis (FEA), Computational Fluid Dynamics (CFD), Multidisciplinary Optimization (MDO) and software customization for Geology, Reservoir Engineering and Microstructural Characterization. The main strength of the joint operation relies on ESSS’ 15-year expertise as the leading computational simulation solution provider in South America, with offices in Brazil, Argentina, Chile and Peru, and partnerships with various universities and research centers; and EnginSoft’s 25 years of experience as a leading European Computer Aided Engineering (CAE) service provider with several offices in Italy, across Europe, and partnerships with industry and universities.

Newsletter EnginSoft Year 7 n°3 -

A Simple Parallel Implementation of a FEM Solver in Scilab Nowadays many simulation software have the possibility to take advantage of multi-processors/cores computers in order to reduce the solution time of a given task. This not only reduces the annoying delays typical in the past, but allows the user to evaluate larger problems and to do more detailed analyses and to analyze a greater number of scenarios. Engineers and scientists who are involved in simulation activities are generally familiar with the terms “High Performance Computing” (HPC). These terms have been coined to indicate the ability to use a powerful machine to efficiently solve hard computational problems. One of the most important keywords related to the HPC is certainly parallelism. Total execution time will be reduced if the original problem can be divided in a given number of subtasks which are then tackled concurrently, that is in parallel, by a number of cores. To completely take advantage of this strategy three conditions have to be satisfied: the first one is that the problem we want to solve has to exhibit a parallel nature or, in other words, it should be possible to reformulate it in smaller problems, which can be solved simultaneously, whose solutions, opportunely combined, give the solution of the original large problem. Secondly, the software has to be organized and written to exploit this parallel nature. So typically, the serial version of the code has to be modified where necessary to this aim. Finally, we need the right hardware to support this strategy. Of course, if one of these three conditions is not fulfilled, the benefits could be poor or even non-existent in the worst case. It is worth to mention that not all the problems arising from engineering can be solved effectively with a parallel approach, if their associated numerical solution procedure is intrinsically serial. One parameter which is usually reported in the technical literature to judge the goodness of a parallel implementation of an algorithm or a procedure is the so-called speedup, which is simply defined as the ratio between the execution time on a single core machine and the same quantity on a multicore machine (S = T1/Tp), being p the number of cores used in the computation. Ideally, we would like to have a speedup not lower than the number of cores: unfortunately this does not happen mainly, but not only, because some serial operations have to be performed during the solution. In this context it is interesting to mention the Amdahl’s law which bounds the theoretical speedup that can be obtained, given the percentage of serial operations (f [0,1]) that has to be globally performed during the run. It can be written as:

It can be easily understood that the speedup S is strongly (and badly) influenced by f rather than by p. If we imagine to

have an ideal computer with infinite number of cores (p=∞) and implement an algorithm whit just 5% of operations that have to be performed serially (f=0.05), we get a speedup of 20 as a maximum. This clearly means that it is worth to invest in algorithms rather than simply increasing the number of cores… Someone in the past has moved criticism to this law, saying that it is too pessimistic and unable to correctly estimate the real theoretical speedup: in any case, we think that the most important lesson to learn is that a good algorithm is much more important that a good machine. As said before, many commercial software propose since many years the possibility to run parallel solutions. With a simple internet search it is quite easy to find some benchmarks which advertize the high performances and high speedup obtained using various architectures and solving different problems. All these noticeable results are usually the result of a very hard work of code implementation. Probably the most used communication protocols to implement parallel programs, through opportunely provided libraries, are the MPI (Message Passing Interface), the PVM (Parallel Virtual Machine) and the openMP (open Message Passing): there certainly are other protocols and also variants of the aforementioned ones, such an the MPICH2 or HPMPI, which gained the attention of the programmers for some of their features. As the reader has probably seen, in all the acronyms listed above there is a letter “P”. With a bit of irony we could say that it always stands for “problems”, in view of the difficulties that a programmer has to tackle when trying to implement a parallel program using such libraries. Actually, the use of these libraries is often and only a matter for expert programmers and they cannot be easily accessed by engineers or scientists who want to easily cut the solution time of their applications. In this paper we would like to show that a naïve but effective parallel application can be implemented without a great programming effort and without using any of the above mentioned protocols. We used the Scilab platform (see [1]) because it is free and it provides a very easy and fast way to implement applications: on the other hand, the fact that Scilab scripts are substantially interpreted and not compiled is paid with a not performing code in absolute sense. It is however possible to rewrite all the scripts using a compiled language, such as C, to get a faster run-time code. The main objective of this work is actually to show that it is possible to implement a parallel application and solve large problems efficiently (e.g.: with a good speedup) in a simple way rather than to propose a super-fast application. To this aim, we choose the stationary heat transfer equation written for a three dimensional domain together with

- Newsletter EnginSoft Year 7 n°3

any loss of generality, we decided to only use ten-noded tetrahedral elements with quadratic shape functions (see [4] for more details on finite elements). The solution of the resulting system is performed through the preconditioned conjugate gradient (PCG) (see [5] for details). In Figure 1 a pseudo-code of a classical PCG scheme is reported: the reader should observe that the solution process firstly requires to compute the product between the preconditioner and a given vector (*) and secondly the The stationary heat transfer equation product between the system matrix and another known vector As mentioned above, we decided to consider the stationary (**). This means that the coefficient matrix (and also the and linear heat transfer problem for a three-dimensional preconditioner) is not explicitly required, as it is when using domain Ω. Usually it is written as: direct solvers, but it could be not directly computed and stored. [1] This is a key feature of all the iterative solvers and we together with Dirichlet, Neumann and Robin boundary certainly can take advantage of it, when developing a parallel conditions, which can be expressed as: code. The basic idea is to partition the mesh in such a way that, more or less, the same number of elements are assigned to [2] each core (process) involved in the solution, to have a well balanced job and therefore to fully exploit the potentiality of The conductivity k is considered as constant, while f the machine. In this way each core fills a portion of the matrix and it will be able to compute some terms resulting from the represents an internal heat source. On some portions of the matrix-vector product, when required. It is quite clear that domain boundary we can have imposed temperatures , given fluxes and also convections with an environment some coefficient matrix rows will be split on two or more characterized by a temperature and a convection processes, since some nodes are shared by elements on coefficient h. different cores. The discretized version of the Galerkin The number of overlapping rows formulation for the above reported equations resulting from this strongly leads to a system of linear equations which can depends on the way we partition be shortly written as (* and **) the mesh. The ideal partition produces the minimum [3] overlap, leading to the lesser number of non-zero terms that The matrix of coefficients [K] is symmetric, each process has to compute and positive definite and sparse. This means that a store. great amount of its terms are identically zero. In other words, the efficiency of The vector {T} and {F} collect the unknown the solution process can depends nodal temperatures and nodal equivalent loads. If large problems have to be solved, it on how we partition the mesh. To solve this problem, which immediately appears that an effective strategy to really is a hard problem to solve, store the matrix terms is needed. In our case we decided to store in memory the non-zero terms we decided to use the partition functionality of gmsh (see [2]) row-by-row in a unique vector opportunely allocated, together with their column positions: which allows the user to partition a mesh using a well in this way we also access terms efficiently. We known library, the METIS (see decided to not take advantage of the symmetry of the matrix (actually, only the upper or lower [3]), which has been explicitly written to solve this kind of part could be stored, requiring only half as much problem. The resulting mesh storage) to simplify a little the implementation. Moreover, this allows us to potentially use the partition is certainly close to the best one and our solver will use same pieces of code without any change, for the solution of problems which lead to a notit when spreading the elements symmetric coefficient matrix. to the parallel processes. Fig. 1 - The pseudo-code for a classical The matrix coefficients, as well as the known preconditioned conjugate gradient solver. It can be An example of mesh partition vector, can be computed in a standard way, noted that during the iterative solution it is required performed with METIS is plot in compute two matrix-vector products involving the performing the integration of known quantities to Figure 3, where a car model mesh preconditioner M (*) and the coefficient matrix K over the finite elements in the mesh. Without (**). is considered: the elements have appropriate boundary conditions. A standard Galerkin finite element (see [4]) procedure is then adopted and implemented in Scilab in such a way as to allow a parallel execution. This represents a sort of elementary “brick” for us: more complex problems involving partial differential equations can be solved starting from here, adding new features whenever necessary.

Newsletter EnginSoft Year 7 n°3 -

been drawn with different colors according to their partition. This kind of partition is obviously suitable when the problem is run on a four cores machine. As a result, we can imagine that the coefficient matrix is split row-wise and each portion filled by a different process running concurrently with the others: then, the matrix-vector products required by the PCG can be again computed in parallel by different processes. The same approach can be obviously extended to the preconditioner and to the postprocessing of element results. For sake of simplicity we decided to use a Jacobi preconditioner: this means that the matrix [M] in Figure 1 is just the main diagonal of the coefficient matrix. This choice allows us to trivially implement a parallel version of the preconditioner but it certainly produces poor results in terms of convergence rate. The number of iterations required to converge is usually quite high and it could be reduced adopting a more effective strategy. For this reason the solver will be hereafter addressed to as JCG and no more as PCG.

approach. All the models proposed in the following have been solved on a Linux 64 bit machine equipped with 8 cores and 16 Gb of shared memory. It has to be said that our solver does not necessarily require so powerful machines to run: the code has been actually written and run on a common Windows 32 bit dualcore notepad. A first benchmark: the Mach 4 model A first benchmark is proposed to test our solver: we downloaded from the internet a funny CAD model of the Mach 4 car (see the Japanese anime Mach Go Go Go), produced a mesh of it and defined a heat transfer problem including all kinds of boundary conditions. The problem has no physical nor engineering meaning: the objective is here to have a sufficiently large and non trivial n° of nodes

n° of tetrahedral elements

n° of unknowns

n° of nodal imposed temperatures

511758

317767

509381

2377

A brief description of the solver structure Table 1: Some data pertaining to the Mach 4 model are proposed in this In this section we would like to briefly describe the structure table. of our software and highlight some key points. The Scilab System Analysis JCG System 5.2.2 platform has been used to develop our FEM solver: we n° of Analysis JCG fill-in time time fill-in only used the tools available in the standard distribution (i.e.: cores time speedup speedup [s] [s] speedup [s] avoiding external libraries) to facilitate the portability of the resulting application and eventually to allow a fast translation 1 6960 478 5959 1.00 1.00 1.00 to a compiled language. 2 4063 230 3526 1.71 2.08 1.69 A master process governs the run. It firstly reads the mesh 3 2921 153 2523 2.38 3.12 2.36 partition, organizes data and then starts a certain number of 4 2411 153 2079 2.89 3.91 2.87 slave parallel processes according to the user request. At this point, the parallel processes read the mesh file and load the 5 2120 91 1833 3.28 5.23 3.25 information needed to fill their own portion of the coefficient 6 1961 79 1699 3.55 6.08 3.51 matrix and known vector. 7 1922 68 1677 3.62 7.03 3.55 Once the slave processes have finished their work the master starts the JCG solver: when a matrix-vector product has to be 8 2093 59 1852 3.33 8.17 3.22 computed, the master process asks to the slave processes to Table 2: Mach 4 benchmark. The table collects the times needed to solve the compute their contributions which will be appropriately model, to perform the system fill-in and to solve the system through the JCG. The speedup are also reported in the right part of the table. summed together by the master. When the JCG reaches the required tolerance the postprocessing phase (e.g.: the computation of fluxes) is performed in parallel by the slave processes. The solution ends with the writing of results in a text file. A communication protocol is mandatory to manage the run. We decided to use binary files to broadcast and receive information from the master to the slave processes and conversely. The slave processes are able to wait for the binary files and consequently read them: once the task (e.g.: the matrix-vector product) has been performed, they write the result in another binary file which will be read by the master process. This way of managing communication is very simple but certainly not the best from an efficiency point of view: writing and reading files, even if binary ones, could take a not-negligible time. Moreover, the Fig. 2 - The speedup values collected in Table 2 have been plotted here against the speedup is certainly badly influenced by this number of cores.

- Newsletter EnginSoft Year 7 n°3

Fig. 3 - The Mach 4 mesh has been divided in 4 partitions (see colors) using the METIS library available in gmsh. This mesh partition is obviously suitable for a 4 cores run.

performs 1202 iterations to converge. It immediately appears that the global speedup is strongly influenced by the JCG solution phase, which does not scale as well as the fill-in phase. This is certainly due to the fact that during the JCG phase the parallel processes have to communicate much more than during the other phases: a guess solution vector has actually to be written at each iteration and the result of the matrix vector product has to be written back to the master process by the parallel runs. The adopted communication protocol, which is extremely simple and easy to implement, shows here all its limits. However, we would like to underline that the obtained speedup is more than satisfactory. In Figure 4 the temperature field computed by ANSYS Workbench (top) and the same quantity obtained with our solver (bottom) working with the same mesh are plotted. A second benchmark: the motorbike engine model The second benchmark involves the model of a motorbike engine (also in this case the CAD file has been downloaded from the internet) and the same steps already performed for the Mach 4 model have been repeated. The model is larger than before (see Table 3) and it can be seen in Figure 6, where the grid is plotted. However, it has to be mentioned that conceptually the two benchmarks have no differences; the main concern was also in this case to have a model with a non-trivial geometry and boundary conditions. The final termination accuracy for the JCG has been set to 106 reaching convergence after 1380 iterations. The Table 4 is analogous to Table 2: the time needed to complete different phases of the job and the analysis time are reported, as obtained for runs performed with increasing number of parallel processes involved. Also in this case, the trend in the reduction of time with the increase of number of cores seems to follow the same law as

n° of nodes

n° of tetrahedral elements

n° of unknowns

n° of nodal imposed temperatures

2172889

1320374

2136794

36095

Table 3: Some data pertaining to the motorbike engine model.

Fig. 4 - Mach 4 model: the temperature field computed with ANSYS Workbench (top) and the same quantity computed with our solver (bottom). No appreciable differences are present.

model to solve on a multicore machine, to compare the results with those obtained with a commercial software and to measure the speedup factor. In Table 1 some data pertaining to the mesh has been reported. The same mesh has been solved with our solver and with ANSYS Workbench, for comparison purposes. In Table 2 the time needed to complete the analysis (Analysis time), to compute the system matrix and vector terms (System fill-in time) and the time needed to solve the system with the JCG are reported together with their speedup. The termination accuracy has been always set to 10-6: with this set up the JCG

n° of cores

Analysis time [s]

System fill-in time [s]

JCG time [s]

Analysis speedup

System fill-in speedup

JCG speedup

33242

2241.0

28698

1.00

20087

1116.8

17928

1.65

2.01

1.60

14679

744.5

12863

2.26

3.01

2.23

11444

545.6

9973

2.90

4.11

2.88

9844

440.9

8549

3.38

5.08

3.36

8694

369.6

7524

3.82

6.06

3.81

7889

319.7

6813

4.21

7.01

4.21

8832

275.7

7769

3.76

8.13

3.69

Table 4: Motorbike engine benchmark. The table collects the times needed to solve the model (Analysis time), to perform the system fill-in (System fill-in) and to solve the system through the JCG, together with their speedup.

Newsletter EnginSoft Year 7 nÂ°3 -

Fig. 5 - A comparison between the speedup obtained with the two benchmarks. The ideal speedup (the main diagonal) has been highlighted with a black dashed line. In both cases it can be see that the speedup follow the same roughly linear trend, reaching a value between 3.5 and 4 when using 6 cores. The performance drastically deteriorates when involving more than 6 cores probably because the machine where runs were performed has only 8 cores.

Fig. 6 - The motorbike engine mesh used for this second benchmark.

before (see Figure 5). The run with 8 parallel processes does not perform well because the machine has only 8 cores and we start up 9 processes (1 master and 8 slaves): this certainly wastes the performance. In Figure 7 a comparison between the temperature field computed with ANSYS Workbench (top) and our solver (bottom) is proposed. Also in this occasion no differences are presents. Conclusions In this work it has been shown how it is possible to use Scilab to write a parallel and portable application with a reasonable programming effort, without involving hard-to-use message passing protocols. The three dimensional heat transfer equation has been solved through a finite element code which takes advantage of the parallel nature of the adopted algorithm: this can be seen as a sort of â&#x20AC;&#x153;elementary brickâ&#x20AC;? to develop more complicated problems. The code could be rewritten with a compiled language to improve the run-time performance: also the message passing technique could be reorganized to allow a faster communication between the

Fig. 7 - The temperature field computed by ANSYS Workbench (top) and by our solver (bottom). Also in this case the two solvers lead to the same results, as it can be seen looking the plots.

concurrent processes, also involving different machines connected through a net. Stefano Bridi is gratefully acknowledged for his precious help.

References [1] http://www.scilab.org/ to have more information on Scilab. [2] The Gmsh can be freely downloaded from: http://www.geuz.org/gmsh/ [3] http://glaros.dtc.umn.edu/gkhome/views/metis to have more details on the METIS library. [4] O. C. Zienkiewicz, R. L. Taylor, (2000), The Finite Element Method, volume 1: the basis. Butterworth Heimemann. [5] Y. Saad, (2003), Iterative Methods for Sparse Linear Systems, 2nd ed., SIAM.

For more information on this document please contact the author: Massimiliano Margonari - Enginsoft S.p.A. info@enginsoft.it

- Newsletter EnginSoft Year 7 n°3

The Need for “Simulation-Quality” Material Data Material testing for simulation is about understanding how to best describe a material’s behaviour as input for the CAE code. Such testing requires expertise and experience beyond testing performed in a typical test laboratory: while the test instruments may be the same, the knowledge of CAE and experience with diverse materials is increasingly important. FEA software such as ANSYS are being increasingly used for nonlinear simulations such as those listed below. We discuss how DatapointLabs’ uncommon material expertise helps you avoid problems when the data is being generated for • Rubber hyperelastic modeling • Foam / hyperfoam and crushable foam modeling • Plastics: elastic-plastic modeling, visco-elasticity and stress-relaxation • Metals: kinematic and isotropic hardening, cyclic plasticity • Crash and drop testing: rate dependent stress-strain models • Metal forming: forming limit diagram (FLD) and spring-back material modeling • Process Simulation including injection-molding, blowmolding and thermoforming CAE More than one method to get the data Obtaining material data for non-linear FEA is not easy because the testing can be highly complicated. Hyperelastic material modeling requires testing in different modes such as uniaxial, biaxial or shear. For use in FEA, DatapointLabs performs these tests with a calibrated load cell to measure the stress, and an extensometer to measure the local strain in the gauge region of the test specimen. Some test labs measure strain using instrument displacement instead of extensometry but this brings error from the test into the FEA. Now, when tests are performed at high speeds for the

calibration of crash material models, careful instrument design is needed to avoid noise and oscillation in the stress-strain data, as presented in our paper at the NAFEMS World Congress, 2009 [1]. If noise exists, the quality of the simulation is degraded. The error here is not due to wrong methodology of testing, but the wrong choice of instrumentation. Understanding the region of interest for your FEA Rubber materials suffer damage by chain breakage during the first deformation (Mullens Effect), which results in a considerably different stress-strain behavior seen between the first pull and the subsequent cyclic loadings [2]. DatapointLabs develops data and model calibration depending on whether the initial deformation is being simulated as compared to cyclic loading. Understand the impact of the environmental conditions of your product. DatapointLabs maintains extensive facilities to test materials at elevated or cryogenic temperature, in saline (for in-vivo biomedical simulation), or other fluids-soaked environments. Understanding how well the model accommodates the real-life simulation Visco-elastic and stress relaxation data acquisition requires understanding of the complex visco-elastic theory: it can be applied only for small strain simulation, but FEA of rubber and plastics is often performed at large strains. DatapointLabs has deep expertise in applying visco-elasticity to real-life simulation. In the modeling of foams, DatapointLabs assists clients with the selection of the material model that is most suitable for the type of foam: crushable, elastic, visco-elastic or hyperfoam. [3]. This service is included with the testing ordered. Experience with diverse materials Products of today utilize an astonishing variety of materials ranging from metals, rubber, plastic, foam to films, fiber, composites, ceramics and glass. Being able to test each of these widely differing materials with the same high level of accuracy demands familiarity with such materials. DatapointLabs has tested over 18,000 materials over the past 15 years for physical properties such as tensile, compressive, shear, high strain rate, hyperelastic, visco-elastic, creep, stress relaxation, fatigue, thermal expansion and conductivity, viscosity, PVT.

Fig. 1 - LS-DYNA MAT24 Crash Material Model Calibration

Understanding material modeling and CAE As we see in the above outlined cases, the material data requirements of the various material models used in CAE are

Newsletter EnginSoft Year 7 n°3 -

often complex and unclear. It is not common for test laboratories to be familiar with CAE. With over a decade-long focus on CAE, DatapointLabs has the unique credentials required to meet the exacting demands of new product development. DatapointLabs works in direct partnership with over 15 of the world’s most prominent CAE software vendors to make TestPaks® which are packages that include the material testing, material model selection, model calibration and validation processes. The CAE user simply requests a TestPak®, sends the material sample and then receives, 5 days later, the material data plus a digital input file ready for the specified CAE. DatapointLabs online catalog offers over 150 TestPaks®. Conclusion It is clear that considerable thought and effort must therefore be paid to correct material modeling and that this part of CAE cannot be taken lightly. Certainly, universities and research institutes possess the scientific understanding to perform material testing. However, their instruments and test technicians are not dedicated to this kind of testing. Their laboratories are usually not ISO 17025 quality certified. The few cases above just serve to illustrate the nature of the problem which is quite wide-spread ranging from rate dependency [1] to process simulation [4]. The data must be clean and free from instrument artifact. It must be correct and appropriate for the simulation. Finally, the process of calibrating these material models is often error prone because, for a variety of reasons, the models cannot accommodate the observed material behavior. This lack of fidelity then results in a limitation in the ability of the model to describe the real life situation in FEA. Ordering TestPaks® from DatapointLabs reduces these risks! About the Author Mr. Hubert Lobo is a recognized leader in the understanding of non-linear material behavior, and how it impacts virtual product design. With >20 years of experience in this area, he brings valuable insights to the product development community in its efforts to design with modern day materials like plastics, rubber, foams and composites. Mr. Lobo has a Masters degree in Engineering from Cornell University. He has authored numerous articles and the

Why clients treat DatapointLabs as the expert partner for product development! Our clients have come to realize that material data used for CAE applications cannot be ordered from a material test laboratory that is not familiar with simulation. The cost for removing this important source of CAE inaccuracy is trivial compared to the risk of product failure. Time wasted by a highly qualified CAE analyst attempting to get a good simulation result with bad test data can also be much more expensive. DatapointLabs makes it easy for CAE users to get good material model calibrations for CAE in a timely and cost-effective way. • Cost savings: only the required tests are performed • Highly pertinent: properties of the actual material being simulated • Save effort: the CAE user does not waste time selecting and calibrating material models • Fast Results: data in 5 business days; a 48 hour RUSH service is available. • High Quality: DatapointLabs has been ISO 17025 certified since 2000, ensuring that the tests are performed on calibrated, traceable instruments by technicians trained to do this job correctly. • Best and cutting edge Technology Center: Online Order Placement Service at www.datapointlabs.com • DHL Sample Pickup Service from countries in Europe, 2 day express delivery to DatapointLabs! • Digital Test Data download available at www.matereality.com via Matereality Data Delivery Service. Each client gets a Personal Material Database to store their material properties on this digital platform. Handbook of Plastics Analysis. In 2002, the Society of Plastics Engineers honored Mr. Lobo, recognizing his pioneering work in quantification of material behavior for CAE. He is the founder and President of two successful companies: DatapointLabs, an expert materials testing company that generates representative properties for CAE, and Matereality, providing material database solutions for virtual product development. DatapointLabs and Matereality are based in Ithaca, New York State, USA, hometown of the famous Cornell University. About DatapointLabs and EnginSoft: DatapointLabs offers expertise for precise “Simulation-Quality” Material Data to EnginSoft and its customers in Italy as part of our Resellers Agreement with EnginSoft SpA.

Fig. 2 - Sophisticated instrumentation and expert technical staff are needed

Stefano Odorizzi, General Manager of EnginSoft: Precise material data and correct material modeling are important for our customers’ sophisticated simulation work, design and product development. We are delighted to collaborate with DatapointLabs and to offer their expertise to our customers in Italy who can now benefit

- Newsletter EnginSoft Year 7 n°3

from the company’s speedy material testing services and knowledge. DatapointLabs is a partner of ANSYS, Inc., Livermore Software Technology Corp. and the TechNet Alliance. For more information about the services in Italy, please contact: Nicola Gramegna - EnginSoft info@enginsoft.it www.datapointlabs.com - mail@datapointlabs.com References [1] "A Robust Methodology to Calibrate Crash Material Models for Polymers." Hubert Lobo and Brian Croop NAFEMS World Congress Crete, Greece. 2009.

[2] "Practical Issues in the Development and Implementation of Hyperelastic Models." Hubert Lobo and Twylene Bethard. Abaqus User Conference. 2001. [3] "Selecting Material Models for the Simulation of Foams." Brian Croop and Hubert Lobo. 7th European LS-DYNA Conference, Austria. 2009. [4] "Closing the Gap: Improving Solution Accuracy with Better Material Models." Hubert Lobo. MUG2000 2000. To download our technical papers, www.datapointlabs.com (click on Research)

please

visit:

Hubert Lobo DatapointLabs, USA

Model of a Multimass Hyperelastic System and its Parametric Identification In the description of systems with non-rigid connections, models with lumped parameters are widely used. In these cases, the number of masses usually does not exceed four and connections are represented by different rheological models. The most frequently used models are two-element models of KelvinVoigt and Maxwell and three-element models of Bingham, Shvedova. However, they are not applicable for the description of dynamics of the systems that contain links from hyperelastic materials, experiencing relevant reversible deformations. Thus, the development of a simple and at the same time highly accurate model of a hyperelastic element is a very relevant task. In the present research, the hyperelastic element is represented by two consequently connected viscoelastic bodies of KelvinVoigt with different elasticity modules and viscosity coefficients. A non-linear damper, which possesses memory, is included in one of the bodies and is parallel to the elements of Hooke and Newton. The damper is based on the generalized Bouc-Wen model of dynamic hysteresis, which is described by an ordinary differential equation of the first order. It is difficult to adjust the mathematical model of the proposed hyperelastic element. It contains 13 coefficients, most of which specify the shape of the hysteresis loop and therefore can not be measured directly during experiments. However, the model allows us to take into consideration the elastic aftereffect, the Bauschinger effect, that enhances the accuracy of the description of nonrigid systems’ dynamics. The unknown coefficients of the model are determined by parametric identification based on prior available information about their admitted region (should not contradict the physical meaning) and on the experimental data, obtained from the studied nonrigid system. The Identification process consists of solving a multiobjective optimization problem, having as constraints the inequalities with the values of models coefficients.

The optimization objectives are: • minimize the root-mean-square deviation of responses of the real system and its model at harmonic input action; • minimax Wald’s criterion of deviation of responses of real systems and its model at step input excitation; • minimize the weighted sum of values of two previous objective functions at the mixed input action. The efficiency of the proposed model of the hyperelastic element and the method of identification of its parameters was estimated using a system consisting of a DC motor, a two-stage parallelshaft reducer and a rotating mass. For their connection, hollow aluminum shafts with hyperelastic inserts in the form of rubber tubing were used. The model of this system was developed in the MatLab / Simulink package and contained 27 unknown coefficients, including the reduction ratio. Antitorque moment was accepted equal to zero; loss due to frictional forces was neglected. The optimization process was carried out in the program modeFRONTIER® with the use of the MOGA-II algorithm (multiobjective genetic algorithm with elitism). The values of velocities and angles of the DC-motor, and rotating mass were considered as responses of the system. As a result of calculations a set of Pareto-optimal solutions was defined, from which a vector of desired parameters of model was selected. When the resulting model worked out mixed input action, the error was of 4-7 % from experimental data, while the errors of a model with elements of Hooke and Kelvin-Voigt were of 16-19% and 11-14% respectively. Therefore, the proposed model of hyperelastic element and the method of identification of its parameters are highly effective and can be used to describe the dynamics of nonrigid systems with high accuracy. Denis Kozlov (MrKozlovDV@rambler.ru), postgraduate student, Department of Electrotechnics and Electrical equipment, Tula State University, Russia

!PPLICAZIONI SEMPRE PIÂ ESIGENTI RICHIEDONO ELABORAZIONI SEMPRE PIÂ VELOCI

% #OMPUTER %NGINEERING SI PROPONE COME UNA REALTA ITALIANA DI ECCELLENZA NELL INTEGRAZIONE DI SOLUZIONI DEDICATE AL CALCOLO AD ALTE PRESTAZIONI (0# L OFFERTA DI % SI BASA SU UN ESTESA GAMMA DI PRODOTTI WORKSTATION GRAFICHE SERVER STORAGE 3!. FINO AI SISTEMI CLUSTER hCHIAVI IN MANOv DI GRANDI DIMENSIONI TUTTI PROGETTATI IN BASE ALLA ESIGENZE DEL CLIENTE E TESTATI SECONDO RIGOROSE PROCEDURE PER OFFRIRE SOLUZIONI SCALABILI ED AFFIDABILI NEL TEMPO GARANTENDO IL RITORNO DEGLI INVESTIMENTI SULL HARDWARE

% #OMPUTER %NGINEERING EXCELS AT INTEGRATING SOLUTIONS FOR THE (IGH 0ERFORMANCE #OMPUTING (0# % S RANGE INCLUDES A BROAD SELECTION OF PRODUCTS FROM COMPUTER GRAPHICS 7ORKSTATIONS TO SERVER STORAGE 3!. UP TO POWERFUL CUSTOM BUILT CLUSTER SYSTEMS EACH ONE OF THEM DESIGNED FOLLOWING THE CLIENT S REQUIREMENTS AND TESTED ACCORDING TO STRICT PROCEDURES IN ORDER TO PROVIDE SCALABLE SOLUTIONS WHICH ARE RELIABLE EVEN AS TIME GOES BY AND GUARANTEE A PROFITABLE RETURN ON HARDWARE INVESTMENTS

% COMPUTER ENGINEERING 3 P ! 6IA -ARTIRI DELLA ,IBERTÜ 3CANDIANO 2EGGIO %MILIA )TALIA 4EL &AX WWW E COMPANY COM E MAIL INFO E COMPANY COM

- Newsletter EnginSoft Year 7 n°3

An Interview with Mr Nazario Bellato, Simulation Manager of Magneti Marelli Powertrain

Intervista con Nazario Bellato, Simulation Manager di Magneti Marelli Powertrain

Magneti Marelli Powertrain is Magneti Marelli’s business unit dedicated to the development and manufacturing of engines and transmission components for cars, motorbikes and light vehicles. Today, Magneti Marelli Powertrain supports four application centers and eleven manufacturing sites on four continents. During our recent FEM Interview Tour in Italy, we had the pleasure to meet Mr Nazario Bellato at Magneti Marelli Powertrain. Mr Bellato’s expertise is of the highest standard, he holds the position of Simulation Manager and is a Veteran User of ANSYS ( since 1993). A key person in the Calculation Department at Magneti Marelli Power Train, Mr Bellato’s ambition has always been to be innovative in every respect, in particular within the company’s European Projects ( 11 European and 5 American patents). Most of his technical expectations from the simulation community are summarized in the following interview. Mr Bellato is a source of knowledge and inspiration in the Ansys Italian Advisory Group! Mr Bellato has a University degree from Università Politecnica delle Marche, he started his career working in applied research for Indesit. After an extremely successful cooperation with Giorgio Fua, a famous Italian economist, Mr Bellato joint Fiat in 1994. This was also the time when he started his work with finite element applications.

Laureato presso l'Università Politecnica delle Marche, ha iniziato il percorso professionale nell'ambito universitario della ricerca applicata a livello europeo e internazionale (Gruppo Indesit). L'incontro nel 1993 con l'economista Giorgio Fuà e la sua scuola di specializzazione (ISTAO, Istituto Adriano Olivetti per la Gestione dell'Economia e delle Aziende), l'ho portano in contatto con il Gruppo Fiat e, in particolare il Centro Ricerche Fiat, dove entra a farne parte nel 1994. Attualmente è il responsabile dell’Ente Calcoli e Simulazione presso Magneti Marelli PowerTrain. Ha al suo attivo 14 brevetti europei e 5 estensioni negli Stati Uniti. Varie soluzioni sviluppate, e relative varianti, sono attualmente in produzione.

Air Intake Manifold Development

1. Che spazio ha (e dovrebbe avere) l’innovazione nel mondo industriale/impresariale? La globalizzazione e le nuove tecnologie di comunicazione hanno avviato e consolidato una rapida trasformazione dei mercati e dei businees correlati. I cicli di vita prodotto che a volte duravano decenni oggi possono durare anche mesi. Questo fa si che l'innovazione sia diventata un fattore dominante per una azienda che vuole sopravvivere o espandersi alle mutevoli regole internazionali e soprattutto per le aziende del "vecchio continente": garantire livelli coerenti della qualità della vita per i propri dipendenti e ragionevoli

Newsletter EnginSoft Year 7 n°3 -

Simulation is the key to state-of-the-art engineering and manufacturing…

1. What is or should be the role of innovation in the industrial and entrepreneurial world ? In the last years, globalization and new communication technologies have created an extremely fast market; business transactions and our business culture evolved in a similar way. In the past, the life cycles of products spanned years, now they are realized in months and in some cases even in weeks. Obviously, the innovation of products has become the main growth factor and is crucial for a company that aims at growing its product range into increasingly competitive markets. Today, we can observe continuous changes in the global market, it’s a very fast and dynamic world out there. Among each company’s goals should be to maintain good working conditions for its workers and to create reasonable profits for its shareholders. Innovation is an important driver for the culture, health and future of a company. New ideas will only lead to successes and competitive advantage when they are transformed into competitive industrial solutions which grow the company turnover in the end. 2. Which are the strategies to be innovative and what are the actions needed to realize innovation? Innovation is usually achieved at the end of a complex development process which involved and counted on various internal and external competencies. Each competence contributes to a unique final result. Strong competitive advantage is based on many reliable technical results. When we want to develop and create an optimized new product, we know that it is essential to examine the same from different points of view: technical performance, organization, product processes, market strategies… In this interview, I would like to focus on the technical aspects. As a first step, a specific strategy has to be defined in order to concentrate and guide the company’s energy. We have to know the reasons why we want to innovate a product and it has to be clear what results we want to obtain. At this point, the managers look at the company’s resources, the final results and then determine the right tools that should lead to those results. The Human Resources responsible has an important role as he/she coordinates the assignment of the technical experts und support teams and in this way, determines efficiency, work load and success of a project. The General Manager and owner carry key responsibilities because they communicate the mission of a company and important projects, especially in product development. The right words and actions motivate and inspire excellence in the team all the way through to the final results. I can say from my experiences through the years that to know, expand and believe in the company’s competencies is key to success and will secure the company’s future.

The block diagram represent the phases for the development of the progressive bore: Before “cutting” the aluminium we made some FEA analysis in order to find a “possible” solution. This process usually take 2 or 3 attempts The final adjustments are made on real parts. Also this phase usually needs 2 or 3 loops to reach the correct bore shape.

profitti per i propri investitori. L'innovazione è figlia della cultura aziendale. Le idee innovative da sole non servono per generare vantaggio competitivo ma solo se generano soluzioni innovative. Le idee anche le più brillanti da sole sono delle "scatole vuote": devono essere in grado di generare soluzioni industriali utili a produrre profitto. 2. Quali sono le strategie per essere innovativi e quali valutazioni spingono all’innovazione? L'innovazione è il risultato di un complesso processo di sviluppo a cui contribuiscono competenze diverse interne o esterne all'azienda, volte ad un unico obiettivo: ottenere nuovi vantaggi competitivi. Ci sono molte vie per innovare: prodotto, organizzazione, processo produttivo, strategie di mercato, etc. in questo ambito parleremo di prodotto. Innovare sul prodotto significa aver scelto a priori una strategia specifica dove far convergere le proprie energie, ciò equivale a chiedersi perché vogliamo innovare e a quali risultati vogliamo convergere. Presa coscienza della scarsità delle risorse, nel senso che non ci sono mai risorse sufficienti per muoversi in ogni direzione, è indispensabile individuare con certezza gli obiettivi da raggiungere. Al centro comunque ci sono le persone: sono necessarie figure professionali particolari e Team ben coordinati: efficienti, efficaci e soprattutto capaci di gestire lo stress, il rischio e soprattutto gli insuccessi. Inoltre per aver successo ovviamente è necessaria la

- Newsletter EnginSoft Year 7 n°3

3. Which are the right positions covered by CAE and Virtual Prototyping? It is well known that virtual analysis ( CAE) is a fundamental tool for time-to-market TTM reduction. The FEM approach helps to control development costs and ultimately specific business and product goals. I would recommend to define specific accurate DOE plans in order to significantly reduce prototyping time and to substitute the same with a VIRTUAL prototyping step. Obviously, the advantage is to simplify the design with six sigma, considerable improvements in product quality will be the results. The approach to reduce experimental time while defining investments and production lines at the same time is probably the best solution to achieve a profitable final product. 4. How did the user’s demand change in the last year? From the 90th until today, the skills of numerical analysts have consistently evolved; in fact, in the past, they used to be advanced computer experts because to create a finite element model was considered “art” If this “wizard” was also a Unix expert, he/she would improve the company’s technical level. Nowadays however, a lot of things have changed. For example, it is possible to work within integrated environments where the core activity is not to have a mesh model but to have a more accurate realization of a simulation of the real physical phenomena and hence to have the correct prediction of all mechanical, CFD and electromagnetic variables to optimize the final product. 5. What benefits did you have in your professional career so far and how did the move to design and production happen? The wide and correct use of new technologies brings us the opportunity to have a global vision of the whole project and the driving of all development forces is an important new capability in our technology box. Coupled analysis now makes it possible to make fast decisions for a project and at the same time, new technologies help us in being more accurate with our decisions in the different processes. 6. How has EnginSoft increased the value, the quality and the capability of your company? EnginSoft is not a just a software provider, we regard EnginSoft as a pro-active technology branch of our Magneti Marelli Powertrain unit. The EnginSoft team has an important role as ‘ Knowledge broker’. The technical standards and levels of EnginSoft are transferred to our company. In fact, EnginSoft has a well-established reputation which is based on their flexibility to adapt their offer, services and expertise to the dynamics and requirements of the market. For example, EnginSoft’s concept of project chain management involving different techniques is second to none. When we discussed new frontier applications, their technical experiences to run simulations which deliver usable results were of utmost importance. We greatly value EnginSoft’s concept to offer a single solution for a single technical problem, but to include

chiara sponsorizzazione della Direzione o della proprietà. Se il mix è ben posto i risultati prima o poi arrivano e spesso sono l'ancora di sopravvivenza o di successo dell'azienda. 3. Che ruolo ricoprono gli strumenti CAE e di prototipazione virtuale in tal senso? Nel ciclo di sviluppo prodotto le analisi predittive (CAE) sono uno degli strumenti fondamentali per ridurre il TTM e mantenere i costi di sviluppo e industrializzazione in linea con gli obiettivi di business. La possibilità di ridurre drasticamente la fase prototipale in funzione di un giusto mix con la prototipazione virtuale permette di definire piani DOE molto più accurati e semplifica il design for six sigma, con notevoli miglioramenti in ottica qualità. La possibilità di ridurre le fasi prototipali e rilasciare gli investimenti in tool e linee di produzione solo nei momenti definiti, diventa in alcuni settori la chiave per avere prodotti profitevoli. 4. Come sono cambiate le esigenze degli utilizzatori negli ultimi anni? Si è passato dall'analista numerico anni '90 che spesso era un mago di informatica perché preparare modelli era veramente un'arte da sartoria e conoscenze unix alla possibilità di lavorare in ambienti integrati dove il core dell'attività non è la mesh ma la rappresentazione, il più possibile coerente del fenomeno fisico in funzione degli obiettivi dell'attività e la corretta interpretrazione dei risultati. 5. Quali vantaggi ha rilevato nella sua esperienza professionale e come è cambiato il suo approccio alla progettazione/produzione? Ora con le nuove tecnologie si ha una visione più di insieme quindi la possibilità di guidare correttamente le energie di sviluppo. La possibilità di poter effettuare analisi accoppiate, l'accesso a tecnologie che per costi o

one-way Fluid Structure Interaction using CFX and Ansys Mechanical

Newsletter EnginSoft Year 7 n°3 -

experiences and knowledge to establish a defined company project workflow. 7. What are the perspectives of computational codes in view of the future challenges of the market? These codes will dominate development cycle processes. Although it is important to keep in mind that a lot of information will have to be managed and this will require considerable human knowledge. Finally, I would like to mention that it is important to be sure about the results of any simulation. This is a complex problem which concerns model quality, input parameters, the modeling of material properties and more…..It is crucial to manage the problem in a consistent way and to keep the economics in mind as well: for example, to upgrade hardware to be able to exam more complex models and to have consistent information and useful methods. In a situation like this, it is only natural to talk to a partner (like EnginSoft) who is able to provide systematic support and competencies in the use of different computational software for the various application areas in engineering. 8. For which projects will you foster the wide use of these tools? I hope to succeed in realizing my strategy in the company which will help to quickly answer to any new business opportunity by providing a robust, flexible and modular solution to create profit instead of small volume products.

9. What are your wishes for the scientific technology world that is searching the right ratio/mix between competition and creativity? This deserves a simple answer: Competition and Creativity are great ingredients for Innovation! It is very interesting to look at the actual general methodology that Magneti marelli has in its primary development project phase: They usually work for FPT, PSA, RSA, GM Dailmer, BMW, VW, Suzuki but it is not possible to look at their specific images but it is extremely remarkable to have a synthetic vision of their work-flow project completly and strongly closed in Ansys workbench philosofy. For more information: Roberto Gonella - EnginSoft info@enginsoft.it

tempi di implementazione in passato non erano utilizzabili in ambito industriale, permette di essere più efficienti ed efficaci sulle scelte progettuali. 6. Qual è stato il contributo di EnginSoft e in che modo ha saputo valorizzare qualità, potenzialità e capacità della sua industria/impresa? Enginsoft è stato sempre per noi non un fornitore ma una estensione della nostra azienda. La competenza dei suoi tecnici e la flessibilità aziendale della società ad adattarsi alle mutevoli condizioni di mercato, sono stati i fattori chiave che hanno reso solida la collaborazione nel corso degli anni. 7. Che prospettive intravede per i codici di calcolo in relazione alle sfide poste dal futuro? Diventeranno sempre più parte dominante del ciclo di sviluppo prodotto ma non dobbiamo dimenticare che la differenza sarà fatta sempre più dall'uomo nella gestione e utilizzo delle maggiori infomazioni a disposizione. Il vantaggio competitvo sarà dato soprattutto dalla preparazione culturale e tecnica delle nuove generazioni, ma questa è un'altra storia... 8. Quali progetti, obiettivi e nuovi traguardi intende perseguire grazie all’uso di questi strumenti? Avere la possibilità di rispondere rapidamente alle nuove opprtunità di business con soluzioni robuste, flessibili e

This methodologies enables to import fluid forces, temperatures, from a steady-state CFD analysis into a the Mechanical application analysis. This one way transfer of temperatures information from a CFD analysis can be used in determining the temperature distribution on a structure in a steady-state analysis.

modulari capaci di rendere profitevoli anche bassi volumi di produzione. 9. E cosa si auspica per il mondo della tecnologia scientifica alla continua ricerca di una dimensione tra creatività e competitività? Che in futuro le due parole siano usate entrambe come sinonimo di innovazione Per maggiori informazioni: Roberto Gonella - EnginSoft info@enginsoft.it

- Newsletter EnginSoft Year 7 n°3

Illumination Analysis and Design Optimization of an Automotive Speed Meter DENSO CORPORATION Japan uses modeFRONTIER® as standard tool for optimization and much more On the display panel in front of the with approximately 120,000 vehicle driver, the instrument employees. In Europe, including Italy, cluster conveys a wide variety of 33 regional DENSO offices and information, such as driving speed, factories are based. DENSO’s main motor rotation number and fuel business is to develop and provide level. These values are shown advanced automotive technologies, clearly on the displays of the panel systems and components, such as, for so that the driver can recognize example, powertrain control systems, each driving condition of the electric systems, electronic systems, vehicle by just one look. For an thermal systems and information & instrument cluster, a high level of safety systems for the world's major visibility and a stress-free display automakers. The automotive speed for long drives that match the meter described above is one of vehicle’s design are required. In DENSO’s main products. As one of the particular, high-luminance and big players in the global automotive consistent illumination are sector, DENSO’s focus is on delivering Fig. 1 - Vehicle interior (above) and meter illumination necessary for the meter display to (below) the highest quality speed meters provide good visibility for highest while shortening development cycles. safety. In fact, the illumination quality is the most Design and development processes continuously have to important design requirement for the automotive speed be adapted and made more efficient. meter. To shorten its product development cycle and to deliver a high quality product to the market as fast as The challenge possible, have become the biggest challenges for DENSO For years, DENSO has done illumination analyses using 3Das global competition in the automotive industry has CAD data to determine the illumination design of increased over the past years. automotive speed meters. DENSO is a world famous automotive parts supplier located in Kariya in the Aichi prefecture region of Japan. The company operates in 33 countries and regions

Fig. 2 - 3D-CAD data of the meter assembly (above) and the pointer (below)

The purpose of the illumination analysis outlined in this article is to predict the illumination brightness and unevenness of the meter by calculating the luminance distribution on the speed meter dial and pointer, and the ray tracing from light sources, and moreover, to design the optimal meter geometry by changing the 3D model in the CAD system as necessary. However, this trial and error design process, to modify the 3D model every time with

Fig. 3 - Luminance distribution result (left) and virtual display (right) obtained by illumination analysis software. Speed meter dial (above) and pointer (below).

Newsletter EnginSoft Year 7 n°3 -

Fig. 4 - modeFRONTIER® Workflow: a)Input setting; b) CAD geometry change and illumination analysis (automated batch process); c) Results output setting

the illumination analysis results, took up considerable time until the required quality was achieved. Typically, there were 4 to 8 design parameters for 1 pointer geometry. Hence it was not easy, even for the experienced engineers, to choose the right design parameters from a number of possible combinations of these parameters so that the 2 different objective functions, average luminance and luminance ratio, could reach an optimum level. Actually, about 10 iteration for one pointer were necessary and the design process took 7 – 10 days for 1 product. The type of work was mostly routine and increased the engineers’ workload. Moreover, the process always relied on the know-how and understanding of the engineers. Therefore, in order to ensure high quality product development for the future, it became extremely important to improve the existing process. To realize time reduction and optimization of illumination quality, DENSO has established a design optimization system for automotive speed meter using modeFRONTIER®. The solution The modeFRONTIER® multi-objective optimization tool and its experimental design methods, were embedded to fully automate the repetition of the 3D design changes following the illumination analysis results. The design process flow to be automated was the following: 1. 3D-CAD data (NX) was translated into IGES and imported into the illumination analysis software. 2. After the meshing was completed and each boundary condition defined, the software calculated the illumination distribution and the result was exported as the input file for the design change inside the 3D-CAD system. To establish the automatic system for this flow, the workflow of the automatic calculation process was created inside modeFRONTIER®. This way, the batch program to change the CAD geometry and to execute the illumination analysis was determined.

As modeFRONTIER® provides an easy-to-use Japanese GUI, the initial settings for the automatic process could be defined very easily even by engineers who were unfamiliar with the software at the time. Then, the actual optimization flow started. The first step was the illumination analysis using modeFRONTIER®’s experimental design method. With this method, the highly accurate approximate function could be obtained using limited calculation time. Then, the optimized result was searched to lead the Pareto front. The challenge was the multi-objective optimization showing the trade-off between the 2 objective functions. So, choosing the most efficient algorithm from the many different multi-objective optimization algorithms that modeFRONTIER® provides, to explore the optimized result effectively, was crucial to establish the new system in DENSO.

As mentioned before, the number of parameters is relatively high with 4 to 8 design parameters and 2 different objective functions of average luminance and luminance ratio. Here FMOGA (Fast Multi-objective Genetic Algorithm which executes the multi-objective optimization by updating the response surface automatically) was selected as it has a wider search range and the Pareto front can be reached quickly. FMOGA has the ability to reduce the actual calculation work by using the response surface and to drastically downsize the total calculation time. Before using FMOGA, it was evaluated by changing the approximate rate by 70%, 80% and 90%, in order to know which rate is the most valid for the response surface, which can explore the better result, and if the required calculation time is reasonable. The goal is within 24 hours. For the final result, an 80% approximate rate was chosen finally as it delivers better performances in a suitable time frame. Results A1 to A6 in Fig.6 are the 3D-CAD design parameters for the pointer geometry and represent the pointer angle, the angle side of the reflecting surface, the angle above the reflecting surface, the angle below the reflecting surface

Fig. 5 - FMOGA evaluation by changing the approximate rate of the response surface

- Newsletter EnginSoft Year 7 n°3

Conclusions The automatic optimization process managed to: • Automate the routine between CAD data change, the illumination analysis and optimization. • Streamline the whole design process by reducing workload and process automation. • Improve product quality by switching from a trial and error solution to theoretical optimization. The general-purpose geometry optimization system can be easily used by the designers and will be introduced not only for the meter design, but also for other optical products of DENSO in the months ahead. Today, modeFRONTIER® is DENSO’s standard tool for optimization and it is expected to support each business area of the company.

Fig. 6 - The pointer geometry (above) and the design parameters (below)

and 2 different heights. For each design parameter, the minimum value, the middle value and the maximum value were defined. The calculation point for the luminance was determined as shown in Fig.7. The optimization was executed in order to find the combination at which both the average luminance and the luminance ratio will become larger.

This article is based on the original case study by Mr. Chiaki Suzumura, DENSO CORPORATION, Japan. The article has been written in collaboration with CDadapco JAPAN Co.,LTD.

Akiko Kondoh, Consultant for EnginSoft in Japan

Fig.8 shows the results of the optimization solution and the Pareto front. The red point represents the result before the optimization system was introduced. The pink point is the optimum result gained from the Pareto front. Though the average luminance is 158 which is the same as before, the luminance ratio has improved by 15% from 0.64 to 0.72.

Fig. 7 - Objective functions (average luminance and luminance ratio)

Yet, not only the luminance result, but also the man hour could be improved significantly. The process which used to require the designer’s constant attention over 8 days in the past, now, after the introduction of the optimization software modeFRONTIER®, only requires 8 hours of manpower for operation and a total of 2 days including the automatic calculations. This means the workload could be reduced by 90% with the help of optimization methods.

Fig. 8 - The Pareto front of the average luminance and the luminance ratio

Newsletter EnginSoft Year 7 n°3 -

Elysium’s CADdoctor Enriches Product Data Quality in PLM 3D interoperability is based on 3D data translation. When 3D CAD data is converted into another format, people expect that not only 3D geometry but also other information such as attributes and annotation is converted perfectly. However, it is a well known fact that some kinds of geometry or information often fail to be converted, or cause errors. To transfer full information

translation, for example, between Japanese and English. Since the grammar and vocabulary are different and some terms don’t have exact equivalents in the other language, it is almost impossible to translate completely. It is not surprising that translation errors of 3D data often occur as in the case of language. Even if you created 3D models that look the same on several different CAD systems and converted them into IGES files, they would make differences. As the English language has various dialects, IGES files converted from different CADs are described in different ways, just like words spoken in different dialects. Dialects also have an effect on translation accuracy.

Fig. 1 - Objects to be translated

from one CAD to another or to other formats for FEM, CAM, RP or DMU, you have to pay more attention to the Product Data Quality (‘PDQ’). Elysium’s ‘CADdoctor’ provides the right solution to leverage 3D CAD data. 3D Data Conversion Methods On a superficial level, conversion methods are divided into ‘Direct’ and ‘Mediate’ translation. ‘Direct’ translation means that user can exchange, read and write an original CAD data among two or more CAD systems. On the other hand, ‘Mediate’ translation is a conversion using an intermediate file format like IGES, STEP or special formats provided by vendors (*). Indeed, those two methods are internally equivalent. Here is how an intermediate file acts and why we need to care about PDQ (**). Errors Caused by Format When you convert a 3D CAD model into another format via an intermediate file, defective geometries are often found. This is because the representation of a 3D geometry is different from CAD to CAD. It is like “language”

Fig. 2 - Wrong 3D data remain wrong after translation

The other reason for errors Nowadays, the auto industry is trying to standardize the notation in IGES to prevent PDQ errors, and each CAD vendor has taken countermeasures against such problems as well. STEP is designed to eliminate ambiguity over 3D data representation. Yet unfortunately even STEP cannot solve all the problems accompanied with 3D data conversion. If original 3D data contains unneeded information or lacks sufficient information, wrong information will be kept in errors throughout the data conversion. As a matter of fact, it is evident that defective 3D geometry, that is lack of PDQ, causes problems later on. Errors caused by Bad PDQ Converting poor-quality 3D data often results in failure. While some apparent errors such as a gaping hole on a surface, face distortion or an untrimmed face are easy to find, many invisible errors are prone to stay undetected. Therefore, an innocuous-looking 3D model can cause problems; for example, you cannot execute operations anymore because the model is not a solid model, or errors

- Newsletter EnginSoft Year 7 n°3

high-accuracy CAD, the gap between faces is considered as ‘Gap’ as the width is larger than the threshold.

Fig. 3 - a missing face in a conversion result

This is the same in the case of ‘Tiny Element’. You can create a microscopic element on high-accuracy CAD. For a precision-level 0.001mm CAD system, a 0.002mm element is sufficient for closing a gap. But when transferred into Fig. 4 - Example of hole in polygon mesh due to bad PDQ another low-accuracy CAD, the element would be recognized as an unusable ‘Tiny element’. Other than tolerance matters, the way to handle analytical representation or nonmanifold varies from CAD to CAD. Major PDQ guidelines recommend generally acceptable values, though it is impractical. For example, while popular PDQ guidelines recommend converting the analytical representation to a generic NURBS surface, a cylinder must be represented as an analytical surface for a kind of motion simulation analysis tool.

Fig. 5 - Typical cause of PDQ deterioration

occur during Boolean operation or offset. In extreme cases, CAD software freezes in the middle of modelling after repeated reworks because of accumulated PDQ problems. 3D data with poor PDQ leads to troubles not only with CAD but downstream FEM analysis, digital mock up (DMU), rapid prototyping (RP) or CAM.

Product Data Quality Guideline ‘PDQ’ literally means the quality of 3D data for product development and manufacturing. To ensure quality control, you will need some criteria and a guideline to judge the quality. One of the most popular guidelines is

Factors that make the PDQ worse Usually, CAD modellers and operators of data conversion don’t pursue the PDQ. But improper 3D geometry is generated at every stage from design to distribution of 3D data. The most common cause of PDQ problems is that each CAD system has unique standards. Even if an original 3D model does not have PDQ errors in the original CAD, the converted model may have PDQ errors in the target CAD because of the difference in accuracy criteria and tolerance. (***) For example, if you create a model in a low-accuracy CAD, a slight gap between faces whose width is smaller than the threshold value is considered as just an edge. And, once the model is converted to

Fig. 6 - Different judgement on ‘Gap’

Fig. 7 - Different judgement on ‘Tiny Face’

Newsletter EnginSoft Year 7 n°3 -

Fig. 8 - Typical examples of poor-quality modelling

the Product Data Quality Guideline for the Global Automotive Industry set by Strategic Automotive product data Standards Industry Group (‘SASIG’). Document Version 2. 1 released in May 2005, which has been also published as ISO/PAS 26183:2006, is widely accepted as a standard.

that the surface is generated when it is exported to IGES format. Embedded Face stands for multiple edges and faces that are overlapping in whole or part. This kind of error is often caused by reworks in the design phase. Copied geometry for a revision remains as an embedded face. Recommended software for PDQ validation and repair There is a remedy for everything. If you are willing to solve the problems regarding PDQ and willing to circulate high-quality 3D data, why don’t you take effective measures to evaluate the PDQ and to correct problems? Elysium’s CADdoctor is one of the most reliable software for PDQ validation and 3D data healing.

Although it is a very big self-intersection, it looks normal when you see it in a CAD window because the trimmed face itself does not have any problems. The base surface has self-intersection outside the trimmed area. It seems

Although there are various healing tools, few can automatically correct errors according to both popular PDQ guidelines and user-defined standards. Through long-term on-site trials with a number of manufacturers, CADdoctor has proven unparalleled performance of detecting and healing errors. Adopted and highly praised among CAD-using industries, it ensures strict compliance with PDQ guidelines and/or specific company standards. Healing requires an extremely high degree of geometry interoperability. CADdoctor allows its users to repair damaged geometry with very simple and quick operations.

Fig. 10 - Embedded Face

Fig. 11 - Invalid 3D geometry detected by Elysium’s CADdoctor SX

Fig. 9 - Self-intersecting Surface

SASIG’s PDQ guideline contains Geometric Quality Criteria that consist of 64 check items and Non-Geometric Quality Criteria which define file naming conventions, data structure including layer or assembly and so on.

- Newsletter EnginSoft Year 7 n°3

with PDQ errors. For FEM, CADdoctor also provides powerful optimization including geometry simplification and polygon handling functions. (****) In addition, ASFALIS, Elysium’s flagship solution, will help automating the entire process. For more information, please visit the ELYSIUM website: http://www.elysium-global.com

Fig. 12 - Some healing tools worsen PDQ rather than improve.

Sakae Morita, ELYSIUM Co.,Ltd., Japan

Tips for Practical Use of CADdoctor Product development is divided into two phases; creation of 3D data and utilization of 3D data. Creation of 3D CAD data Ideally, the original 3D CAD data created during the design phase should have no errors inside and should comply with common PDQ guidelines. CADdoctor is a desirable option for automatic detection and correction of PDQ problems.

(*) For example, Elysium Products including CADdoctor SX, CADdoctor EX and ASFALIS provide Elysium Neutral File as the intermediate file to users. (**) Please note that the conveyable information differ from each intermediate file. Especially, attribute information and 3D annotations tend to be neglected.

Fig. 13 - CADdoctor Healing Example [Original (NX) -> Dr (before) -> Dr (After) -> Result (V5)]

Fortunately, most file formats can convey 3D geometry information.

Fig. 14 - 3D data distribution with PDQ tool

Utilization of 3D CAD data As the tolerance and other standards differ from application to application, engineers in the FEM or experimental stages have to prepare 3D CAD data in accordance with the intended use. Regarding PDQ, if data is not healed at all in the design phase, the data will cost considerable man hours to repair errors with CAD operation. On the other hand, if data is validated and repaired in the design phase, it will need just a few minutes to deal

(***) PDQ items about tolerance were originally classified as Geometric Quality Criteria. However, such items are practically subject to the rules of an individual company or set between the parties concerned. So, the items are classified as Non-Geometric Quality Criteria. In this article, the term ‘Tolerance’ means identical tolerance. If the distance between two vertices is within the identical tolerance, the CAD system recognizes that vertexes are at the same place. (****) Regarding CAE, a complex CAD model can cause considerable damage to FEM analysis, such as generation of irregular mesh, longer calculation time, and analysis errors. The CADdoctor SX FEM Suite will help you as an excellent tool for downsizing CAD files to produce high-quality mesh data for FEM by automatically repairing the PDQ defects and removing unnecessary features. Mesh generator is needed separately. Fig. 15 - CADdoctor helps to create regular mesh for FEM

Newsletter EnginSoft Year 7 n°3 -

The Culture of Wood rigid structures made from concrete. Owing Wood is very close to the people in Japan, to its flex structure and central pillar, the where two-thirds of the land is covered by Pagoda can absorb vibrations and has forest, and wood has been applied to withstood major earthquakes in the last people’s life naturally since the early times. thousand years. Today, the same technique Geographically, Japan stretches a long way is employed by skyscraper architects, not from the northeast to the southwest. Within only in Japan but also in many other short distances, the country has incredible countries worldwide. People’s skills and altitudes and changing landscapes. This is knowledge of how to use the individuality of also why we can find many different types of wood and this architectural method, have trees in the country. Today, not only metallic evolved further during the last centuries. materials but also lightweight and high Japan has developed many new materials strength composite materials have been and techniques, but people still love developed. Such high-performance materials wooden architectures and most houses are are used in different industries, from built from wood. aerospace, automotive, energy, environmental to bio engineering and entertainment, etc. At a time when we are Image1: Gojyu-no-To, five-story Pagoda in Wood can be seen in many different MONODUKURI. One of the reasons is that offered such sophisticated new materials Horyu-ji the latest processing technologies are able to produce many from science and technology, we once again appreciate wood wooden materials with even characteristics that are heat and for the natural beauty, warmth and gentleness it brings to humidity resistant. Wood is now applied to a wider range of our homes and workplaces. products. Another reason is that traditional wooden products made by old manufacturing methods are still popular, and Unlike metallic and composite materials, wood is a living their popularity is growing ! These products unify in harmony material with individual characteristics. Sometimes, different techniques and craftsmanship with the natural engineers and designers face difficulties when using wood as material that people admire, and that we can’t expect from a material. It erodes under the influence of humidity, and products made by machines. becomes distorted and cracks from drying. Its strength and durability depend on the position and direction of use. But For example: Hashi (chopsticks), Wan (bowl), Ohitu, wood also has many benefits as a material. Wood provides Suribachi (mortar) and Seiro (steamer) are some of the heat insulation, moisturizing and good humidity capabilities. typical wooden tools that many people use at home. Its natural soft look and scent gives us a feeling of wellRecently, Muku (unprocessed wood) has become popular for being. With all this in mind, the Japanese have produced flooring and furniture. Wood is also used after the burning beautiful and functional wooden goods, making up process, as Mokutan (charcoal). disadvantages of the material and revealing its full beauty. Famous examples can be found in Japan’s traditional It is used for char-grilled cooking like Yakitori (grilled architecture. In the Nara prefecture (region in the center of chicken), it turns tap water into mineral water, and thanks to Japan, on the Kii Peninsula) people celebrate this year the its humidity condition, it deodorizes our homes with its air 1300th anniversary of Heijo-kyo, the prefecture’s capital. cleaning capabilities. It also protects Heijo-kyo is home to many traditional wooden computers and electric devices from architectures, some are listed as World Heritages by electromagnetic rays. Unesco. Horyu-ji and Todai-ji are well known all over the world as the world’s oldest and biggest We appreciate the living material wood for wooden architectures. Gojyu-no-To (five-story its natural beauty, it has become a treasured Pagoda) in Horyu-ji exists for 1300 years without companion in many areas of our lives. To leaning or breaking despite the wooden structure, people in innovative manufacturing its delicate figure has not changed since the day it industries using CAE, Wood MONODUKURI was finished. Gojyu-no-To was designed and erected will bring a fresh understanding and based on an old architectural method of piling the enthusiasm for nature… flex structure joined wooden parts using a special assembling technique. Of course, this is very Image2: Ohitu (bowl for keeping Akiko Kondoh different from the approaches architects use for cooked rice)

- Newsletter EnginSoft Year 7 n°3

EnginSoft Joined the E2BA On the 8th of July, 2010, the General Assembly of the E2BA (Energy Efficient Buildings Association), held in Brussels, admitted EnginSoft S.p.A. as a new member of the association. The new membership to the association will allow EnginSoft to extend its research network and to have a proactive and collaborative role across the building industry value chain, furthermore, EnginSoft will reinforce its active role in the research activities related to the construction industry. EnginSoft, BENIMPACT and the sustainable buildings industry Recently the EnginSoft R&D team strongly focused in the eco-sustainable buildings design sector, thanks to the development of its applied research project BENIMPACT (Building’s ENvironmental IMPACT evaluator & optimizer). The project is co-funded by the autonomous Province of Trento (Northern Italy) by means of the ERDF (European Regional Development Fund). BENIMPACT mainly aims at the development of methodologies (and of a related prototypical software platform) to support architects and engineers in the design of ecosustainable buildings (both new ones and modifications of the existing ones). Based on the integration of various analysis tools within the Process Integration and Design Optimization platform modeFRONTIER®, the BENIMPACT suite will allow to identify the optimal trade-off between costs and environmental performances of the buildings. please get more information on the BENIMPACT research project from: http://www.enginsoft.com/research/prgbenimpact.html The E2B Association The E2BA (Energy Efficient Buildings Association) was created in the 2008 by the founding members of the E2B EI (Energy Efficient Buildings European Initiative) as a non-profit, international, industrial association. The E2BA was created in order to prepare and manage a PPP (Public-Private Partnership) with the European Commission, seek and demonstrate industry engagement and represent and coordinate members’ research interests within the PPP. The E2BA will focus, strengthen and give coherence to an overall effort in Europe, with the objective of accelerating

innovation in cutting edge European low carbon technologies. The association will also pool its members’ efforts in order to support the mission of the E2B EI. The E2B EI is a Europe wide, industry driven, research and demonstration programme for energy efficient buildings and districts, with the ambitious vision that all the European buildings will be designed, built or renovated to high energy efficiency standards by 2050. Its overall vision is to deliver, implement and optimize building and district concepts that have the technical, economic and societal potential to drastically decrease energy consumption and reduce CO2 emissions in both new and existing buildings across the European Union (EU). The E2B EI aspiration is to manage a € 2bn research and demonstration programme from 2009 until 2019. To date the EC has committed € 500m for the period 2010 to 2013 in the framework of the E2B PPP. The E2B EI will increase the level of research into key technologies and develop a competitive industry in the fields of energy-efficient construction processes, products and services. With the outcome of this research, the European community will be equipped to address climate change and improve its energy independence. The E2B EI will work to achieve the following objectives: • deliver high quality, cost effective research that secures confidence from industry, public and private investors, decision makers and other stakeholders; • leverage further industrial, national and regional RTD investment; • build close cooperation with research being carried out at national and regional levels; • enable the market entry of energy efficiency technologies, allowing commercial market forces to drive the associated public benefits; • place Europe at the forefront of energy efficient buildings and district technologies worldwide; • focus on achieving long-term sustainability and industrial competitive targets for cost, performance and durability aimed to overcome critical technology problem areas; • stimulate innovation and the emergence of new value chains including SMEs; • facilitate the interaction between industry, universities and research centers; • encourage the participation of the new Member States and candidate countries; • perform broadly conceived socio-techno economic research aimed to assess and monitor technological progress;

Newsletter EnginSoft Year 7 n°3 -

• target non-technical barriers to leverage markets and carry out research modes to support the development of new regulations; • review existing standards to eliminate artificial barriers to markets; • provide reliable information to the general public on the benefits of new technologies to the environment, security of supply, energy costs and employment. The impacts that will arise while pursuing the achievement of the aforementioned objectives are important and numerous, the most effective are: • impacts on energy consumption and on renewable energy installations: lowering the total primary energy requested by buildings and generating at least 20% of the total primary energy from renewable resources; • impacts on the environment: the use of the new innovative and cost effective technologies will reduce

the production of CO2 of nearly 65 million tons per year; • impacts on the society: generating 90,000 to 150,000 new jobs and improving buildings indoor comfort (thermal comfort, indoor air quality, acoustics and visual comfort); • impacts on the economy: cost savings in the range of 12% for lighting, 55% for heating and 20% for hot water, that results in an estimate of € 126,000 - € 150,000 per year.

Please get more information on the E2BA from: http://www.e2b-ei.eu For further information please contact: Angelo Messina - EnginSoft info@enginsoft.it

Feat Group: We Forge All You Need La storia di Feat trova il suo principio agli inizi degli anni 70. Si colloca nel cuore della Brianza, nella zona del lago di Como, una terra abitata da genti con solide competenze meccaniche, dedite al lavoro e portate alla creatività. L'uomo che ha guidato lo sviluppo di Feat è il sig. Cogo che, dopo esperienze in Svizzera e con un gruppo americano, ha preso in mano le redini di quella che allora era una piccola stamperia dotata più di entusiasmo che di risorse. Erano anni di grandi opportunità , dove l'Italia era considerata sul piano internazionale un fornitore di qualità e competitivo.

Fin dall'inizio Feat ha delineato tre elementi strategici alla base del proprio sviluppo nel settore dello stampaggio a caldo dell'acciaio: 1) I clienti vanno cercati in tutto il mondo e devono essere ditte di primo livello che esigono fornitori di primo livello. 2) Non bisogna dare solo una prodotto ma anche un servizio tecnico di sviluppo con competenze specifiche nei vari campi di applicazione. 3) È un falso luogo comune che la qualità si dia per scontata ed il prezzo sia l'unica componente distintiva. Per applicazioni di sicurezza o ad alto contenuto di sollecitazioni strutturali, un componente forgiato viene scelto per le sue caratteristiche di affidabilità. Tutti gli elementi che vanno a concorrere nella qualità devono, di conseguenza, essere curati con la massima attenzione. Basti pensare che i clienti, in Feat, hanno a disposizione un ufficio tecnico commerciale con background internazionale con il quale è possibile sviluppare in collaborazione le migliori soluzioni produttivo – funzionali. Inoltre trovano un metallurgista per poter definire correttamente le caratteristiche dei materiali e dei trattamenti termici necessari e possono avvalersi di

- Newsletter EnginSoft Year 7 n°3

un laboratorio interno completamente equipaggiato. Oggi Feat è fornitore strategico per importanti marchi internazionali produttori di primo impianto. Le nostre forniture interessano tutti i paesi europei, gli Stati Uniti ed il Sud Africa. Inoltre consideriamo con vivo interesse i nuovi mercati di sbocco: Brasile, Corea e Medio Oriente. La produzione si è specializzata nei settori del valvolame per l'industria chimica ed energetica, dei ganci per le macchine da sollevamento e dei componenti per le macchine da costruzione e movimento terra. La tendenza seguita è quella di completare lo stampato con lavorazioni meccaniche di precisione fino a consegnare un componente finito , certificato e pronto per l'assemblaggio. Le risorse produttive non hanno mai smesso di evolvere ed i processi sono messi costantemente in discussione. Oggi disponiamo di linee di stampaggio capaci di lavorare componenti fino a 100 kg, trasformiamo tutti i tipi di acciaio con una specializzazione per le super-leghe e l'inox. Gestiamo geometrie complesse e cambiamo produzione con flessibilità. La sfida del prossimo decennio è quella di robotizzare le operazioni di stampaggio riuscendo a preservare la flessibilità e mantenendo ragionevoli i costi delle attrezzature. In questo modo potremo garantire l' omogeneità dei parametri qualitativi e produttivi ed elevare i passati standard artigianali a quelli propri dell'ingegneria industriale. Visitate il sito di FEAT all’indirizzo: www.featgroup.com L’utilizzo di FORGE L'introduzione di Forge in Feat risale al 2003. Avevamo approcciato quest'applicazione consentendo a due studenti dell' università di Padova, di svolgere la loro tesi presso di noi. Alla fine dell' esperienza ci eravamo convinti che questo tipo di software potesse aiutarci ad ottenere risparmi di materia prima e un approccio più metodico alla progettazione degli stampi con conseguente codifica del knowhow aziendale acquisito. A tal fine abbiamo organizzato all'interno del dipartimento stampi, una funzione specifica dedicata all'utilizzo di Forge e dotata degli strumenti hardware più moderni.

L'utilizzo delle simulazioni è diventato un passaggio d'uso comune ogni volta che vogliamo studiare una nuova famiglia di prodotto oppure re-ingegnerizzare uno stampato esistente. È ormai diventata una prassi per tutto lo staff tecnico, incontrarsi in sala corsi per potere analizzare e discutere in team le simulazioni sviluppate. L'utilizzo costante e l'applicazione nei casi concreti sono condizioni fondamentali per sfruttare al massimo le potenzialità dello strumento. Ad esempio, nel caso illustrato in figura vediamo un componente valvola destinato a settori dell'industria farmaceutica, si tratta di applicazioni estremamente delicate, ove ogni imperfezione superficiale potrebbe dare luogo a disastrose non-conformità. In questo caso Forge ci ha permesso di individuare preventivamente un'insidiosa ripiega del materiale e di sviluppare una soluzione tramite la modifica dello stampo.

Perché EnginSoft e FORGE in FEAT Ad oggi Feat ha raggiunto un buon livello di esperienza che consente di fare simulazioni utili, conoscendo la loro affidabilità e limiti.

Dà soddisfazione vedere i responsabili di produzione essere oggi i primi a volere simulare prima di emettere una quotazione o definire un'attrezzatura. Riteniamo che Forge ci abbia aiutato ad ottenere benefici nei seguenti campi: riduzione sfrido, prevenzione problematiche di stampaggio, individuazione difettologie come: mancato riempimento , cricche o ripieghe. Analisi dell'andamento delle fibre, robotizzazione della movimentazione, vita stampi. Il beneficio più importante è stato quello di stimolare il lavoro di gruppo, con un linguaggio scientifico comune per integrare le varie competenze in soluzioni innovative. Il supporto di EnginSoft, attraverso l’assistenza telefonica ed affiancamenti dedicati, è fondamentale per essere aggiornati sui miglioramenti dello strumento e ci consente di applicarlo a problemi nuovi e sempre più complessi.

Newsletter EnginSoft Year 7 n°3 -

Enginsoft ha partecipato allo Users’ Meeting Europeo di FORGE il 7 e 8 giugno 2010 a Sophia Antipolis, Francia Nei giorni 7 e 8 giugno 2010 presso il centro congressi AGORA EINSTEIN a Sophia Antipolis si è tenuto il tradizionale Users’ Meeting Europeo degli utilizzatori di Forge, organizzato da Transvalor, produttore dei due programmi. In questa occasione sono state presentate le ultime novità del software e Transvalor si è confrontata con i propri clienti, raccogliendo suggerimenti per migliorare le funzionalità di Forge e ColdForm. Nutrita è stata la partecipazione italiana (vedi foto dell’“italian team”), con utilizzatori in campi di applicazione anche molto diversi, interessati a confrontarsi con Transvalor e con gli altri utilizzatori presenti per migliorare il modus operandi e la qualità dei risultati. Tra le novità di Forge 2009 sono state evidenziate il nuovo modulo di ottimizzazione automatica, in grado di fornire risultati molto più affidabili della progettazione sperimentale, la significativa riduzione dei tempi di calcolo, ottenuta grazie alla revisione delle funzioni di contatto, le nuove funzioni di tracciatura delle ripieghe (vedi www.enginsoft.it/software/forge/ per ulteriori dettagli) Per quanto riguarda le modifiche a breve termine, in corso di implementazione per la prossima versione, sono stati mostrati nuovi modelli più completi per l’impostazione di processi particolari (laminazione), nuovi modelli di presse meccaniche, miglior controllo di volume, nuovi risultati nei file .vft, ma soprattutto un nuovo strumento per la generazione in automatico dei report di calcolo in formato Word, PowerPoint o pagine HTML. Per quanto riguarda lo sviluppo a medio termine, è in fase iniziale di sviluppo una interfaccia completamente nuova, “Forge workbench”, che integrerà pre-, post-processore e launcher: flessibilità, personalizzazione, usabilità ed ergonomicità sono gli obiettvi di questo sviluppo, che verrà testato prima della release ufficiale con un gruppo ristretto di utenti. Un altro tema di ricerca è l’integrazione in Forge di un modulo per il calcolo del riscaldamento ad induzione, che verrà validato attraverso casi industriali suggeriti dagli utenti e rilasciato ufficialmente nel 2012. Per ultimo, recentemente è stato concluso un progetto di ricerca con il CEMEF basato sul metodo bi-mesh, in grado di ridurre significativamente i tempi di calcolo, soprattutto nella simulazione di processi stazionari e non stazionari, come ad esempio la laminazione circolare.

Per quanto riguarda le presentazioni degli utenti, in tutti i lavori è stato sottolineato che Forge consente di ottenere un notevole ritorno sugli investimenti. Alcuni esempi: Forge è stato utilizzato da un utente tedesco per una simulazione di forgiatura incrementale e da un secondo utente per lo studio dei fattori che possono influenzare il ciclo di vita dello stampo. Un utente spagnolo ha invece impiegato il software per simulare la formazione di un disco con la forgiatura orbitale. In un altro caso, Forge ha calcolato l’evoluzione della dimensione della grana in una lega a base di nickel per migliorarne la qualità. Altri utenti hanno mostrato applicazioni del software per la simulazione del processo di laminazione, la laminazione trasversale planetaria KRM e l’analisi della deflessione della pressa meccanica in un processo di forgiatura multistage. A contorno delle sessioni tecniche, Transvalor ha organizzato una eccellente cena di gala sul lungomare di Cannes, che è stata molto apprezzata da tutti i presenti. Ing. Marcello Gabrielli (info@enginsoft.it) Responsabile in Enginsoft della attività di simulazione con il software Forge Meeting Italiano Forge Montichiari, 22 Ottobre 2010 Come ogni anno, Enginsoft propone, a tutti gli utilizzatori che non hanno potuto essere presenti al Meeting Europeo di Transvalor, un Meeting Italiano degli utilizzatori di Forge, che avrà luogo in occasione della 2010 Enginsoft International Conference a Montichiari (BS), i giorni 21 e 22 ottobre. La seconda giornata vedrà presente Transvalor per un riassunto di quanto mostrato in Francia, e alcuni clienti italiani e stranieri, che mostreranno come viene utilizzato il software nella propria azienda. Per maggiori informazioni, consultate il programma dell’evento all’indirizzo http://www.caeconference.com/

- Newsletter EnginSoft Year 7 n°3

EnginSoft Event Calendar ITALY September 2010- Automotive WEBINARS on Modelling Metal Cutting and Machining Simulation with AdvantEdge™ EnginSoft is pleased to announce the next Webinars on Modelling Metal Cutting and Machining Simulation with AdvantEdge. To register, please visit: www.enginsoft.com 21-22 October 2010 – EnginSoft International Conference 2010. CAE Technologies for Industry. Fiera Montichiari, Brescia. Be part of Europe’s major CAE event where today's limitless applications of Simulation based Engineering and Sciences will be discussed! “Believe in innovation: simulate the world” www.caeconference.com FRANCE 7 October- Journée Simulation Numérique «Organisation et rentabilité de la fonction calcul». Paris http://www.af-micado.com/ 12-13 October – Congrès Nafems. «Simulation numérique: moteur de performance». Paris http://www.nafems.org/events/nafems/2010/francecongres/ 18 November – French Flowmaster and modeFRONTIER® Users Group Meeting. Forum Utilisateurs. Enginsoft France vous invite à participer à l'édition 2010 de son Forum Utilisateurs qui se déroulera le 18 novembre à Paris. Cette journée sera résolument orientée témoignages clients sur les solutions modeFRONTIER® et Flowmaster! Hotel Saint James et Albany, Paris, Lieu : Saint James et Albany - 202, rue de Rivoli 75001. www.enginsoft-fr.com EnginSoft France 2010 Journées porte ouverte dans nos locaux à Paris et dans d’autres villes de France, en collaboration avec nos partenaires. Prochaine événement: Journées de présentation modeFRONTIER® Pour plus d'information visitez : www.enginsoft-fr.com, contactez: info.fr@enginsoft.com GERMANY Please stay tuned to www.enginsoft-de.com, contact info.de@enginsoft.com for more information. 2-4 November - AIRTEC 2010, Messegelände Frankfurt Visit EnginSoft GmbH in the exhibition of AIRTEC 2010, booth E 20 in „Design & Engineering“. We are delighted to present modeFRONTIER® and „Design and Numerical Optimization of Winglets of a Piaggio Aircraft” by Ubaldo Cella, Piaggio Aero Industries; Francesco Franchini,

EnginSoft SpA, in the parallel Conference „Supply on the wings“. Please note the presentation in your diary: 3rd November 2010, 15:00hrs, Session „Improved Simulations/Experiments“ http://www.airtec.aero/ modeFRONTIER® Seminars 2010 EnginSoft GmbH, Frankfurt am Main • 26 October • 30 November 24 – 25 November 2010 - NAFEMS European Conference: Simulation Process and Data Management Holiday Inn Frankfurt Airport-North http://www.nafems.org/events/nafems/2010/ EuropeSDPM2010/ Seminars Process Product Integration EnginSoft GmbH, Frankfurt Office How to innovate and improve your production processes ! Seminars hosted by EnginSoft Germany and EnginSoft Italy Please stay tuned to: www.enginsoft-de.com UK Please stay tuned to www.enginsoft-uk.com, contact Bipin Pastel at: b.patel@enginsoft.com for more information. modeFRONTIER® Workshops at Warwick Digital Lab • 18 October • 10 November • 7 December Please register for free on www.enginsoft-uk.com 28-29 September -The InfoWorks user meeting. Crowne Plaza Hotel, Reading, Berkshire. EnginSoft UK are official sponsors as well as co-presenting with Wessex Water 10-11 November 2010 - WaPuG. Hilton Hotel, Blackpool EnginSoft UK will be attending. www.ciwem.org/groups/wapug 25 November 2010 - modeFRONTIER® Workshops with InfoWorks CS at Warwick Digital Lab. SWEDEN Training Courses: November 3-4 - Introduction to modeFRONTIER®. For more information, please contact Adam Thorp, a.thorp@enginsoft.se. For training registration, please visit http://nordic.enginsoft.com/training/index.html October 26-27 - NAFEMS Nordic Regional Conference 2010,

Newsletter EnginSoft Year 7 n°3 -

Göteborg. EnginSoft Nordic will be giving the presentation "Multi-objective. Optimization of Dual-Antenna Handhelds for MIMO Communications" together with Ericsson and Efield. Please visit http://www.nafems.org/events/nafems/ 2010/NORDIC2010/ for more information, abstract submission and registration. SPAIN 28 - 29 September - Introductory Course on the use of modeFRONTIER®. The 2-day course provides a practical introduction to design optimization using modeFRONTIER®. The course combines lectures but most of the time is dedicated to hands-on sessions so that the attendees complete the course with the basic skills in using many of the modeFRONTIER® functions. More information can be found on http://www.aperiotec.es/agenda.php Programa de cursos de modeFRONTIER® and other local events. Please contact our partner, APERIO Tecnología: g.duffett@aperiotec.es and stay tuned to: www.aperiotec.es 4 November - NAFEMS-Iberia Awareness Seminar on Organised by: NAFEMS-Iberia. Department. of Aeronautics, Polytechnic University of Madrid (UPM), Madrid. Gino Duffett to represent Aperio Tecnología and present: Experimental and Simulation Evaluation of Material Properties Related to Mechanical Components PORTUGAL 24 November 2010 - NAFEMS Awareness Seminar on Finite Elements and Numerical Optimization in Engineering. Department of Mechanical Engineering, University of Aveiro, Aveiro. Organised by: Research Group GRIDS (Department of Mechanical Engineering, University of Aveiro) and NAFEMSGino Duffett to represent Aperio Tecnología / ESTECO and present: Multi-Disciplinary Optimization and Automatic Design Process using CAE software and modeFRONTIER®. USA 15-16 November - NAFEMS Virtual Conference: 2020 Vision of Engineering Analysis and Simulation Hosted online by NAFEMS North America http://www.nafems.org/events/nafems/2010/NA2010/ 8 December – Workshop on Optimization hosted by Stanford University, Cascade Technologies and EnginSoft USA. A unique program on the benefits and use of optimization in today’s product design and development, conducted by Gianluca Iaccarino, Assistant Professor, Stanford School of Engineering. Courses on: Design Optimization with modeFRONTIER®. Ozen Engineering, Sunnyvale – Silicon Valley, CA. Learn about Optimization coupled with ANSYS. OZEN can easily help you out automating the search for the optimal design. The primary audience for this course includes ANSYS Classic and Workbench users as well as new modeFRONTIER® users who want to have a complete overview to all software capabilities. www.ozeninc.com

EUROPE, VARIOUS LOCATIONS modeFRONTIER® Academic Training. Please note: These Courses are for Academic users only. The Courses provide Academic Specialists with the fastest route to being fully proficient and productive in the use of modeFRONTIER® for their research activities. The courses combine modeFRONTIER® Fundamentals and Advanced Optimization Techniques. For more information, please contact Rita Podzuna, info@enginsoft.it To meet with EnginSoft at any of the above events, please contact us at: info@enginsoft.com

EnginSoft Contributes to the LION5 Conference This meeting, which continues the successful series of LION events, is aimed at exploring the intersections and uncharted territories between machine learning, artificial intelligence, mathematical programming and algorithms for hard optimization problems. The main purpose of the event is to bring together experts from these areas to discuss new ideas and methods, challenges and opportunities in various application areas, general trends and specific developments. EnginSoft will contribute to the conference because of its interest in design optimization and machine learning. Silvia Poles, Optimization Consultant at EnginSoft, will give a tutorial on “Multiobjective Optimization for Innovation in Engineering Design” as a survey on methodologies to approach the design optimization process, a set of best practices intended for rapid delivery of high-quality products, with a specific focus on the numerical algorithms and post-processing used for selecting optimal design configurations.

Moreover, EnginSoft, together with the University of Trento, the University of Udine and Microsoft Research, is organizing a special session on “Software and Applications” as part of LION5. All customers’ contributions on solving design optimization problems using dedicated software (e.g. modeFRONTIER®, ANSYS, …) are welcome. The conference takes place at “Sapienza Università di Roma, Dipartimento di Informatica e Sistemistica Antonio Ruberti”, on January 17th-21st, 2011. More information is available at: http://www.intelligent-optimization.org/LION5/