Newsletter EnginSoft - 11-1

Year 8 n째1 Spring 2011

ICEPAK 13.0: buone notizie per i progettisti elettronici APERS CALL FOR P EN IS NOW OP

CAE-based tablet design

Combined 1D & 3D CFD approach for GT Ventilation System analysis

Analisi di un meccanismo link-drive per presse con tecnologia multibody in ANSYS

Multi Variate Analysis in Systematic Impeller Design Applying modeFRONTIER at Sulzer Pumps

Interview with Paolo Nesti, Piaggio Group

Ottimizzazione Termofluidodinamica di un forno da cucina Indesit

Newsletter EnginSoft Year 8 n°1 -

3

EnginSoft Flash Researchers of Trinity College Dublin have With this 1st edition of the Newsletter in significantly improved the fatigue resistance 2011, we extend a Special Invitation to of components using modeFRONTIER. GE Oil our readers, to meet us at the EnginSoft & Gas Italy adopted a combined 1D and 3D International Conference 2011 from 20th numerical approach with Flowmaster and - 21st October in Verona. There could ANSYS Fluent to study ventilation systems. hardly be a better venue for the community We interviewed Mr Paolo Nesti, engineer at of simulation and VP (Virtual Prototyping) Piaggio Group, one of the major players users than Verona. A UNESCO world worldwide in the two-wheeler vehicles heritage site, famous for its operas, the sector, and feature a case study on the use of ancient amphitheatre built by the Romans, ANSYS Workbench at Piaggio. Landi Renzo Romeo and Juliet, and a diverse cultural S.p.A., a global leader in components, LPG wealth. Today, Verona is a vibrant city and CNG fuel systems for motor vehicles, dedicated to innovation. Verona’s airport spoke to us about the use of modeFRONTIER offers daily nonstop flights to Europe’s Ing. Stefano Odorizzi in their product development. Componeering hubs which will facilitate travel for our EnginSoft CEO and President Inc. Finland presents the Opencell Delta™ guests from around the world! concept which provides a brand new way to The Conference, which is one of the major construct metal sandwich panels.

Get-togethers for CAE and VP users worldwide, will again The Event Calendar features conferences, fairs and courses present a parallel event: the ANSYS Italian Users’ Meeting. across Europe, in the USA and Japan...

EnginSoft is delighted to collaborate with ANSYS, Inc. and When we hear about Japan in these days, above all our ANSYS Italia, our key partners, to offer an interactive heart and best wishes go out to the Japanese people who platform to the ANSYS developers and users to share battle and will overcome the consequences of a terrible knowledge, experiences and to enhance the use of ANSYS natural disaster that hit their country. Our Japan Column in the various industrial fields. brings to our readers an article on the novel approach of In this edition, we report on the progress of the EnginSoft CAE-based tablet design of Mr. Hideaki Sato of ASAHI Americas Project. EnginSoft has recently strengthened its BREWERIES, LTD. Elysium presents news on the use of North American operations by expanding its base in Palo ASFALIS at Nissan. We close the Column with an article on Alto, Silicon Valley. Moreover, EnginSoft has joined the Tokyo, a unique metropolis…and some ideas on how each TFSA (Thermal and Fluid Sciences Affiliate) Program of one of us can help.

Stanford University.

Finally, the Editorial Team would like to recommend the Cascade Technologies, Inc, EnginSoft’s partner, is a spinnew book “Reactive Business Intelligence. From Data to off of the Center for Turbulence Research at Stanford Models to Insight” by Prof. Roberto Battiti and Prof.

University. Cascade develops and supports state of the art Mauro Brunato to our readers. While the book is easy-toCFD analysis tools for various engineering applications. To read, it guides us from the very basics to such advanced stimulate the discussion on optimization, EnginSoft and topics as supervised learning, data-mining, optimization, Cascade have sponsored a One-Day Seminar on statistics and interactive visualizations.

Optimization, which was held on 1st February at Stanford To continue our discovery of the immense opportunities of Campus. The driving force behind Cascade Technologies is CAE and VP in our today’s world, EnginSoft and ANSYS Prof. Gianluca Iaccarino, who was recently awarded the invite our readers to the International Conference 2011.

Presidential Early Career Award for Scientists and Please follow the Announcements, Call for Papers and Engineers (PECASE) by President Barack Obama. Our Program on www.caeconference.com readers will find more information on Prof. Iaccarino’s work and the prestigious award in this issue.

We look forward to welcoming you to Verona this October!

EnginSoft’s growing international business is also reflected in the highlights of this issue: Sulzer Pumps, one of the world's leading centrifugal pump manufacturers Stefano Odorizzi based in Winterthur, Switzerland, speaks about Multi Editor in chief Variate Analysis in Systematic Impeller Design.

4 - Newsletter EnginSoft Year 8 n°1

Sommario - Contents CASE STUDIES

6 10 13 15 19

Multi Variate Analysis in Systematic Impeller Design Applying modeFRONTIER at Sulzer Pumps Analisi di un meccanismo link-drive per presse con tecnologia multibody in ANSYS Ottimizzazione termofluidodinamica di un forno da cucina Indesit Combined 1D & 3D CFD Approach for GT Ventilation System Analysis ANSYS WB and a Review of the Design Metrics in Piaggio: the Case of the Motor Shaft

INTERVIEWS

21

EnginSoft Interviews Ing. Paolo Nesti, Piaggio Group

CASE STUDIES

26 27

modeFRONTIER Used in the Design of Fatigue-Resistant Notches A Multi-Objective Optimization with Open Source Software

SOFTWARE NEWS

32 33 36 38

ANSYS 13: Il punto sui solutori per modelli di grandi dimensioni nelle simulazioni meccaniche La simulazione di sistema in ANSYS: Simplorer ICEPAK 13.0: buone notizie per i progettisti elettronici Development of the Novel Opencell™

IN DEPTH STUDIES

41

Componenti forgiati di qualità necessitano di un approccio CAE integrato – esperienze di simulazione di processo nel campo Energia e Nucleare

TESTIMONIAL

46

Landi Renzo: the global leader in the sector of components and LPG and CNG fuel systems

JAPAN CAE COLUMN

47 48 52

The CAD-CAM Cooperation in Nissan Achieved by ASFALIS CAE-based tablet design Tokyo a Metropolis

The EnginSoft Newsletter editions contain references to the following products which are trademarks or registered trademarks of their respective owners: 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) modeFRONTIER is a trademark of ESTECO srl (www.esteco.com) Flowmaster is a registered trademark of The Flowmaster Group BV in the USA and Korea. (www.flowmaster.com) MAGMASOFT is a trademark of MAGMA GmbH. (www.magmasoft.com)

ESAComp is a trademark of Componeering Inc. (www.componeering.com) Forge and Coldform are trademarks of Transvalor S.A. (www.transvalor.com) AdvantEdge is a trademark of Third Wave Systems (www.thirdwavesys.com)

.

LS-DYNA is a trademark of Livermore Software Technology Corporation. (www.lstc.com) SCULPTOR is a trademark of Optimal Solutions Software, LLC (www.optimalsolutions.us) Grapheur is a product of Reactive Search SrL, a partner of EnginSoft For more information, please contact the Editorial Team

Newsletter EnginSoft Year 8 n°1 -

PRESS RELEASE

54 56

President Obama Honors EnginSoft’s Partner with the Presidential Early Career Award for Scientists and Engineers Formazione a distanza sugli elementi finiti

BOOK REVIEWS

57

REACTIVE BUSINESS INTELLIGENCE: From Data to Models to Insight

60 61 62

Newsletter EnginSoft Year 8 n°1 - Spring 2011 To receive a free copy of the next EnginSoft Newsletters, please contact our Marketing office at: newsletter@enginsoft.it 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

EVENTS

58

5

EnginSoft at the Optimization Day: Research and Applications NAFEMS World Congress 2011 - Preliminary Agenda Announced EnginSoft alla Fiera Made in Steel di Brescia EnginSoft Event Calendar

PAGE 6 MULTI VARIATE ANALYSIS IN SYSTEMATIC IMPELLER DESIGN APPLYING MODEFRONTIER AT SULZER PUMPS

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

PAGE 15 COMBINED 1D & 3D CFD APPROACH FOR GT VENTILATION SYSTEM ANALYSIS

EnginSoft GmbH - Germany EnginSoft UK - United Kingdom EnginSoft France - France EnginSoft Nordic - Sweden Aperio Tecnologia en Ingenieria - Spain www.enginsoft.com

ASSOCIATION INTERESTS NAFEMS International www.nafems.it www.nafems.org TechNet Alliance www.technet-alliance.com RESPONSIBLE DIRECTOR Stefano Odorizzi - newsletter@enginsoft.it PRINTING Grafiche Dal Piaz - Trento

PAGE 36 ICEPAK 13.0: BUONE NOTIZIE PER I PROGETTISTI ELETTRONICI

The EnginSoft NEWSLETTER is a quarterly magazine published by EnginSoft SpA

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

ESTECO srl 34016 TRIESTE Area Science Park • Padriciano 99 Tel. +39 040 3755548 • Fax +39 040 3755549 www.esteco.com

6 - Newsletter EnginSoft Year 8 n°1

Multi Variate Analysis in Systematic Impeller Design Applying modeFRONTIER at Sulzer Pumps The most important pump component - its heart - is the impeller which transforms kinetic energy into pressure and therefore generates the required head. The impeller geometry is defined by more than 50 parameters requiring experienced hydraulic design engineers. Even if only 20 of these parameters have a major influence, it is obvious that a severe variation yields an excessive database which should be made use of.

Fig. 1 - Two stage pump with detailed view of the first stage impeller.

estimate for the new impeller and properly conditions the variable ranges for an optimization which is likely to follow. This article presents an approach based on classification of existing impeller designs with Multi Variate Analysis through Self Organizing Maps (SOM) by use of modeFRONTIER. Systematic impeller design The parameters defining an impeller include the main dimensions like outer diameter D2 and shaft diameter D0 as also the meridional contour and blade shape (Figure 2).

The impeller design is done for a specified operating point with given flow rate Q and head H for a certain rotational speed n. Efficiency Ρ is one criterion for an optimal impeller design not only at best efficiency point bep but also over a certain operating range (Figure 3, left).

Suction capability, which is the pressure available at pump inlet NPSH, is another criterion (Figure 3, centre).

Decreasing NPSH affects the pump head which needs to be considered.

Good suction capability and high efficiency both over a broad operating range are conflictive design goals.

Increasing suction capability at maximum operating point reduces efficiency at minimum operating point. This is an important fact when using optimization techniques in impeller design.

Fig. 2 - Dimensions of the impeller.

A proper classification of the available designs in the database gives the developer a better understanding of the complex parameter correlation and enables the prediction of not yet available impellers by interpolating among the existing designs. This gives a first parameter

Characteristic numbers Impellers can be classified by characteristic numbers enabling a comparison among the designs. The specific speed nq defines the form of the impeller (Figure 4) and is calculated from flow rate Q, head H and the rotational speed n. Figure 5 lists the main design parameters and shows their conversion into characteristic numbers. The outer diameter D2 of the impeller is selected according an

Fig. 3 - left: Efficiency Ρ; centre: suction capability NPSH within the operating range of the impeller dependent on the flow rate Q; right: NPSH3% criterion

Newsletter EnginSoft Year 8 n°1 -

optimal head coefficient Ψ for the specific speed nq. The inlet diameter D1 influences the suction capability and depends on the flow coefficient at inlet ϕ1. The suction capability can either be expressed by a characteristic number σ or the suction specific speed nss which both depend on the suction head at pump inlet NPSH. Similar relations exist for other dimensions.

Using these characteristic numbers and dimensionless values, impellers with different outer diameters D2 can be

Fig. 4 - Impeller form in meridional view dependent on specific speed nq

compared and new designs can be calculated based on these values. This facilitates a classification of the impellers and the use of the SOM technique.

7

“pixel” being colored to reflect different properties of the input data, e.g. specific speed nq, impeller width B2 or efficiency η. This way SOM overcomes the problem of visualizing multivariate data: input data are projected onto a 2D grid (Figure 6).

Another advantage of the SOM is related to its intrinsic interpolation capability. There might exist regions of unexplored zones, the performance of the impeller is not yet predicted by CFD. This is reflected in SOMs by empty hexagons separating others that are filled up by different families of designs. When this happens, it is reasonable to look at the prototype vector of the empty unit as kind of forecast of a design family that have still to be realized, and that represent a reasonable interpolation of the ones that are available. The benefit of the SOM is that, apart from the input variables characterizing such design families, also a complete forecast of the performances is immediately available. This predictive use of the SOM is really powerful when handling designs that are described by a high number of parameters (inputs and outputs) so that any other interpolations approach, like Response Surface Modeling, becomes heavy to implement.

For the pump impellers described here, a unique SOM is trained on the existing database and used to forecast new design families able to provide certain performances.

Self-Organizing Map Any existing impeller design is described through a multidimensional vector, where each component represents a Impeller design and multi variate analysis defining parameter (input) or a performance index The dimensionless values for the main parameters and the (output). design objectives efficiency and suction capability (η, The Self-Organizing Map (SOM) is an unsupervised Neural nSS,bep, nSS,max) are selected as input for the training Network algorithm capable to group and classify such already available impeller designs in a two-dimensional (classification) of the SOM. Within this test case, the grid space. Each node of this grid is called “Unit” and it results of six different impeller optimizations with three groups (includes) vectors (impeller designs) that are similar with respect to all their parameters (inputs and outputs) simultaneously.

SOM preserves the topology of the data, so that similar data items will be mapped to nearby units on the map. To do so, units are hexagonal-shaped, and hence each unit has 6 neighbors and is labeled by a “prototype vector” that in fact represents all the vectors included in the unit itself, as a kind of average.

Fig. 6 - Left: SOM is the blue network that adapts to real data (red points) in a X-Y-Z space, SOM lives in the multi-dimensional data note that some nodes are far from any real point, hence the related unit will be empty. space, but its visualization capabilities are Centre: the same SOM in its 2D conventional representation: each of the nodes is a hexagonal unit. Right: each unit is colored with respect to the X-value of its prototype vector, and the built on the top of its representation in the square on its center represents the number of real design enclosed in the unit (see the empty grid space. Each of the hexagons becomes a units).

Fig. 5 - Correlation between dimensional and non dimensional values

8 - Newsletter EnginSoft Year 8 n째1 back from dimensionless to dimensional parameters.

A new impeller is then generated with the conventional design tools and its performance is checked by CFD.

Fig. 7 - Specific speed (nq) in the SOM with distribution of existing designs

Fig. 8 - Selected designs in the SOM (Color: specific speed)

different specific speeds between nq13 and nq60 are used as data basis. Goal of this study is to develop new impellers in this range with high suction capability and high efficiency for four different specific speeds (nq16, nq24, nq30, nq47) under the assumption that for each nq both operating point and impeller diameter D2 are given and the shaft diameter is pre-defined from mechanical calculations.

Figure 7 illustrates the trained SOM of the specific speed ranging from nq13 (blue) to nq60 (red). The squares describe the density of input parameters available. The larger the square the more data exists, no square signifies that parameter values are based on pure interpolation.

The selection of the new impellers is undertaken in regions with purely interpolated data (Figure 8). For each new impeller, existing designs with a similar nq in the SOM table are compared. This is necessary as three objectives need to be fulfilled, and the SOM designs might only achieve one.

The advantage of this technique is the access to every single parameter defining the impeller geometry. With an amount of over 50 parameters, the entire meridional impeller contour and the blade shape are approximated by the SOM. This method allows a complete impeller design within a few minutes just by giving the operating point and selecting an appropriate outer diameter D2. The impeller parameters are taken from the SOM and converted

Table 1 - Comparison of coefficients of obtained and predicted objectives for the selected designs

Table 1 shows a comparison of the performances obtained by CFD and predicted by SOM for the selected nq. A coefficient is defined with: objectiveCFD / objectiveSOM ,

describing the ratio of the CFD result to the prediction of the objective by the SOM. A value equal to one signifies an error of zero; the CFD performance matches the predicted one. For a coefficient smaller than one, the performance is over predicted, if it is larger than one, the design is under predicted by the SOM. nq24 The first impeller modeled with the SOM is nq24. Therefore the best possible solution in compliance with the specific speed and the design goals is selected. CFD calculations are performed according the CFD in the optimization process. The results are excellent, both efficiency and suction performance are better than predicted by the SOM. nq16 For the second impeller (nq16) two different designs are selected from the SOM table, one with high efficiency predicted and a second with a lower efficiency but a better expected suction capability. The impeller with high efficiency reaches almost the suction capability at bep while the impeller with the lower efficiency exceeds the suction capability. Both impellers outperform the expected suction capability at the maximum operating point efficiency. nq30 For the impeller nq30 two different interpolated designs are selected from the SOM table, differing in suction capability at maximum operating point. After calculating performances of the interpolated designs, the aspired suction capability at overload is not achieved, the other targets are outperformed. For this reason two more designs from the SOM table are selected, now with lower efficiency than the previous designs. With the lower efficiency target, the suction capability is reached for both operating points. This proves the conflicting design targets efficiency and suction capability.

Newsletter EnginSoft Year 8 n°1 -

Fig. 9 - Efficiency (red = high, blue = low)

Fig. 10 - suction capability at bep (based on σ, blue = good, red=bad)

nq47 Two designs from the SOM table are selected. The second design with the higher suction capability target at bep misses the required suction capability at maximum operating point. Even if the first design does not fulfill the requirements at bep, the deviation from the target value is only small.

Summary for all designs For all designs, the calculated efficiency is higher than SOM predicted. The suction capability misses the requirements for some designs because of contradicting objectives. In these cases it is possible to select new designs from the SOM with compromises in efficiency but achieving the required suction capability.

Figures 9-11 present the objectives efficiency, suction capability at bep and max OP. It can be clearly seen that efficiency and suction performance pattern are completely different. This discrepancy makes it difficult to fulfill all three objectives and either a compromise is required or one objective has to be prioritized.

Conclusions The article describes a novel methodology to design impellers starting from the well assessed knowledge at

Sulzer Pumps Sulzer Pumps is one of the world's leading centrifugal pump manufacturers. Intensive research and development in fluid dynamics, process-oriented products and special materials as well as reliable service helps Sulzer Pumps maintain its leading positions in its key markets. Its customers come from the oil and gas, hydrocarbon processing, power generation and pulp and paper sectors as well as from water distribution and treatment and other general industries. The products are internationally reputed for their technical excellence. www.sulzerpumps.com

9

Fig. 11 - suction capability at max OP (based on σ, blue = good, red=bad)

Sulzer Pumps. The concept has proven to provide the designer a new and effective tool to speed up the design process of the pumps core part - the impeller.

Self-Organizing Maps (SOM) have been trained on the already available impeller designs and corresponding performance indicates: such a SOM embeds the so far available pump designer knowledge, and provides a complete interpolation in regions in which designs are still missing. Each input or output variable of the impeller design can be represented through a conventional twodimensional SOM map. This allows the use of SOM as an extremely powerful tool to suggest new designs in regions in which the design space has not yet been explored to forecast their performances. In fact, the methodology allows a new and complete impeller design in some minutes, just assigning a few parameters, as the operating point and the outer diameter. Any new design proposed by the SOM can be validated through high-fidelity fluid dynamic simulations (CFD) and then be used as starting point for further refinements by directly linking the CFD model to a modeFRONTIER optimizer, [2].

References [1] Gülich J.: "Centrifugal Pumps", Springer, 2010 [2] Krüger S., Maurer W.: "How to use modeFRONTIER within the daily hydraulic design process: Sulzer Pumps’ experiences with automated impeller design", modeFRONTIER 2008 Users’Meeting, Trieste, 14th-15th October 2008

Wolfgang Maurer, Susanne Krueger Sulzer Pumps, Winterthur, Switzerland Luca Fuligno EnginSoft SpA, Trento, Italy Francesco Linares EnginSoft GmbH, Frankfurt am Main, Germany

10 - Newsletter EnginSoft Year 8 n°1

Analisi di un meccanismo link-drive per presse con tecnologia multibody in ANSYS Analysis of a link drive mechanism for presses using ANSYS MBD multibody technology This paper presents a test case in which the tool “Rigid Dynamics” of ANSYS Workbench 13 is used to investigate the kinematics and the dynamics of a link drive mechanism equipping a deep drawing press.

The device is first analyzed by developing the kinematic motion equations. This step highlights the difficulty which arises when we have to manually manipulate the equations of a complex multibody system.

Then, a parameterized multibody model of the link drive is built in ANSYS. This approach is much more straightforward and allows the user to understand the mechanism’s behavior in a shorter time. In addition, the software makes it possible to watch the working mechanism animation at the end of the solution.

The multibody model returns information about the dynamics, which is useful for structural design purposes. Moreover, thanks to the easy parameter management in ANSYS, we can automatically investigate and compare multiple alternatives of the same mechanism.

Questo articolo presenta un test case significativo nel quale attraverso una “Rigid Dynamics” di ANSYS Workbench 13 vengono efficacemente analizzate la cinematica e la dinamica di una pressa meccanica per imbutitura profonda.

L’imbutitura è un processo di formatura a freddo attraverso il quale una lamiera metallica viene trasformata in un oggetto cavo, con buone caratteristiche dimensionali e di finitura. Nello schema tradizionale, l’imbutitura si realizza attraverso un pun-

zone che spinge la lamiera, eventualmente fissata con un premilamiera, all'interno di una matrice. Il processo è intrinsecamente delicato perché deve indurre nel materiale elevate deformazioni plastiche, senza raggiungere la condizione di rottura.

La qualità del prodotto finale è fortemente influenzata dai parametri di processo, tra i quali spicca per importanza la velocità di discesa del punzone nel tratto di corsa in cui lavora la lamiera.

Idealmente, la velocità del punzone dovrebbe essere bassa, per realizzare una deformazione graduale del materiale, e costante, per evitare la formazione di pieghe e striature superficiali.

In una pressa meccanica con tradizionale schema slider crank, la riduzione della velocità del punzone è ottenibile solo aumentando il tempo ciclo, con ovvie conseguenze negative sulla produttività dell’impianto. Pertanto, se si vogliono ottenere buone performance di processo senza penalizzare la produzione, è opportuno predisporre un meccanismo più raffinato, che consenta maggiori libertà nella gestione della velocità del punzone. Una soluzione è il meccanismo link drive illustrato, già utilizzato da alcuni produttori di presse. La Figura 1 confronta le curve di velocità del punzone per una pressa tradizionale e una pressa link Drive di pari dimensioni (con la stessa corsa massima e lo stesso tempo ciclo).

La velocità di discesa del punzone, per la pressa link drive, presenta un tratto con andamento regolare a velocità quasi costante. Inoltre, grazie al maggior numero di membri, questo meccanismo è molto versatile: variando le dimensioni del meccanismo si possono ottenere diverse curve di velocità. Risulta, pertanto, evidente che il link drive presenta caratteristiche e prestazioni molto più vantaggiose dello schema tradizionale.

Studio analitico del meccanismo link drive Il paragrafo precedente ha messo in luce l’importanza della velocità di discesa del punzone nella messa a punto del processo di imbutitura. Pertanto, è opportuno focalizzare l’attenzione sull’analisi cinematica del meccanismo link drive, che permette di comprendere come il meccanismo trasformi la rotazione del motore elettrico nella traslazione a velocità variabile del punzone.

Fig. 1 – confronto tra pressa “Slider–Crank”e “link drive”: velocità del punzone e zona di lavoro

Gli approcci per condurre uno studio cinematico sono diversi.

Per un meccanismo relativamente semplice come il link drive, è possibile, seppure con qualche difficoltà, derivare analiticamente le equazioni del moto. Lo schema cinematico cui si farà riferimento è riportato in Figura 2.

Newsletter EnginSoft Year 8 n°1 -

11

Simulazione multibody del meccanismo link drive Una valida alternativa per studiare meccanismi in modo più veloce e con minor rischio di errore è la simulazione tramite codice multibody. ANSYS Workbench 13, attraverso il modulo “Rigid Dynamics”, consente di creare e gestire modelli multibody a corpi rigidi. L’utilizzo di questo strumento, permette, inoltre, di integrare i risultati dell’analisi cinematica con tutte le grandezze dinamiche fondamentali per la progettazione strutturale del dispositivo.

ANSYS Workbench 13 consente di gestire in modo parametrico qualsiasi modello di calcolo. Con riferimento all’analisi multibody del link drive, la parametrizzazione permette di simulare varie alternative del meccanismo, consentendo la scelta di quella più adatta alle esigenze.

Fig. 2 – schema cinematico della pressa link drive

I vettori L1, L3, L4, L5 e L6 rappresentano i membri mobili del meccanismo, il vettore L2 rappresenta il telaio e il vettore s rappresenta la posizione del punzone (misurata da un riferimento arbitrario). Lo studio analitico della cinematica inizia con la scrittura delle equazioni di chiusura in forma vettoriale:

Ciascun vettore può essere formalmente descritto nella forma L= L cos (φ), dove L è la lunghezza e φ è l’angolo di inclinazione. Sostituendo nelle precedenti e manipolando opportunamente è possibile esprimere la posizione s del punzone in funzione dell’angolo φ4 di rotazione dell’eccentrico (variabile indipendente del problema). Successivamente, si ricavano per derivazione rispetto al tempo la velocità v e l’accelerazione a. In sintesi, l’analisi cinematica mediante approccio analitico restituisce tre equazioni nella forma:

Nella fase di pre-processing avviene l’assemblaggio del modello multibody. Le geometrie dei corpi possono provenire da CAD esterni oppure possono essere create direttamente all’interno di Design Modeler. Per questa applicazione abbiamo provveduto a creare integralmente la pressa ed il meccanismo, parametrizzando le grandezze di cui andremo ad analizzare gli effetti.

Sono quindi state scelte e create le connessioni tra i corpi (Figura 3).

I revolute joint consentono la rotazione relativa tra i membri connessi, mentre i general joint lasciano liberi i gradi di libertà scelti esplicitamente dall’utente.

Per consentire la soluzione di un modello multibody a corpi rigidi, i vincoli, inseriti sottoforma di connessioni, non devono essere ridondanti. ANSYS mette a disposizione lo strumento “Redundancy Analysis” che permette di individuare automaticamente la presenza di vincoli in eccesso e che fornisce indi-

Queste espressioni hanno una struttura molto articolata e pertanto ne omettiamo la scrittura estesa. Si noti che i risultati dipendono dalla legge di moto assegnata al movente e dai parametri dimensionali del meccanismo.

Benché l’approccio analitico consenta di pervenire ai risultati cinematici in tempi accettabili, va precisato che la manipolazione di equazioni con questo livello di complessità è una operazione alquanto delicata: se non si dispone di strumenti per la manipolazione simbolica, il rischio di errore è decisamente elevato.

Fig. 2 – vista 3D dell’assieme e schema delle connessioni

12 - Newsletter EnginSoft Year 8 n°1 Confronto di 3 configurazioni del meccanismo link drive A titolo di esempio, proponiamo lo studio comparativo della risposta cinematica restituita dal meccanismo link drive modificando la coordinata orizzontale della cerniera di collegamento della biella L1 al telaio (Figura 2 e Figura 4).

Nello specifico, abbiamo ipotizzato di passare da un valore di 1400 mm a 1100 mm, con un valore intermedio di 1250 mm.

Le tre analisi sono condotte semplicemente modificando il parametro corrispondente nell’interfaccia utente. ANSYS MBD aggiorna automaticamente le geometrie e ripete la simulazione multibody.

Fig. 4 – modelli di pressa ottenuti tramite parametrizzazione

cazione di quali modifiche si debbono apportare per rendere il modello consistente.

Il meccanismo virtuale viene azionato applicando una legge di moto all’albero dell’eccentrico. ANSYS consente di assegnare a qualsiasi connessione sia leggi di moto, sia azioni dinamiche.

In entrambi i casi, le funzioni sono gestibili in forma tabulata o tramite espressioni analitiche. Nel caso del meccanismo link drive, abbiamo imposto al movente una velocità di rotazione costante.

L’analisi multibody comporta l’integrazione delle equazioni del moto che il software ha sviluppato automaticamente durante l’assemblaggio del modello. ANSYS dispone di due integratori, con diverse opzioni per la gestione del passo, della convergenza e della qualità della soluzione.

Nella fase di post-processing ANSYS Workbench permette di estrarre numerosi risultati dai corpi e dalle connessioni. Le grandezze disponibili per i corpi sono di natura cinematica (posizione, velocità, accelerazione), mentre per le connessioni possiamo diagrammare le grandezze cinematiche dei gradi di libertà consentiti e le reazioni vincolari dei gradi di libertà annullati dal joint.

Ad esempio, su un “revolute joint” possiamo leggere angolo, velocità ed accelerazione angolare lungo l’asse di rivoluzione, e, in aggiunta, possiamo estrarre forze radiali, forze assiali e momenti trasversali che i corpi si scambiano mutuamente.

Fig. 5 – confronto della velocità del punzone per i tre modelli di pressa

La Figura 5 illustra gli effetti del parametro scelto sulla velocità del punzone. Lo zoom mette in particolare evidenza la comparsa del tratto a velocità quasi costante, passando dalla pressa A alla pressa C. Pertanto, se l’obiettivo è quello di usare la pressa per un processo di imbutitura, la soluzione C è la migliore. Naturalmente, sfruttando la parametrizzazione del modello, è possibile individuare configurazioni con prestazioni ulteriormente migliorate.

Per maggiori informazioni: Fabiano Maggio - EnginSoft info@enginsoft.it L’esempio del meccanismo link drive solleva una serie di problematiche tipiche della modellazione multibody. Infatti, l’utente deve scegliere con cura numero e tipologia di vincoli se non vuole pervenire a risultati incompleti o addirittura errati. L’utilizzo di strumenti come “ANSYS Transient Structurla MBD” presuppone che l’utente possieda adeguate nozioni di meccanica applicata e calcolo numerico che gli consentano di tradurre correttamente un sistema fisico in un modello virtuale. La schematizzazione può avvenire in modo più o meno raffinato, con conseguenze dirette sull’efficacia della simulazione. È compito del modellista scegliere dimensione, grado di complessità e dettagli del modello che vuole creare, considerando simultaneamente obiettivi da raggiungere, onere computazionale e tempo a disposizione.

Il miglior modello non è quello più dettagliato, ma quello che risponde in modo più veloce ed esauriente alle esigenze.

Questa regola, che vale in generale per tutte le dimensioni del CAE, assume un ruolo decisivo per la simulazione multibody.

EnginSoft propone un corso di modellistica multibody della durata di 2 giorni a tutti i progettisti che affrontano quotidianamente problemi di cinematica e dinamica. Il corso è pensato e strutturato in modo da trasferire in breve tempo le conoscenze che servono a formulare consapevolmente le principali scelte di modellazione multibody. Il corso verrà tenuto dal prof. Roberto Lot dell’Università di Padova in collaborazione con l’ing. Fabiano Maggio di EnginSoft.

Per informazioni sui contenuti consultare il sito del consorzio TCN www.consorziotcn.it Per iscrizioni e informazioni generali consultare la sig.ra Mirella Prestini della segreteria del consorzio. E-mail: info@enginsoft.it Tel: 035 368711

Newsletter EnginSoft Year 8 n°1 -

13

Ottimizzazione termofluidodinamica di un forno da cucina Indesit Indesit Company è tra i leader in Europa nella produzione e commercializzazione di grandi elettrodomestici: lavabiancheria, asciugabiancheria, lavastoviglie, frigoriferi, congelatori, cucine, cappe, piani di cottura e forni.

Proprio su quest’ultimo prodotto, il forno da cucina, si è recentemente concentrata un fase di sviluppo volta a migliorarne efficienza in termini di consumi e di uniformità di cottura.

Lavorando già da tempo con strumenti di modellazione numerica (nello specifico ANSYS ICEM CFD e ANSYS CFX) Fig. 1 per coadiuvare le prove sperimentali, Indesit Company ha deciso di utilizzare tali strumenti per tutta la fase di ottimizzazione della termodinamica interna del forno, avvalendosi del software di ottimizzazione modeFRONTIER e della collaborazione dei tecnici della EnginSoft S.p.A. A valle di studi sperimentali e considerazioni basate su know how interno, si è pensato che, per migliorare le prestazioni del forno dal punto di vista energetico e funzionale, l’attenzione maggiore doveva essere posta sulla paratia forata che si trova tra la ventola e la zona di cottura. Il lavoro presentato, è quindi quello dell’ottimizzazione geometrica di tale paratia, alla ricerca della configurazione tale da avere minori consumi ed una maggiore uniformità di temperatura nella muffola.

lativi alla dimensione degli elementi della griglia di calcolo, sia a livello globale che locale. Per tutte le configurazioni analizzate, 3 strati di prismi sono stati estrusi su tutte le pareti del dominio di calcolo. Le griglie di calcolo che tipicamente si possono ottenere con questa impostazione constano di circa due milioni di elementi tetraedrici ed un milione di prismi. È chiaro tuttavia che il numero di elementi è leggermente diverso per ogni configurazione, essendo la geometria parametricamente variata per ogni design.

L’attività sul modello baseline, oltre che rappresentare il riferimento per quantificare il margine e la direzione del miglioramento nella fase di ottimizzazione, è servita anche per la fase di taratura, indispensabile in attività di questa portata per determinare il miglior compromesso tra numero di elementi, qualità degli stessi e numerica più efficace all’ottenimento di risultati affidabili del calcolo CFD. Variabili di Input ed Output modeFRONTIER è un ottimizzatore multi disciplinare e multi obiettivo. L’esperienza che vanta EnginSoft nell’utilizzo di questo strumento accoppiato a software di analisi numerica, ha consentito la messa a punto di un flusso logico di macro-operazioni che hanno portato ad una vera e propria “customizzazione” per il problema qui illustrato (ottimizzazione forno).

I parametri geometrici di ingresso sono stati in tutto sedici. Essi hanno permesso di controllare dimensione, distribuzione e numero di fori sui quattro bordi della paratia. Lo spazio delle possibili configurazioni è stato delimitato in tre modi: dai valori minimi e massimi che ciascuna variabile di input doveva rispettare, da vincoli di costruzione, e da vincoli geometrici che evitassero geometrie degeneri.

Modellazione Si è partiti dalla modellazione di tutto il “volume bagnato” del forno nella configurazione baseline di partenza. Il forno, in questa configurazione, presenta anche una leccarda nella parte inferiore della zona di cottura. Sfruttando il fatto che la parte geometrica soggetta a modifica parametrica è la sola paratia, si è pensato di dividere il volume in due parti con un piano che separi il dominio di calcolo in prossimità della paratia. In questo modo, la parte di modello a valle della paratia, comprendente Discorso un po’ più dettagliato meritano le variabili di output. A anche tutta la leccarda, rimane invariata, e viene perciò prepamonte del lavoro di ottimizzazione è stata valutata molto attenrata (geometria + griglia di calcolo) una sola volta. L’altra parte tamente la modellazione delle variabili in uscita. Esse infatti dedel modello, è a sua volta suddivisa in un volume che rimane invono rappresentare un indice sia dell’efficienza del forno in tervariato, il dominio rotante con la ventola, e tutto il resto, commini di uniformità della temperatura che del suo consumo di prendente fra gli altri la paratia, la cui geometria e griglia di calcolo sono state parametrizzate per poter essere rigenerate automaticamente di volta in volta sulla base delle scelte operate dal’ottimizzatore.

Il software utilizzato per le modifiche geometriche e la generazione della griglia di calcolo è stato ANSYS ICEM CFD. Mettendo a punto una sequenza di istruzioni opportune, ICEM modifica la geometria della paratia e genera la griglia del volume ottenuto. In questo insieme di istruzioni (script) sono presenti oltre a tutti i parametri geometrici, anche quelli re- Fig. 2

14 - Newsletter EnginSoft Year 8 n°1 energia. Si è deciso di scegliere come grandezza per l’uniformità di temperatura, il suo scarto quadratico medio (RMS di T) misurato su una nuvola di punti situata “ad hoc” nella zona di cottura, mentre per la potenza elettrica necessaria si è optato per la portata d’aria ricircolante all’interno della muffola. Con queste assunzioni, gli obiettivi implementati in modeFRONTIER diventano perciò la ricerca di un design che renda minimo il valore dello scarto quadratico medio delle temperature sulla nuvola di punti garantendo allo stesso tempo una portata d’aria superiore ad un va- Fig. 3 lore limite precedentemente stabilito come limite inferiore.

Il processo logico che lega tutti i passaggi è il seguente: • Partendo dal file replay.rpl che modeFRONTIER ha aggiornato con i valori delle variabili di ingresso, lo script, eseguito in batch da ICEM, si occupa di modificare la paratia con i fori, copiare il resto della geometria che resta immutata e generare la griglia di calcolo.

• Terminata questa prima fase, la griglia di calcolo generata viene caricata assieme a quella che resta inalterata nel preprocessor di CFX. Al modello così aggiornato viene quindi applicato il setup fisico e numerico dell’analisi cfd, scritto il file di lancio e lanciato il run, specificando eventualmente se il calcolo deve essere eseguito in modalità parallela. Anche questa fase è interamente eseguita in modalità batch da tutti i software coinvolti.

• Finita l’analisi, sempre in batch, un altro script del post-processore di CFX, calcola le grandezza di output dal file di risultati, concludendo così l’iterazione per il singolo design Doe e Ottimizzazione Il DOE (Design Of Experiment) di partenza è stato realizzato mediante l’algoritmo RANDOM tra i più adatti per un’ottimizzazione mono-obiettivo. In effetti anche se gli output sono due, solo la minimizzazione dello scarto quadratico medio della temperatura è un vero e proprio obiettivo, dato che il controllo del valore di portata d’aria smaltita dalla ventola sia sempre superiore ad un valore minimo è considerato come un vincolo del problema. Il numero di design iniziale, dipendente anche dal tipo di algoritmo scelto per l’ottimizzazione, è in questo caso calcolato come la somma dei parametri di ingresso più uno, come richiesto dall’algoritmo di ottimizzazione utilizzato.

A seguito della campagna di analisi sui risultati del DOE di partenza, è iniziata la fase di ottimizzazione, mediante l’utilizzo dell’SIMPLEX. Il numero di design simulati in questa fase, utili all’ottenimento di una buona convergenza dell’algoritmo stesso, è stato di 107.

Ottenuto il design “ottimo”, sfruttando le potenzialità di postprocessing e statistica presenti all’interno di modeFRONTIER è

stato possibile estrarre dalla considerevole mole di dati resisi disponibili, delle informazioni importanti per capire il legame tra le varibili di input e tra ciascuna di esse e l’obiettivo. Più in generale, questa fase consente di raccogliere delle preziose informazioni per capire più approfonditamente il problema studiato, soprattutto laddove esso dipenda da diversi input e output anche fra loro in forte contrasto, tipicamente con comportamento non lineare.

Strumenti quali matrici di correlazione, grafici a coordinate parallele, filtri, bubble multidimensionali, hanno consentito di stabilire quale sia il peso dei singoli ingressi sui risultati, e quale sia il migliore intervallo di utilizzo tra i valori ammissibili, per ogni variabile.

Infine, conoscendo come ogni variabile influisce sul risultato finale, partendo dal miglior design sono state implementate modifiche aggiuntive, che, integrando quelle previste dallo spazio parametrico sopra illustrato, hanno portato ad un ulteriore miglioramento delle performance del forno.

Dall’attività nel suo complesso, ne è scaturita una profonda conoscenza del fenomeno fisico, delle relazioni fra le variabili, con soddisfacenti riscontri sperimentali.

Risultati e Conclusioni Grazie alle simulazioni numeriche e all’esperienza dei tecnici Indesit nell’indirizzare decisioni e assunzioni da prendere, le prestazioni del forno da cucina sono migliorate nell’ordine del 20% sullo scarto quadratico medio rispetto alla configurazione iniziale.

Le prove sperimentali sul miglior design hanno confermato i risultati numerici ed un risparmio energetico dovuto ad una portata smaltita dalla ventola superiore a quella del design di partenza.

Gianluca Mattogno, Indesit - gianluca.mattogno@indesit.com Fabio Damiani, EnginSoft - info@enginsoft.it

Fig. 3

Newsletter EnginSoft Year 8 n°1 -

15

Combined 1D & 3D CFD Approach for GT Ventilation System analysis The Gas Turbine ventilation system is designed to supply the necessary amount of air for cooling and to prevent the accumulation of hazardous gases in the enclosure by maintaining a slight over-pressure. The classical GE approach to studying ventilation system operating conditions consists of modeling the whole system as a series of discrete losses, where the ASHRAE duct-fitting database provides the corresponding pressure loss coefficients. The system is solved by means of a onedimensional flow simulation tool (Flowmaster).

The goal of this work was to improve the critical points that affect the above-mentioned procedure, such as modeling of complex fittings and bend interactions. For this purpose, dedicated CFD analyses were performed to characterize the loss coefficient for splitting and bend interactions at different operating conditions (split percentage and inlet flow rate) for two different ventilation systems. The resulting loss coefficient curves have been implemented within the corresponding onedimensional Flowmaster models. Finally, to characterize off-design conditions, a variable Heat Rejection model (obtained from previous CFD analyses) and real fan curves were used.

This new approach produces more accurate results, as confirmed by the close agreement with experimental measurements. Among the benefits of using this new approach is the ability to characterize the flow behavior of complex fittings. This would be useful in the event of a fitting redesign or for noise reduction analyses.

Current GE approach to studying Gas Turbine Ventilation Systems A ventilation system must provide a continuous source of cooling air over the entire Gas Turbine operation range in order to: • maintain a uniform and constant airflow through the flange-to-flange Gas Turbine at all ambient conditions; • remove heat and maintain the air temperature in the compartment below the operating limit. (The operating limit is set according to the temperature rating of the components located in the compartment); • eliminate stagnation zones and prevent the accumulation of hazardous gases; • prevent the ingress of dust and sand in gas turbines located in regions prone to sandstorm conditions by means of proper compartment pressurization.

Specific Design Practices provide a general description, acceptance limits and design criteria that a ventilation system must meet for Oil & Gas applications (e.g., enclosure design temperature ranges, design pressure ranges, purging ranges, etc.).

As mentioned, the current GE approach to studying GT ventilation systems consists of modeling the whole system as a series of “blocks”. Each block represents a source of pressure loss (concentrated loss) due to changes in shape (e.g., elbow, transition, etc.), flow direction or the presence of physical obstacles within the system. The ASHRAE duct-fitting database provides the corresponding pressure loss coefficients.

Following the net balancing by means of a onedimensional flow tool (Flowmaster), the system is characterized in terms of velocities, pressures, and flow rate split.

Critical points for this approach are the modeling of complex fittings and bend interactions. In order to improve the current Ventilation System calculation procedure, dedicated CFD analyses were performed for these critical points. A combined 1D & 3D CFD approach was adopted to study two different GE Ventilation Systems, called for simplicity System A and System B. Numerical calculations for System A The current System A Flowmaster network, modeled as a series of discrete losses, is shown in Figure 1. The

Fig. 1 - System A Flowmaster model based on discrete losses.

16 - Newsletter EnginSoft Year 8 n°1 enclosure is modeled as two heaters and the fan as two 3 flow sources with a flow rate of 65000 m /h, estimated by using the enthalpy balance equation:

K12 (GT Compartment)

K13 Load Compartment

ASHRAE Database

0.4

0.71

modified ASHRAE model (experience based)

1.15

1.5

CFD

0.4-0.6

2.28

Table 1: Loss coefficients used for standard calculations, System A.

were: = mass air flow [Kg/s], = enclosure heat rejection [W] = specifiwec heat at constant pressure [J/Kg °C] = maximum allowable outlet air temperature [°C] = max ambient temperature [°C] In order to develop a more suitable model (taking into account interactions, 3D characteristics of the fluid, etc.), dedicated ANSYS FLUENT CFD analyses were performed. In particular, a critical point for the discrete losses modeling is the flow split into the Load Compartment and the Gas Turbine Compartment (see Figure 2).

Fig. 2 - Analyzed split (left) and Load Compartment final section (right), System A

defined in (2), as a function of the flow rate split (see Figure 3).

Subsequently, these coefficients were implemented within the corresponding one-dimensional Flowmaster model.

A comparison between the loss coefficients obtained using CFD and those coefficients used for standard calculations is summarized in Table 1.

Finally, to better simulate the ventilation system a bend interaction analysis was performed on the Load Compartment final section, which is highlighted in Figure 2 (for System B the geometry of this section is the same).

The total loss coefficient as a function of the inlet velocity is shown in Figure 4. The loss coefficient decreases as the inlet flow velocity increases, and a good agreement with the ASHRAE database value was found for a velocity of about 5m/s. For higher velocity values the difference between the two curves (CFD and ASHRAE) starts to be significant. Again, the loss coefficient curve obtained was implemented within the new model.

The fan, previously modeled as two flow sources, was replaced by the “FAN” element with the corresponding real operating curve.

The final System A Flowmaster model including the main differences from the standard approach is shown in Figure 5.

It is useful to define the coefficients K12 and K13 as:

; where: P01 = inlet total pressure P02 = GT Compartment total pressure P03 = Load Compartment total pressure V2 = GT Compartment mean velocity V3 = Load Compartment mean velocity For the characterization of the flow split at different operating points, two test campaigns were performed.

In both cases the inlet flow rate was 3 3 fixed (65000 m /h and 130000 m /h, respectively) and, for each of these, a variable split percentage between the GT and Load compartments was used.

These analyses provided K12 and K13,

The results obtained with the new model were compared with the results obtained by the ADV (Air Ducts and Ventilation) department using a model based on the ASHRAE loss coefficient with appropriate corrections based on experience and with the results obtained with a pure ASHRAE model (see Table 2). The reliability of each approach was evaluated through comparison with experimental data.

Fig. 3 - K12 and K13 as a function of flow rate split for two different inlet flow rates, System A.

Newsletter EnginSoft Year 8 n째1 -

17

lower than 5%). On the contrary, the pure ASHRAE model produced an error of 18%.

Numerical calculations for System B Also for the System B split, several tests were performed to determine the split loss coefficients for different operating conditions. Figure 7 shows K12 and K13 as a function of the split flow rate percentage between the GT and Load compartments for an inlet flow rate equal to 3 70000 m /h (design flow rate). As one can see, both curves follow a linear trend.

Fig. 4 - Loss coefficient curve for Bend Interaction.

Similar to the System A model, the new System B Flowmaster model contains the loss coefficient curves obtained from CFD analyses (including the bend interaction curve) and the real fan operating curve.

Finally, in order to better simulate the heat removal, the heat rejection was modeled as a function of the mass flow rate, in accordance with recent studies performed by the SYS-OPT (System Optimization) department, that is:

where: HR = heat rejection HR0 = reference heat rejection = mass flow rate 0 = reference mass flow rate n = reference exponent

Fig. 5 - New System A Flowmaster model.

Table 2 summarizes the results obtained for the enclosure pressure. Using the new approach we got a favorable level of approximation with respect to the measured value (error equal to 7%). The other two approaches yielded errors higher than 25%.

The final System B Flowmaster model is shown in Figure 8.

Figure 9 shows the GT and Load Compartment velocity obtained with the STD model (previous calculations) and the new model for dirty and clean filter house conditions.

Figure 6 shows for each model the load compartment velocity and the corresponding error from the measured value at clean filter house conditions. The measured mean velocity is 12.47 m/s.

Both the new model and the modified ASHRAE model (experience-based) led to a high level of agreement (error Enclosure Pressure[mmH2O]

Measured value[mmH2O]

Error[%]

Discrete loss model (ASHRAE)

54.40

43.0

26.5

Discrete loss model (Experience based)

54.86

43.0

27.6

New model (Flowmaster+CFD)

40.00

43.0

-7.0

Table 2: Enclosure pressure, clean filter house conditions, System A.

Fig. 6 - Load compartment velocity, clean filter house conditions, System A.

18 - Newsletter EnginSoft Year 8 n째1 In both cases, the load compartment velocity obtained with the new approach is significantly higher than the old value (+49%). In particular, for the new approach, we got a split of 89-11% compared to a value of 92.7-7.3% obtained from previous calculations.

Considering that the target flow rate is 90-10%, the new approach again provides more accurate results.

Fig. 7 - K12 and K12 as a function of flow rate split, System B.

No significant variations between the two approaches in terms of enclosure pressure and temperature were found.

Conclusions In this work, a combined 1D and 3D numerical approach was adopted to study two GE ventilation systems.

This approach, compared to the current one-dimensional approach, improves the simulation of the actual operating conditions in terms of inlet flow rate, duct velocity and enclosure Fig. 8 - New System B Flowmaster model. pressure, as confirmed by the References close agreement with experimental [1] Miller, D.S.: Internal Flow Systems; 2nd Edition, Miller measurements.

Innovations, 2008 Among the benefits of using this new approach is the [2] Idelchik I.E.: Handbook of hydraulic resistance, 3rd ability to characterize the flow behavior of complex Edition, CRC Begell House, 1994 fittings. This would be useful to support the redesign of [3] ASHRAE Duct Fitting Database, Version 2.5.0. ASHRAE fittings or for noise reduction analyses.

About GE Oil & Gas GE Oil & Gas (www.ge.com/oilandgas) is a world leader in advanced technology equipment and services for all segments of the oil and gas industry, from drilling and production, LNG, pipelines and storage, to industrial power generation, refining and petrochemicals. We also provide pipeline integrity solutions, including inspection and data management, and design and manufacture wire-line and drilling measurement solutions for the oilfield services segment. As part of our 'Innovation Now' customer focus and commitment, GE Oil & Gas exploits technological innovation from other GE businesses, such as Aviation and Healthcare, to continuously improve oil and gas industry performance and productivity. GE Oil & Gas employs more than 12,000 people worldwide and operates in over 100 countries.

L. Barbato, M. Blarasin, S. Rossin GE Oil & Gas, Via Felice Matteucci 2, Florence, Italy

Figure 9 - GT and Load compartment velocity for STD and New approach, System B.

Newsletter EnginSoft Year 8 n°1 -

19

ANSYS WB and a review of the design metrics in Piaggio: the case of the motor shaft I still can remember the time when, looking at a 3D CAD model of an engine block, I would start thinking about the best way to translate it into a PREP7 procedure. I would come up with something to mesh, but the next time I would have to start from scratch again. In those times, beam representations in conjunction with SIFs were the best way to go with crankshafts. Other components required similar efforts.

Things evolved in a continuous fashion, but a discontinuity came when ANSYS changed its face completely with Workbench. At first I thought that dealing with it would have been a dive into a bottomless ocean, just like the first time I met a CONTA174. But I’ve always been a fundamentalist when it comes to new CAE techniques, so I tried to move to WB as quickly as I could and to the maximum possible extent.

And my way of working radically changed: geometry import and meshing issues sharply decreased, and past

models could be used as templates for new, similar analyses. This last aspect evolved dramatically with the introduction of WB projects, where bunches of interconnected analyses form now real CAE procedures, laid down with nearly no effort.

Such a case happened just a few weeks ago when I came up against a crankshaft simulation.

I had to use WB both as stand-alone application and as part of a CAE chain, including MBS and durability analyses.

The simulations I had to perform required both linear and nonlinear models, involving the simulation of neighboring components, in addition to the crankshaft proper. WB allowed me to quickly setup a baseline model: DM fixed a few CAD issues and Mechanical Automatic Contact detection feature greatly speeded up the assembly setup.

The CAD interface can sense CAD simplified representations, allowing to perform partial CAD imports, really useful when dealing with big CAD models.

Generating all the other models I needed from the baseline one was really easy at a project level, duplicating when a different topology was needed and linking when only different load systems or different analysis types were required.

That way, I could assess both the frictional load transfer capability and the fatigue performance of the crankshaft assembly.

For the former I used nonlinear models, exploiting the WB contact features, whose default settings are much more error-proof than navigating among all the keyoptions and

20 - Newsletter EnginSoft Year 8 n°1 bodies included in an assembly, of the contact interfaces, or the available postprocessing quantities, to name a few.

From a table you can jump to the relevant model element with a simple click. So Worksheets prove to be a nice tool to check what you’ve done.

The fatigue analyses required the computation of load/stress transfer functions and of a Craig-Bampton modal representation [1]. The latter was calculated by means of a simple combination of rigid Remote Displacement features and of a Commands object: no more messing around with CERIGs, since WB did the job totally in the background.

The results of the activity were not only fatigue safety factors and stress distributions; besides them, and above all, a neat trace of what I did has been left both at a Project and System level, in the WB jargon. The next time a similar component will have to be simulated, it will be an update process, not a generation one. Opening the WB project, the modeling procedure will be easily recognizable.

real constants of the good old CONTA family. I could check the functionality of friction couplings with both standard and custom postprocessing quantities. The latter are easily definable with the aid of another WB feature: the Worksheets.

With them you have an overview of your modeling stages in a neat tabular form. You can check the properties of the

In the past the CAE techniques had a hard time trying to be simultaneously fast and accurate; that slowed their integration into the development process of complex systems. I think that WB has been one of the major milestones in overcoming these problems, therefore allowing the simulations to be perceived as standard and required activities.

For more information: Roberto Gonella - EnginSoft info@enginsoft.it [1] R.R. Craig, M.C.C. Bampton - Coupling of substructures for dynamic analyses - AIAA Journal,vol. 6, n.7, 1968

Newsletter EnginSoft Year 8 n°1 -

21

EnginSoft interviews Ing. Paolo Nesti, Piaggio Group

EnginSoft intervista l’ing.

Paolo Nesti del Gruppo Piaggio

ABOUT THE PIAGGIO GROUP The Piaggio Group was founded in 1884 and nowadays is the European leader and one of the major players worldwide in the sector of twowheeler vehicles. The Group is also a global leader in the development and manufacture of commercial vehicles. The product range of the Piaggio Group includes scooters, motorbikes and motorcycles from 50 to 1,200 cc and the trademarks: Piaggio, Vespa, Gilera, Aprilia, Moto Guzzi, Derbi, Scarabeo. In the light commercial vehicle market, Piaggio is represented with its three- and four-wheeler vehicles and the trademarks: Ape, Porter and Quargo.

The Group’s head office is based in Pontedera (in the province of Pisa, Italy). Since 2003, it is managed by the industrial holding Immsi S.p.A. Roberto Colaninno is the President and Managing Director of the Piaggio Group.

The manufacturing plants are located in: Pontedera (Pisa), Noale and Scorzè (Venice), Mandello del Lario (Lecco), Martorelles (Barcelona, Spain), Baramati (India, in the Maharashtra nation), and Vinh Phuc (Vietnam).

Furthermore, there is a joint venture in China (in Foshan, in the province of Guangdong). In 2009, the Piaggio Group sold 607.700 vehicles worldwide: 410,300 two-wheeler vehicles and 197.400 commercial vehicles. Motorcycle racing is a very important business for the Piaggio Group. Some Piaggio models are well known for some world records: 95 world championships won in different fields and more than 500 victories in different competitions. Among the most successful brands is Aprilia. With its 45 global qualifications and 278 wins in the GP, Aprilia is the most successful brand in the history of the GP Motorcycle Racing in Italy and Europe.

IL GRUPPO PIAGGIO: PROFILO Il Gruppo Piaggio, fondato nel 1884, è il più grande costruttore europeo e uno dei principali player mondiali nel settore dei veicoli motorizzati a due ruote. È inoltre protagonista internazionale nel settore dei veicoli commerciali. La gamma di prodotti del Gruppo comprende scooter, moto e ciclomotori da 50 a 1.200cc con i marchi Piaggio, Vespa, Gilera, Aprilia, Moto Guzzi, Derbi, Scarabeo, e veicoli commerciali leggeri a tre e quattro ruote con le gamme Ape, Porter e Quargo.

Il Gruppo ha sede a Pontedera (Pisa, Italia) e dal 2003 è controllato da Immsi S.p.A., holding industriale facente capo a Roberto Colaninno, che ricopre la carica di Presidente e Amministratore Delegato del Gruppo Piaggio.

Sul piano della produzione, il Gruppo Piaggio opera nel mondo con una serie di stabilimenti situati a: Pontedera (Pisa), Noale e Scorzè (Venezia), Mandello del Lario (Lecco), Martorelles (Barcellona, Spagna); Baramati (India, nello stato del Maharashtra), Vinh Phuc (Vietnam).

Il Gruppo Piaggio opera inoltre con una società in joint venture in Cina (a Foshan, nella provincia del Guangdong). Il Gruppo Piaggio nel 2009 ha complessivamente venduto nel mondo 607.700 veicoli di cui 410.300 nel business due ruote e 197.400 nel business dei veicoli commerciali.

Di grande rilievo, per la produzione motociclistica del Gruppo Piaggio, sono le attività racing. Il Gruppo vanta infatti, nel proprio portafoglio di brand, marchi facenti parte a pieno titolo della storia del motociclismo sportivo mondiale, con un palmarès complessivo di 95 campionati mondiali conquistati nelle varie specialità e oltre 500 vittorie nelle varie specialità. Tra i marchi del Gruppo, Aprilia con 45 titoli mondiali e 278 vittorie nei G.P. è il marchio italiano e europeo più vincente nella storia del Motomondiale.

The engineer Paolo Nesti graduated in Mechanical Engineering. Since 1988, the focus of his work has been on engines. Presently he is responsible for the design of engines and for computational systems at the Centro Tecnico Motori 2 Ruote of Pontedera. This Piaggio Group Center is dedicated to the development of two-wheeler vehicle power engines.

1. What is or should be the role of innovation in the industrial and entrepreneurial world?

Nowdays the market has enlarged itself and Global competition is our new work environment and we are not

L'ing Paolo Nesti è laureato in ing. Meccanica, in Piaggio dal 1988, si è sempre occupato di motori. Oggi è il responsabile della Progettazione Motori e Sistemi di Calcolo del Centro Tecnico Motori 2 Ruote di Pontedera, dove si sviluppano tutti i motopropulsori per i veicoli a 2 ruote del Gruppo.

1. Che spazio ha (e dovrebbe avere) l’innovazione nel mondo industriale/impresariale?

La competizione globale a cui siamo chiamati non può es-

22 - Newsletter EnginSoft Year 8 n°1 able to win our challenge playing on the simple grounds of cost reduction. A large group like ours, which operates on all major world markets, must be in a position of competitive advantage based on providing products more attractive to the customer, most original, high quality: we can achieve this goal looking at the expressed and latent Customers requirements even if they are also extremely different from themselves because economic, lifestyle and consumption reasons: in a word, we absolutely must innovate. Only in this way we can create products that ensure profitable business of the Company. Our group was very clear this need.

Finally in order to achieve this goal, anything but simple, we must put attention to the market, to extensive technical expertise often interdisciplinary, to ability to express and synthesize new solutions, to fast implementation of ideas into products, to process control and to organization.

2. What are the strategies for innovation and quality assessments pushing innovation?

In order to innovate the right prerequisites are the following ones: • to be very knowledgeable about both the product and customer; • to have the expertise on technology and product knowledge of the best competitors; • to be ensure that the work environment is "creative", giving space to persons who may contribute to the creation of ideas and having care to them encouraging their development aims by providing preferential channels for such projects.

The MP3 version and Hybrid are strictly examples of this approach: the original idea was quickly passed to the development objective by providing our customers the best technology available on the market today (from a vehicle structure enhanced by a revolutionary hybrid engine based on a 4-stroke ie Euro3 with integrated management of the two engines, ride-by-wire, lithium batteries, plug-in without external power supply, electric reverse, etc.)…

sere giocata sul semplice terreno della riduzione dei costi.

Un grande Gruppo come il nostro, che opera su tutti principali mercati mondiali, deve porsi in una posizione di vantaggio rispetto alla Concorrenza basandosi sull'offerta di prodotti più attraenti per il Cliente, più originali, di elevata qualità globale, intendendo con questo termine anche e soprattutto la rispondenza ai bisogni espressi o latenti di Clienti anche estremamente diversi tra di loro per esigenze, contesti economici, stili di vita e di consumo: in una parola, deve fare innovazione. Solo in questo modo potrà realizzare prodotti profittevoli che garantiscano l'attività dell'Azienda. Il nostro Gruppo ha ben chiara questa necessità.

Per raggiungere questo obiettivo, tutt'altro che semplice, occorrono attenzione al mercato, vaste competenze tecniche spesso interdisciplinari, capacità di esprimere e sintetizzare nuove soluzioni, velocità nell'implementazione delle idee in prodotti, controllo dei processi ed organizzazione.

2. Quali sono le strategie per essere innovativi e quali valutazioni spingono all’innovazione?

Per poter fare innovazione bisogna prima di tutto essere profondi conoscitori sia del prodotto sia della clientela, avendo le competenze specifiche sulle tecnologie e la conoscenza dei prodotti dei migliori concorrenti; occorre poi far sì che l’ambiente di lavoro sia “creativo”, dando spazio alle persone che possono contribuire alla nascita delle idee e favorendone lo sviluppo predisponendo canali preferenziali per tali progetti.

L’MP3 prima e la versione Hybrid dopo sono un esempio di questo tipo di approccio: dall’idea originale si è passati in breve tempo allo sviluppo finalizzato, mettendo a disposizione dei clienti la miglior tecnologia disponibile ad oggi sul mercato (un veicolo dalla struttura rivoluzionaria impreziosito da un motore ibrido basato su un 4T i.e. Euro 3, dotato di gestione integrata dei due propulsori, rideby-wire, batterie al litio, plug-in senza alimentatore esterno, retromarcia elettrica, etc.).

3. Che ruolo ricoprono gli strumenti CAE e di prototipazione virtuale in tal senso?

L'innovazione in campo industriale parte da un'idea che, attraverso stadi successivi di maturazione, si trasforma in uno o più prodotti che generano profitto per l'impresa. Questo processo non è spontaneo; passa bensì per una serie di verifiche ed eventuali correzioni, che determinano il tempo necessario per passare da idea a profitto. È quindi cruciale che queste attività siano veloci senza perdere accuratezza; gli strumenti CAE contribuiscono ad accelerare il processo, consentendo di ridurre i cicli di sperimentazione fisica che richiedono notevole dispendio di tempo e quin-

Newsletter EnginSoft Year 8 n°1 -

3. What role do the CAE and virtual prototyping tools in this regard?

Innovation in industry starts with an idea that, through successive stages of maturation, changes in one or more products that generate profits for the company. This process is not spontaneous, but it is composed by multiple checks and corrections, which determines the time it takes to go from idea to profit.

Therefore it is crucial that these activities are fast without losing accuracy and the CAE tools help speed up the process, reducing, for example, the cycles of physical experimentation saving considerable time and therefore money.

In the early stages of development of an innovative idea the checks mentioned above are possible only at a conceptual level, the details are not yet available for physical prototypes At this level the simulation is the only way to correct any abuses of the process of maturation.

4. How user needs have changed in recent years?

User requirements have been prompted by a significant level of integration of CAE techniques and actually these technologies are completely integrated in the process design of product development.

It obviously implies that the complexity of the simulation increase and, at the same time, the numerical results should be produced in more little time; and it seems to be a paradoxical situation.

CAE Users therefore need intensive tools interfaced with those used in other business areas such as CAD software and processing of experimental data, another requirement is the ability to simulate not only the separate components, but also systems to study the interactions between components. This work field requires software tools fast, robust and accurate, and hardware with the maximum possible power from various points of view (processors, RAM, CPU, etc.)… All these things are strictly necessary without ever losing sight of the overall objective and having clear theoretical principles that govern the software used.

5. What are the advantages pointed out in his professional experience and how has it changed its approach to the design/production?

The advantages are actually well known to all actors (engineers, managers): first of all the savings in time and obviously money, having care to make virtual models (CAD, Multibody, FEM) analysis and CFD studies for engine performance, analysis of cooling, etc…

The primary purpose of this type of activity, but not the only one, is to minimize the construction of physical prototypes, thus avoiding the long series of validation tests, limiting this activity to a stage "ripe" for the project.

The construction of prototypes takes place when the phase

23

di di denaro. Nelle fasi precoci di elaborazione dell'idea innovativa le verifiche prima citate sono inoltre possibili solo a livello concettuale, non essendo ancora disponibili i dettagli per realizzare prototipi fisici: a questo livello la simulazione rappresenta l'unico strumento per correggere eventuali derive del processo di maturazione.

4. Come sono cambiate le esigenze degli utilizzatori negli ultimi anni?

Le esigenze degli utenti sono state indotte dal sensibile livello d'integrazione che le tecniche CAE hanno ottenuto nel processo di sviluppo dei prodotti. Questa circostanza ha richiesto e continua a richiedere simulazioni sempre più complesse e, sebbene questo possa sembrare paradossale, sempre più veloci. Gli utenti CAE hanno perciò bisogno di strumenti fortemente interfacciati con quelli usati in altre aree aziendali, come i software CAD e di elaborazione dei dati sperimentali; un'altra esigenza è la possibilità di simulare non solo componenti isolati, ma anche sistemi, per studiare le interazioni fra componenti. Questo quadro richiede strumenti software veloci, robusti e accurati, e hardware con la massima potenza possibile sotto vari punti di vista (processori, memoria RAM, CPU, ecc.).

Il tutto senza mai perdere di vista l’obiettivo globale ed avendo chiari i principi teorici che sovrintendono gli applicativi utilizzati.

5. Quali vantaggi ha rilevato nella sua esperienza professionale e come è cambiato il suo approccio alla progettazione/produzione?

I vantaggi sono oramai noti a tutti: prima di tutto il risparmio di tempo e quindi di denaro facendo modelli virtuali (CAD, Multidody, FEM), analisi e studi CFD per le prestazioni dei motori, analisi di raffreddamento, etc.

Lo scopo primario di questo tipo di attività, ma non il solo, è ridurre al minimo la costruzione di prototipi fisici evitando così la lunga serie di prove di validazione, limitando questa attività ad una fase “matura” del progetto.

La costruzione dei prototipi avviene quando la fase di calcolo (e di ottimizzazione) è stata completata, riducendo così la probabilità di dover apportare costose correzioni

24 - Newsletter EnginSoft Year 8 n°1 calculation (and optimization) has been completed, thus reducing the likelihood of having to make costly adjustments to physical parts.

The second key issue concerns the quality of the project: using CAE tools intelligently, we can quickly arrive at a definition of optimal projects, thus laying the foundation for a quality product.

The third aspect is rewarding management and sharing of information: in this development process, it is essential to share information in real time and records management, to speed up internal processes of decision, correction, approval, etc… and then once again reduce the risk of introducing errors.

6. What was the contribution of EnginSoft and how it has helped to enhance quality, capability and capacity of its industry/company?

EnginSoft was invaluable in training, both basic and learning new software, even with the help of TCN. Basic education is key to preventing senseless and unconscious use of CAE tools: a real risk because of the simplicity and widespread use of software that claims to be based on complex theoretical apparatus.

EnginSoft has also played and still plays a vital role in the innovation process, enabling access to technical and software embedded in a significant way with other development activities, and promoting effective networking among users, for example conferences and meetings with the resonance increasing year after year.

7. What prospects he sees for the calculation codes in relation to the challenges of the future?

It 'is now widely believed that competitive, if not survival, of different firms in the Old Continent and in the western world, in general, must be based on knowledge and innovation content of the products: the old standards of competitiveness are indeed falling down because the real big power of the new countries and these new ‘world’ will become strong players in the world economy. The instruments that promote the enrichment of science and technology products are becoming so indispensable to be

su pezzi fisici. Il secondo aspetto fondamentale riguarda la qualità del progetto: utilizzando con intelligenza gli strumenti CAE, si può arrivare rapidamente ad una definizione ottimale dei progetti, ponendo così le basi per un prodotto di qualità.

Il terzo aspetto premiante è la gestione e la condivisione delle informazioni: in questo processo di sviluppo è fondamentale la condivisione delle informazioni in tempo reale e la gestione della documentazione per velocizzare i processi interni di decisione, correzione, approvazione, etc. e quindi ridurre ancora una volta i rischi di introdurre errori.

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 stata preziosa nel campo della formazione, sia a livello base che di apprendimento di nuovi software, anche con il contributo di TCN. La formazione di base è fondamentale per evitare usi scriteriati e inconsapevoli degli strumenti CAE: rischio concreto e diffuso a causa della semplicità d'uso di software che pure si basano su apparati teorici complessi.

Enginsoft ha inoltre ricoperto e tuttora ricopre un ruolo basilare nell'ambito dell'innovazione di processo, rendendo possibile l'accesso a tecniche e software integrabili in modo significativo con le altre attività di sviluppo, e promuovendo con efficacia il networking fra utenti, per esempio con conferenze e incontri di risonanza crescente anno dopo anno.

7. Che prospettive intravede per i codici di calcolo in relazione alle sfide poste dal futuro? È ormai convinzione diffusa che la competitività, se non la sopravvivenza, delle imprese del Vecchio Continente e del mondo occidentale in genere dovrà poggiare sulla conoscenza e sul contenuto innovativo dei prodotti; i vecchi canoni di competitività si stanno infatti sfaldando di fronte ai paesi che sono divenuti protagonisti dell'economia mondiale. Gli strumenti che favoriscono l'arricchimento scientifico e tecnologico dei prodotti stanno divenendo perciò indispensabili per riuscire a rimanere nel mercato; i codici di calcolo appartengono a tale categoria ed è perciò prevedibile che la loro evoluzione da prodotti di nicchia a strumenti di uso corrente si completi nei prossimi anni, accompagnata dall'accelerazione della loro potenza ed efficienza, resa possibile dall'analoga tendenza in campo hardware.

8. Quali progetti, obiettivi e nuovi traguardi intende perseguire grazie all’uso di questi strumenti?

Ritengo che nel medio termine non si verificheranno cambiamenti sostanziali nel-

Newsletter EnginSoft Year 8 n°1 -

25

able to stay in the market, the computer codes belong to this category and it is therefore like absolutely sure that their evolution from niche products to use current tools will be completed in the nextg years, and it will be accompanied by the acceleration of their power and efficiency, and it will be made possible from the analogous trend in the field hardware.

8. What plans, objectives and new goals will be pursued through the use of these tools?

I believe that in the medium term there will be no substantial changes in the use of CAE tools in Piaggio; during trouble-shooting in phase its usage has become a minority compared to the prediction, and it is evident that the products are developed today with growing contribution of the simulations.

To predict the spread of multi-disciplinary optimization techniques will be Piaggio road map in this area, and the integration between CAE tools available today from various related subjects could push this design-way just as it happens between fluid dynamics and mechanics cold thematic.

The number of experimental procedures for the qualification and a relative fee CAE activity will increase in the next future, so that physical tests will be conducted on prototypes already developed and optimized in terms of simulation, thereby increasing the likelihood of success of the test.

In general, the use of CAE techniques will "dominate" more and more products in the sense that you can know more deeply the physical behavior, with clear benefits in terms of quality and reliability.

9. And what we hope for the world of scientific technology to the continuing search for a dimension of creativity and competitiveness?

The synthesis between creativity and competitiveness gets through the tools that technology provides. The task of scientific technology is just to make simple evaluations and complex activities that hinder the creativity of the people, thus removing the obstacles that typically prevent you from looking up to new ideas. It will be increasingly necessary to have integrated multidisciplinary tools, in order to further improve the aspects that are now viewed individually.

Optimizers, being able to interact with different disciplinary tools, they have a high development potential that any company can develop applications specifically for doing their business. Also in this area Piaggio has already made some experiences and others are being set (modeFRONTIER and application specific simulation engine) to maximize engine performance and reduce development time for new products.

l’uso degli strumenti CAE in Piaggio: l'uso in fase troubleshooting è divenuto minoritario rispetto a quello predittivo, indicando che i prodotti vengono sviluppati oggi con contributo crescente da parte delle simulazioni.

Prevedo per Piaggio una diffusione delle tecniche di ottimizzazione multi-disciplinare, favorita dall'integrazione spinta oggi disponibile fra strumenti CAE afferenti da diverse materie, come ad esempio la fluidodinamica e la meccanica fredda.

Aumenterà il numero di procedure di qualifica sperimentale con un corrispettivo CAE, in modo che i test fisici siano condotti su prototipi già studiati e ottimizzati a livello di simulazione, aumentando così la probabilità di successo della prova.

In generale, l'uso delle tecniche CAE consentirà di "dominare" sempre più i prodotti, nel senso che sarà possibile conoscerne sempre più a fondo il comportamento fisico, con evidenti benefici in termini di qualità ed affidabilità.

9. E cosa si auspica per il mondo della tecnologia scientifica alla continua ricerca di una dimensione tra creatività e competitività?

La sintesi tra creatività e competitività passa proprio attraverso gli strumenti che la tecnologia mette a disposizione. Compito della tecnologia scientifica è appunto quello di rendere semplici le valutazioni e le attività complesse che ostacolano la creatività delle persone, rimuovendo così gli ostacoli che tipicamente impediscono di alzare lo sguardo verso nuove idee. Sarà sempre più necessario avere strumenti multidisciplinari integrati, per poter migliorare ancora gli aspetti che ora sono visti singolarmente.

Gli ottimizzatori, potendo interagire con diversi strumenti multidisciplinari, hanno un elevato potenziale di sviluppo che ogni azienda può sviluppare facendo applicazioni ad hoc per le proprie attività. Anche in questo settore Piaggio ha fatto già alcune esperienze e altre sono in corso di impostazione (modeFRONTIER e applicativi specifici di simulazione motore) per massimizzare le prestazioni dei motori e per ridurre il tempo di sviluppo dei nuovi prodotti.

26 - Newsletter EnginSoft Year 8 n째1

modeFRONTIER Used in the Design of Fatigue-Resistant Notches evolved variable-radius notches in trees, bones, etc; the German engineer Claus Mattheck showed that similar concepts could be applied to engineering structures.

Professor David Taylor at Trinity College Dublin in Ireland wondered whether the variable-radius notch could be treated as an optimisation problem, and decided to use mF to solve it. Working with Matteo Toso and Professor Luca Susmel of the University of Ferrara in Italy, they considered the problem of a 90째 fillet and used mF to seek for solutions within a design space consisting of different variable-radius fillets. They were able to find solutions better than those previously obtained using other methods, achieving reductions in the maximum stress of more than a factor of two.

Fig. 1 - Researchers at Trinity College Dublin used modeFRONTIER in the Design of Fatigue-Resistant Notches

Researchers at Trinity College Dublin in Ireland have used modeFRONTIER (mF) software to reduce stress concentration effects of notches and thus significantly improve the fatigue resistance of components. Many engineering components contain features such as notches and fillets, which are usually designed with a constant radius of curvature.

However it has long been known that this is not the best solution.

Variable-radius notches, in which the radius of curvature changes with position along the notch, can achieve much lower stress concentration factors with negligible change in the weight or size of the component. Nature has

Experimental work conducted on steel samples showed that these predicted reductions translated exactly into real improvements in the fatigue strength of the components. Reductions in stress concentration factors can be highly beneficial, allowing designers to save weight, with consequent reductions in energy and material costs, without sacrificing reliability. The approach developed at Trinity College could be automated for use in industrial design, using mF in conjunction with FEM, to achieve real improvements in components of the future.

Prof. David Taylor M.R.I.A., Mechanical Engineering, Trinity College Dublin For more information, please contact: Professor David Taylor, DTAYLOR@tcd.ie

Department of Mechanical and Manufacturing Engineering The Department of Mechanical and Manufacturing Engineering undertakes research in a number of selected themes, including; Bioengineering, Fracture and Fatigue of Materials, Fluids, Acoustics and Vibration, Fluids and Heat Transfer, Manufacturing Technology and Systems, and Tribology and Surface Engineering.

Fig. 2 - Finite element analysis of a variable-radius fillet in a bracket component.

http://www.tcd.ie/mecheng/research/

Newsletter EnginSoft Year 8 n°1 -

27

A Multi-Objective Optimization with Open Source Software Sometimes it happens that a small-to-medium sized firm does not benefit from the advantages that could be achieved through the use of virtual simulation and optimization techniques. This represents in many cases a great limitation in the innovation of products and processes, and this can lead, in a very short time, to a complete exclusion from the market and to an inevitable end.

Nowadays, it is mandatory to reduce as much as possible the time-to-market, while always improving the quality of products and satisfying the customer needs better that the competitors. In some fields it is a question of “life or death”.

According to our opinion, the main reasons that limit or, in the worst case, make impossible the use of the virtual simulation and optimization techniques can be grouped into three categories: 1. These techniques are not yet sufficiently known and the possible users do not have a great confidence in the results. Sometimes physical experimentation, guided by experience maturated through many years of practice, is thought to be the only possible way to proceed. This is actually wrong in the great majority of cases, especially when new problems have to be solved. A change of vision is the most difficult but essential step to take in this context.

Rough Phase

Fine Phase

License

Many possibilities are available

GNU license largely used or similar versions with some restrictions

Development

Continuous improvement and a clear guideline

Left to the community

Available features

State of the art

It strongly depends on “who” leads the development. Sometimes, very advanced features can be available.

Technical support

Usually the distributor offers a technical support

Usually no support is available but in some cases forums can help

Usability

Easy-to-use and smart GUIs

Some effort could be required to the user

Customization

Only in some cases

If the source code is available the possibility of customization and development is complete

Table 1: The table compares some key features that characterize commercial and open source software, according to our opinion.

2. Adequate hardware facilities considered necessary to perform an optimization are not available and therefore the design time becomes too long. We are convinced that, in many cases, common personal computers are enough to efficiently solve a large variety of engineering problems. So, this second point, which is often seen as an enormous obstacle, has to be considerably downsized.

3. The simulation software licenses are much too expensive given the firm’s financial resources. Even though the large majority of commercial software houses offer a low-cost first entry license, it is not always immediately evident that these technologies are not just an expense, but rather a good investment.

As briefly stated above, the second point often does not represent a real problem; the most important obstacle is summarized in the first point. People actually find it hard to leave a well-established procedure, even if obsolete, for a new one which requires a change in the everyday way of working. The problem listed in the third point can be solved, when possible, by using open source (see [1]), free and also home-made software. It is possible to find, with an accurate search on internet, many simulation software systems which are freely distributed by the authors (under GNU license in many cases). Some of them also exhibit significant features that usually are thought to be exclusive to commercial software.

As usual, when adopting a new technology, it is recommended to consider both the advantages and the disadvantages. We have compared in Table 1 some aspects that characterize the commercial and the open source codes which should be considered before adopting a new technology.

Open source codes are usually developed and maintained by researchers; contributions are also provided by advanced users all over the world or by people who are supported by research projects or public institutions, such as universities or research centers. Unfortunately, this can lead to a discontinuous improvement, not driven by a clear guideline, but rather left to the free contributions given by the community. On the contrary, commercial software houses drive the development according to wellknown roadmaps which generally reflect specific industry trends and needs.

Commercial software is usually “plug-and-play”: the user has just to install the package and start using it. On the contrary - but not always - open source software could

28 - Newsletter EnginSoft Year 8 n째1 only open source software, are presented in require some skill and effort in this paper. We decided to use Scilab (see compiling the code or adapting a [2]) as the main platform to drive the package to a specific system configuration. optimization process through its genetic Software houses usually offer to the algorithm toolbox. For the solution of the structural problem, presented in the customer technical support, which following, we adopted two packages. The can be, in some cases, really helpful first one is the Gmsh (see [3]) to manage a to make the investment profitable. An internet forum, when it exists, is the parametric geometry and mesh it; the only way to have support for a user of second one is CalculiX (see [4]), an FEM solver. It is important to remember that an open source code. this choice is absolutely not mandatory, but Another important issue is the usability of the simulation software, is strictly a matter of taste. which is mainly given by a userfriendly graphical interface (often The structural optimization problem In this work a C-type press is considered, as referred to as GUI). The commercial software usually has sophisticated the one shown in Figure 2. This kind of geometry is preferred to other ones when graphical tools that allow the user to Fig. 1 - An example of C-type press. The easily manage and explore large steel C-shaped profile which will be optimized the force that has to be expressed by the in this work is highlighted with a red line. hydraulic cylinder is not very high, usually models in an easy and smart way; the not greater than roughly 200 [Ton]. The main advantages open source codes rarely offer a similar suite of tools, but of this type of press are essentially the relative low weight they have simpler and less easy-to-use graphical and volume of the machine and the possibility of interfaces. accessing the workbench The different magnitude of the investment can explain all from three sides. these differences between the commercial and open The dimensioning of the source codes. lateral C-shaped profiles is probably one of the However, there are some issues that can make the use of most challenging phases a free software absolutely profitable, even in an industrial in the design process; context. Firstly, no license is needed to run simulations: the large majority of the in other words, no money is needed to access the virtual weight and cost, for the simulation world. Secondly, the use of open source mechanical part at least, software allows to break all the undesired links with third is actually concentrated party software houses and their destiny. Third, the number there. Consequently, a of simultaneous runs that can be launched is not limited, good designer looks for and this could be extremely important when performing the lightest profile which an optimization process. Last, but not least, if the source code is available all sorts of customizations are in is able to satisfy all the Fig. 2 - The C-shaped plate geometry has structural requirments. been modeled using the dimensioning principle possible.

Moreover, an economical drawn in this picture, together with the The results of a structural optimization, performed using configuration is also thickness TH of the plate.

Plate thickness [mm] Plate max dimensions [m] Available steel codes desired, in order to reduce as much as possible the production cost.

Vertical <4 20 A, B, C Horizontal <2 When optimizing, the designer should also take into 30 account some aspects which are not strictly related to Vertical <3 40 structural issues but are however important or, in some A, B Horizontal <2 50 cases fundamental, to deal with an optimal design. These aspects could be related to the availability of materials Table 2: The table collects some limitations in the steel plate provision. and components on the market, technical norms that have to be satisfied, marketing indications and more. In our Young modulus Maximum stress / Steel code Cost [$/Kg] case the steel plate supplier, for example, can provide only [MPa] Yield limit [MPa] the configurations collected in Table 2.

A 200 (220) 1.2 It is clear that an optimization process that does not take 210000 B 300 (330) 2.3 into consideration these requisites could lead to configurations which are optimal only from a theoretical C 600 (630) 4.0 point of view, but which cannot be practically Table 3: The maximum desired stress, the yield limit and the cost per unit implemented. For example, a very light configuration is weight for the three available steels.

Newsletter EnginSoft Year 8 n째1 -

Fig. 3 - A possible version of the C-shaped plate meshed in Gmsh.

Fig. 4 - The CalculiX GraphiX window, where the von Mises stress for the Cshaped plate is plotted.

not of practical interest, if it requires a steel characterized by a yield limit greater that 600 [MPa].

For this reason all the requirements collected in Tables 2 and 3 have been included in order to drive the optimization algorithm to feasible and interesting solutions.

Moreover, it is required that the hollow in the plate (H2max(R1,R2) x V2, see Figure 2) is at least 500 x 500 [mm] to allow the positioning of the transversal plates and the hydraulic cylinder.

Another technical requisite is that the maximum vertical displacement is less than 5 [mm] to avoid excessive misalignments between the cylinder axis and the structure under the usual operating conditions. This limit has been chosen arbitrarily, in the attempt to exclude the designs that are not sufficiently stiff, taking into account, however, that the C-plate is a part of a more complex real structure which will be much more stiff than what is calculated with this simple model.

A designer should recognize that the solution of such a problem is not exactly trivial. Firstly, it is not easy to find a configuration which is able to satisfy all the requisites listed above; secondly, it is rather challenging to obtain a design that minimizes both the volume of the plate (the weight) and the production cost.

29

The open source software for the structural analysis Gmsh has been used as a preprocessor to manage the parametric geometry of the C-shaped plate and mesh it in batch mode. Gmsh has the ability to mesh a non-regular geometry using triangular elements; many controls are available to efficiently define the typical element dimension, the refinement depth and more. It is a very powerful software tool which is also able to manage complicated threedimensional geometries and efficiently mesh them using a rich element library.

The mesh can be exported in a formatted text file where the nodes and the element connectivities are listed together with some useful information related to the so-called physical entities, previously defined by the user; this information can be used to correctly apply, for example, boundary conditions, domain properties and loads to a future finite element model.

The CalculiX finite element software has been used to solve the linear elastic problem. Also in this case a batch run is available; among the many interesting features that this software offers are the easy input text format, and the ability to perform both linear and non-linear static and dynamic analyses.

CalculiX also offers a pre and post processing environment, called CalculiX GraphiX, which can be used to prepare quite complicated models and, above all, display results.

These two software tools are both well documented and also some useful examples are provided for new users. The input and output formats are, in both cases, easy to understand and manage.

In order to make the use of these tools completely automatic, it is necessary to write a procedure that translates the mesh output file produced by Gmsh into an input file readable by CalculiX. This translation is a relatively simple operation and it can be performed without a great effort using a variety of programming languages; a text file has to be read, some information has to be captured and then rewritten into a text file using some simple rules. For this, a simple executable file (named translate.exe) has been compiled and it will be launched whenever necessary.

A similar operation has also to be performed in an optimization context to extract the interesting quantities from the CalculiX result file and rewrite them into a compact and accessible text file.

As before, an executable file (named read.exe) has been produced to read the .dat results file and write the volume, the maximum vertical displacement and the nodal von Mises stress corresponding to a given design into a file named results.out.

Many other open source software codes are available, both for the model setup and for its solution. Also for the results visualization, there are many free tools with powerful features. For this reason the interested reader

30 - Newsletter EnginSoft Year 8 n°1 can imagine the use of other tools to solve this problem in an efficient way.

These routines are extremely flexible and they can be modified by the user according to his or her own needs, H1 250 150.

5 since the source code is available. This H2 500 1500 5 The optimization process driven is exactly what we have done, V1 250 1500 5

by Scilab modifying the optim_moga.sci script to V2 500 1500 5 The genetic algorithm toolbox, by handle the constraints (with a penalty V3 250 1500 5 Yan Collette, is available in the approach) and manage the infeasible R1 50 225 5 standard installation of Scilab designs efficiently (i.e.: all the R2 50 225 5 and it can be used to solve the configurations which cannot be multi-objective optimization computed); we have then redefined the TH 20 50 10 problem described above. This Table 4: The lower and upper bounds together with the coding_ga_binary.sci to allow the step for the input variables. toolbox is composed of some discretization of the input variables as listed in Table 4. Other max vM max vertical Volume Design H1 H2 V1 V2 V3 R1 R2 TH Cost stress displacement small changes have been name [mm] [mm] [mm] [mm] [mm] [mm] [mm] [mm] [$] [mm ] [MPa] [mm] made to the routines to A 670 665 575 500 490 165 110 20 1304 577.7 4.93 3.53â&#x20AC;˘10 perform some marginal operations, such as writing B 1155 695 725 545 840 185 165 30 1097 199.8 1.73 1.06â&#x20AC;˘10 partial results to a file.

Table 5: The optimal solutions.

When the genetic algorithm requires the evaluation of a routines which implement both a MOGA and a NSGA2 given configuration, we run a Scilab script which is algorithm and also a version for the operations that have charged to prepare all the text files needed to perform the been performed when running a genetic algorithm, that is run and then launch the software (Gmsh, CalculiX and the the encoding, the crossover, the mutation and the other executables) through a call to the system in the selection. right order. The script finally loads the results needed by the optimization.

Variable

Lower bound [mm]

Upper bound [mm]

Step [mm]

3

7

8

Fig. 5 - The Cost of the computed configurations can be plotted versus the Volume. Red points stand for the feasible configurations while the blue plus indicates the configurations that do not respect one constraint at least.

The two green squares are the Pareto (optimal) solutions (A and B in Table 5).

Fig. 6 - The vertical displacement for the design A.

Fig.7 - The von Mises stress for the design A.

It is important to highlight that this script can be easily changed to launch other software tools or perform other operations whenever necessary.

In our case, eight input variables are sufficient to completely parameterize the geometry of the plate (see Figure 2): the lower and upper bounds together with the steps are collected in Table 4. Note that the lower bound of variable V2 has been set to 500 [mm], in order to satisfy the constraint on the minimal vertical dimension required for the hollow.

We can use a rich initial population, (200 designs randomly generated), considering the fact that a high number of them will violate the imposed constraints, or worse, be unfeasible. The following generations will however consist of only 50 designs, to limit the optimization time.

Newsletter EnginSoft Year 8 n°1 -

31

After 50 generations we obtain the results plotted in Figure 5 and Table 5, where the two Pareto (optimal) solutions are collected. We finally decided to choose, between the two optimal ones, the configuration with the lowest volume (named as “A” in Table 5).

In Figures 6 and 7 the vertical displacement and the von Mises Fig. 9 - The von Mises stress for the modified design. stress are plotted for the Fig. 8 - The vertical displacement for the modified design. optimal configuration named “A” (see Table 5). Note that during the optimization, the maximum value of the von input variables, since the output does not strongly Mises stress computed in the finite element Gauss points dependent on them, and this leads to a simpler has been used, while in Figure 7 the von Mises stress optimization problem. extrapolated by CalculiX to the mesh nodes is plotted; this The cost does not change; actually it represents the cost is the reason why the maximum values are different. of the rectangular plate needed to produce the C-shaped However, they are both less than the yield limit profile. corresponding to the steel type C, as reported in Table 3.

Other configurations with a lower volume can be probably Horizontal and vertical max vM max Vertical Volume found with some other runs; however, the reader has to Cost displacement length of cuts starting stress [$] [mm ] consider that these improvements are not really [mm] from corners [mm] [MPa] significant in an industrial context, where, probably, it is H1/3 1304 581.3 4.80 3.33•10 much more important to find optimal solutions in a very short time.

Table 6: The modified design. It can be seen that there is an interesting 3

7

reduction in the volume with respect to the original design, the “A” configuration in Table 5. Other output quantities do not present significant variations.

Another interesting consideration is that the Pareto front in our case consists of just two designs: this shows that the solution of this optimization problem is far from trivial.

The design of the C-shaped plate can be further improved.

If we run other generations with the optimization algorithm better solutions could probably be found, but we feel that the improvements that might be obtained in this way do not justify additional computations.

Substantial improvements can be achieved in another way.

Actually, if we look at the von Mises stress distribution drawn in Figure 7 we note that the corners of the plate do not have a very high stress level and that they should not influence the structural behavior very much.

A new design can be tested, cutting the corners of the plate; for the sake of simplicity we decided to use four equal cuts of horizontal and vertical dimensions equal to H1/3, starting from the corners. The results are drawn in Figures 8 and 9, which can be compared with Figures 6 and 7.

As expected, there is a reduction in volume with respect to the original design, but no significant variations are registered in the other outputs. This corroborates the idea that the cut dimensions can be excluded from the set of

Conclusions In this work it has been shown how it is possible to use open source software to solve a non-trivial structural optimization problem.

Some aspects which characterize the commercial and the open source software have been stressed in order to help the reader to make his or her own best choice. We are convinced that there is not a single right solution but rather that the best solution has to be found for each situation.

Whichever the choice, the hope is that virtual simulation and optimization techniques are used to innovate.

References [1] Visit http://www.opensource.org/ to have information on open source software [2] Scilab can be freely downloaded http://www.scilab.org/ [3] Gmsh can be freely downloaded http://www.geuz.org/gmsh/ [4] Calculix can be freely downloaded http://www.calculix.de/

more from: from: from:

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

32 - Newsletter EnginSoft Year 8 n°1

ANSYS 13: Il punto sui solutori per modelli di grandi dimensioni nelle simulazioni meccaniche Una delle grandi sfide per i software di simulazione basati sul metodo degli elementi finiti è l’efficienza nel trattare modelli di grandi dimensioni, e quindi: • di conservare i dettagli presenti nei modelli CAD; • di avere risultati accurati; • di disporre di algoritmi veloci e robusti per la meshatura; • di disporre di CPU sia su desktop che su cluster.

In particolare è sempre più vero che il costo della CPU è inferiore al costo uomo, e cioè che, in presenza di algoritmi affidabili per la meshatura di modelli di grandi dimensioni, conviene ricorrere a questi piuttosto che svolgere un pesante lavoro di “defeaturing” e di meshatura semiautomatica.

via numerica nel campo delle applicazioni meccaniche. Nel passato infatti l’elaborazione numerica è sempre stata assegnata alle potenzialità della CPU in una architettura in cui la CPU svolgeva un ruolo centralizzato. Oggi, invece, i principali produttori di schede grafiche hanno messo a punto nuove architetture che permettono incrementi consistenti delle prestazioni grafiche: la novità sta nella capacità di sfruttare il gran numero di ALU (Arithmetic Logic Unit) per eseguire algoritmi in parallelo. Seguendo la comparsa nel tempo di queste soluzioni si può far riferimento alle schede progettate da nVIDIA con tecnologia CUDA e dalla ATI con tecnologia STREAM. ANSYS a sua volta ha sviluppato algoritmi paralleli che eseguono calcoli in doppia precisione sfruttando le risorse relative alla tecnologia “GPU based”.

In generale e per comprendere l’attenzione dedicata da ANSYS al problema di accelerare l’analisi si può far riferimento anche alle tecnologie disponibili nelle precedenti versioni. Queste tecnologie, del resto, sono state rese ancora più efficaci nell’ambito degli aggiornamenti sviluppati da ANSYS. Si richiamano: SMP (Shared memory Processing), tecnologia consistente in un’architettura di più CPU che condividono la stessa memoria.

DMP (Distributed Memory), tecnologia consistente in un’architettura di più CPU con una porzione di memoria dedicata a ciascuna CPU.

All’aspetto dei costi si affianca - ed è più importante di questo - l’aspetto dell’affidabilità dei risultati. Esso dipende in larga parte dalla qualità del modello e quindi anche della mesh.

La release 13 del codice ANSYS Mechanical offre una risposta molto convincente a questi problemi, trattati in ottica industriale. Infatti: • orienta all’High Performance Computing consentendo l’uso di CPU diverse; • produce modelli schematizzati automaticamente in maniera adeguata; • contiene acceleratori specifici.

La novità della release R13 di ANSYS, infatti sta nel saper adattare il nuovo paradigma di processamento del calcolo per

Se si confrontano queste due architetture solamente sulla base del tempo di calcolo risulta più efficiente la tecnologia DMP, sua volta più costosa dal punto di vista dell’hardware.

Il vantaggio della tecnologia DMP rispetto al SMP risulta però inferiore se il tempo di calcolo è misurato come intero “elapsed time”, intendendo con questo il tempo relativo all’intero processo incluse le fasi precedenti e successive all’analisi vera e propria.

Se quindi si ragiona in base al “dpd” (design produced per day) il rapporto di efficienza DMP/SMP è meno significativo.

Più precisamente questo confronto è valido fino ad 8 CPU perché al di sopra di questo numero la SMP satura ed la DMP diventa oltremodo vantaggiosa anche in termini di “dpd”.

Altra tecnologia del software ANSYS per ridurre i tempi di esecuzioni delle analisi è la VT (Variational Technology) che implementa gli algoritmi di Taylor, Pade’ e Roms.

Newsletter EnginSoft Year 8 n°1 -

Essa non può però essere utilizzata in tutti i casi. Si applica ad analisi termiche, ad analisi in frequenza e anche ad alcuni problemi di non linearità strutturali quali le tematiche di creep. Non possono, però, essere presenti nel modello elementi CONTACT e questo, per certi versi, è il limite di tale approccio.

Infine è utile ricordare quanto certe metodologie di analisi di ben solida ed antica applicazione siano utili e rinnovabili nelle situazioni più complesse e quanto queste stesse tecniche possano beneficiare comunque delle generali accelerazioni legate all’aumento delle capacità di calcolo (esempio “GPU technique”).

Ricordandole brevemente esse sono: • Submodelling: valutazione locale del gradiente delle tensioni in situazioni alla De Saint Venant • Substructuring: riduzione del modello a condensazioni di masse e rigidezze e ripetizione delle stesse per parti del modello ripetitive

• CMS: Component Mode Synthesis è una specie di sottostrutturazione con accoppiamento delle zone di interfaccia tramite equazioni di ‘coupling’ I tre sistemi sono stati rivisti nell’ottica dei miglioramenti progettati per la nuova release.

Per concludere uno sguardo al futuro: le tecnologie per l’accelerazione della velocità di calcolo richiedono parallelamente il potenziamento dell’hardware. Gli sviluppi previsti in ANSYS tengono in conto l’evoluzione nell’hardware per migliorare le prestazioni. La release 13 rappresenta quindi un primo passo nell’adattamento delle metriche della tecnologia GPU, finalizzato a migliorare la scalabilità del software in tutte le applicazioni seguendo così quanto già fatto con successo ad esempio nelle applicazioni relative alla CFD. Per maggiori informazioni: Emiliano D’Alessandro - EnginSoft info@enginsoft.it

La simulazione di sistema in ANSYS: Simplorer La release 13 di ANSYS lega nello stesso ambiente le tecnologie per la simulazione elettromagnetica in bassa ed alta frequenza prodotte da Ansoft. Di queste, abbiamo dato informazione in edizioni precedenti della newsletter.

Ci occupiamo qui invece di Simplorer, la tecnologia per la simulazione di sistema in ANSYS. Simplorer è un software di simulazione Multi-Domain che consente di modellare, simulare, analizzare e ottimizzare sistemi complessi, come sistemi elettromagnetici, elettromeccanici, elettrotermici, e più in generale, meccatronici e cibernetici.

Se quindi da un lato ANSYS fornisce strumenti per la modellazione e la simulazione di singoli componenti in diverse fisiche e discipline, dall’altro, attraverso una tecnologia come Simplorer, essa consente la simulazione a livello di sistema. In altre parole vengono messe a disposizione metodologie e tecniche affinché singoli componenti possano essere analizzati simultaneamente in un unico modello, tenendo in considerazione le mutue interazioni tra di essi.

L’utilizzo delle caratteristiche di modellazione di Simplorer consente quindi ai progettisti di realizzare prototipi virtuali considerando tutti gli aspetti ed i componenti di un sistema, quali ad esempio i componenti elettronici, i sensori, gli attuatori, i motori elettrici ed i generatori, i propulsori ibridi, i convertitori di potenza così come i con-

33

Fig. 1 - Simplorer in interfaccia ANSYS WB.

Fig. 2 - Modelli e linguaggi per la simulazione di sistema in Simplorer.

34 - Newsletter EnginSoft Year 8 n°1

Fig. 3 - Link dinamico tra HFSS-Simplorer (a) e SIwave-Simplorer (b).

trolli ed i software embedded.

Una tale metodologia si realizza attraverso l’implementazione di schemi circuitali, modelli analitici ed a parametri concentrati, circuiti equivalenti, tecniche di cosimulazione, reti multi-livello ecc…

Di seguito vengono analizzate alcune delle principali caratteristiche di Simplorer: 1) Tecniche di modellazione Simplorer offre diverse tecniche di modellazione inclusi circuiti, diagrammi a blocchi, macchine a stati, equazioni e linguaggi di modellazione come il linguaggio VHDL-AMS, il SML (Simplorer Modeling Language) ed il C/C++.

L’impiego simultaneo di tali strumenti consente di modellare sistemi caratterizzati da segnali analogici, digitali o analogico-digitale. Questo approccio elimina la necessità di effettuare trasformazioni matematiche, tipicamente soggette ad errori.

Una tale flessibilità nelle tecniche di modellazione fa di Simplorer uno strumento molto efficace all’interno di un gruppo di lavoro poiché professionisti di diversa estrazione tecnica possono far confluire modelli realizzati con linguaggi diversi all’interno dell’unica piattaforma di simulazione di Simplorer.

In Figura 2 una sintesi delle tecniche di modellazione a disposizione di Simplorer.

ti realizzati in SIwave e HFSS. Tale procedura si basa sulla valutazione e caratterizzazione a parametri S di un modello FEM. Tipicamente, una volta definite opportunamente le porte di input ed output nei modelli agli elementi finiti, SIwave e HFSS consentono infatti di esportare la matrice di scattering verso Simplorer.

Come mostrato in Figura 3 il link SIwave-Simplorer consente di effettuare in Simplorer analisi di Signal integrity su schede PCB, mentre l’integrazione di HFSS permette l’analisi elettromagnetica di sistema anche in alta frequenza.

Le tecniche con le quali è possibile includere modelli realizzati con altri software di casa ANSYS sono in particolare due: La tecnica della cosimulazione e la tecnica della model order reduction (MOR).

La cosimulazione o “Co-Simulation” (co-operative simulation) è una metodologia di simulazione che consente a componenti singoli di essere simulati in maniera simultanea da differenti software. Questa tecnologia permette quindi ai due software di scambiarsi informazioni, quali ad esempio boundary conditions o time steps, in maniera collaborativa e sincronizzata.

2) Physics-based modeling Per modelli per i quali è richiesto un elevato livello di accuratezza, Simplorer fornisce un link diretto ad altri software ANSYS, tra questi: Maxwell, Q3D Extractor, RMxprt, PExprt, HFSS, SIwave, ANSYS Icepak, ANSYS Rigid Dynamics e ANSYS Mechanical.

In particolare dall’ultima versione (Simplorer 9.0.1) è possibile integrare all’interno della simulazione di Simplorer i modelli agli elementi fini-

Fig. 4 - Control Design per il modello multy-body di un braccio di un escavatore in ANSYS WB.

Newsletter EnginSoft Year 8 n°1 -

35

analisi parametriche, di ottimizzazione, di sensitività, e di tuning al fine di ottenere un progetto ottimizzato, in relazione a criteri di performance fissati, raggiungendo il miglior trade-off possibile.

In questo senso Simplorer può usufruire di tutta la potenza di calcolo a disposizione perché in grado di effettuare analisi distribuite.

In particolare, dall’ultima release, Optimetrics viene incluso all’acquisto nel pacchetto software.

Fig. 5 - Analisi di sistema e circuit coupling in ANSYS.

Questo tipo di tecnica è implementabile con i modelli di Maxwell e con i modelli Multy Body di ANSYS, per i quali Simplorer fornisce gli eventuali controlli (Fig 4).

La tecnica della Model Order Reduction (MOR) è una disciplina della teoria dei sistemi e dei controlli che studia le proprietà dei sistemi dinamici in modo tale da ridurne la complessità, preservandone il comportamento agli ingressi e alle uscite (input ed output).

Attraverso la tecnica della Model Order Reduction è possibile trasferire nell’ambiente di simulazione di Simplorer modelli a parametri concentrati estratti da modelli agli elementi finiti realizzati tra gli altri in ANSYS Mechanical, ANSYS Fluent e ANSYS ICEPack.

In Figura 5 viene sintetizzato lo stato dell’arte dell’integrazione di sistema in ANSYS. 3) Cosimulazione con tool esterni ad ANSYS Programmi in C/C++, MATLAB® / Simulink®, ModelSim®, QuestaSim® e Mathcad® possono essere integrati direttamente in Simplorer attraverso la tecnica della cosimulazione.

Questo permette una semplice e rapida implementazione di modelli realizzati anche con software esterni al portafoglio dei prodotti ANSYS. La diretta integrazione dei modelli nel loro ambiente di simulazione evita la traduzione del modello, consente di risparmiare tempo per la progettazione, e permette la comunicazione e lo scambio di informazioni tra diversi progettisti.

4) Analisi Statistiche e di Ottimizzazione.

Optimetrics, tool embedded in Simplorer, consente di effettuare

5) Tools di caratterizzazione di power devices.

Simplorer supporta strumenti e sistemi per la caratterizzazione di dispositivi di potenza quali IGBT e converter AC/DC. Per quanto riguarda l’analisi degli IGBT, Simplorer fornisce due diversi metodi di caratterizzazione: dinamica e mediata (dynamic and average). La caratterizzazione dinamica consente una maggiore precisione nel descrivere i fenomeni di switching che avvengono in questi dispositivi, mentre la caratterizzazione average permette una modellazione del dispositivo tale da consentire tempi di simulazione più ridotti, fornendo comunque una stima delle perdite medie durante le fasi di switching.

Entrambe le metodologie descritte leggono gli input necessari alla caratterizzazione del dispositivo direttamente dai datasheet messi a disposizione dai fornitori.

Per quanto riguarda la definizione e la caratterizzazione di dispositivi elettronici inoltre è possibile accedere in rete (http://model.simplorer.com) ad una vasta libreria di modelli di componenti quali Diodi, MOFSFETs ed IGBTs. In Figura 6 viene illustrato un esempio di simulazione multi dominio in ANSYS Simplorer.

Per maggiori informazioni Emiliano D’Alessandro - EnginSoft info@enginsoft.it

Fig. 6 - Simulazione elettro-termica di un commutatore IGBT in Simplorer.

36 - Newsletter EnginSoft Year 8 n°1

ICEPAK 13.0: buone notizie per i progettisti elettronici I malfunzionamenti legati al surriscaldamento, le eccessive sollecitazioni termiche sulla struttura e la gestione non ottimale della distribuzione dei flussi di calore sono problemi comuni nell’ambito della progettazione dei dispositivi elettronici. Tutti questi problemi possono essere studiati e risolti utilizzando ANSYS Icepak.

Icepak è un codice di fluidodinamica computazionale robusto e completo creato specificatamente per il controllo termico dei dispositivi elettronici. Caratterizzato da un’interfaccia estremamente intuitiva, permette di eseguire analisi termo fluidodinamiche stazionarie e tempo transienti simulando tutte le modalità di trasferimento del calore (conduzione, convezione, radiazione e scambio termico coniugato).

Presente sul mercato da ormai più di dieci anni, Icepak è giunto alla release 13.0 che ha aggiunto nuove funzionalità in grado di aumentarne la facilità di utilizzo e di renderlo ancora più attraente al progettista elettronico.

Di seguito sono indicate le principali novità introdotte nella versione corrente suddivise per aree tematiche: Integrazione con ambiente Workbench Il graduale inserimento di Icepak all’interno dell’ambiente di lavoro Workbench, iniziato con la release 12.0, continua nella versione attuale.

In relazione alla fase di pre-processamento geometrico, sono stati implementati in Design Modeler alcuni strumenti (riuniti nel gruppo dei tools denominato Electronics – vedi Figura 1) che permettono di semplificare entità geometriche anche complesse e di convertirle direttamente in oggetti nativi di Icepak con un notevole risparmio di tempo da parte dell’utente.

Fig. 1 - Strumenti di semplificazione e conversione geometrie in Design Modeler: Electronics tools

Fig. 2 - Strumenti di semplificazione e conversione geometrie in Design Modeler: Electronics tools

Per quanto riguarda la fase di post-processing, tutte le funzionalià del visualizzatore interno ad Icepak sono ora disponibili nel software unico di postprocessing di Workbench (CFD-Post) che è uno strumento in generale più potente e in grado di gestire griglie di calcolo anche di notevoli dimensioni (vedi Figura 2).

Sempre nell’ambito di una progressiva integrazione di Icepak nell’ambiente di lavoro Workbench è da registrare il miglioramento dell’algoritmo di trasferimento del campo termico verso il solutore strutturale con una conseguente velocizzazione dei tempi di calcolo per le analisi FSI 1way (vedi Figura 3).

Mesh Il generatore di griglie di calcolo interno ad Icepak è stato ulteriormente sviluppato al fine di automatizzare il processo di meshing. Tra le nuove features ricordiamo: • La nuova opzione di meshing multi level (2D zero Cut Cell) permette un notevole risparmio di elementi di griglia nel caso di modelli 2.5D ed è molto utile per la modellazione delle tracce sulle printed circuit board (PCB) (vedi Figura 4); • il metodo di estrusione di griglia (Extruded Mesh) è stato implementato al fine di controllare il numero di celle posizionate nello spessore delle PCBs e dei Packages; • L’utilizzo degli O-Grid per la generazione della griglia di calcolo è stato ora esteso anche ad oggetti 2D (precedentemente era applicato solo a entità geometriche tridimensionali) con notevole beneficio sia della qualità della griglia che della risoluzione dei boundary layer termici; • È ora possibile creare mesh non conformi per i singoli componenti di un Assembly mettandoli a diretto contatto tra di loro (Zero Slack Assembly). Questo miglioramento nella gestione delle interfacce tra mesh porta ad una notevole riduzione del numero degli elementi da usare per discretizzare oggetti quali Heat sink, BGA etc. dove i

Newsletter EnginSoft Year 8 n°1 -

37

• Sono stati introdotti due nuovi modelli di radiazione: il “surface to surface” e il “ray tracing”. Il primo modello è semplice ed economico, adatto alla maggior parte delle applicazioni con mezzi trasparenti. Il secondo modello è molto più generale ed accurato ma dai costi computazionali elevati. Da sottolineare inoltre il supporto per il carico termico solare (compatibile con tutti i modelli di radiazione) che permette di tenere in conto della radiazione proveniente dall’ambiante esterno e del suo orientamento nello spazio.

Fig. 3 - Trasferimento diretto dei dati da Icepak ad ANSYS Meshanical

Fig. 5 - Icepak permette di eseguire analisi termiche a varie scale dimensionali

Fig. 4 - Multi Level 2D Cut Cell Meshing

singoli componenti hanno dimensioni caratteristiche molto differenti.

Modellazione Uno dei punti di forza di Icepak è la possibilità di generare modelli fisici rappresentativi dei singoli componenti elettronici da includere nella simulazione numerica quando il focus dello studio non è a livello della componentistica (scala [m]) ma a livello di sistema (scala [m]) – vedi Figura 5. Questo punto di forza è stato ulteriormente sviluppato con: • Il miglioramento del sistema di caratterizzazione dei packages elettronici denominato Delphi Extractor che permette di creare in modo automatico una rappresentazione RC (Resistiva Capacitiva) di packages quali BGA, Exposed Die BGA e QFP (Figura 6); • L’introduzione di una nuova macro per assistere nella modellazione delle Heat Pipes (Figura 7).

La versione 13 è inoltre supportata per oggetti 2D quali Fans, Grilles, Walls e Openings la geometria derivante da CAD. Ciò permette di utilizzare forme geometriche realistiche anziché approssimazioni a geometria poligonale per tali oggetti.

Solutore Le novità fondamentali del solutore possono essere riassunte in 2 punti:

Fig. 6 - Caratterizzazione dei componenti: Delphi Extractor

• È ora possibile importare ed esportare oggetti networks cioè modelli bidimensionali che rappresentare circuiti integrati tramite file CSV/Excel.

Per concludere questa visione di insieme della principali novità introdotte con Icepak 13 è da segnalare la possibilità di interagire in modo bidirezionale con SIwave, software per il calcolo dell’integrità di segnale. Icepak può ora dialogare con SIwave fornendo mappe di temperatura e ricevendo campi di potenza.

Interazione che risulta fondamentale per analisi dove le proprietà elettriche del dispositivo dipendono dalla temperatura.

Fig. 7 - Caratterizzazione dei componenti: Heat Pipes

Per esempi, materiale e richieste di informazioni: Ing. Matteo Nobili - EnginSoft info@enginsoft.it

38 - Newsletter EnginSoft Year 8 n°1

Development of the Novel Opencell™ A completely new metal sandwich panel concept, Opencell™, has been developed. Instead of the conventional three constituent panel structure (sheet/core/sheet), an integral cut-and-formed face sheet and the core, and a solid face sheet are used. This concept provides a reduction in the number of joining components and thus manufacturing phases can be decreased for a more cost-efficient process. An increased number of design variables means potential for tailored properties.

Unlike many traditional metal sandwich panels, the structure can have equal mechanical properties in the longitudinal and transversal directions, and in specific applications this concept provides stiffer solutions than the conventional sandwich panels.

TECHNOLOGY REVIEW Metal sandwich panels offer a number of outstanding properties allowing the designer to develop light and efficient structural configurations for a large variety of applications. The most established type of all-metal sandwich panels is the use of directional stiffeners

connected to each other using connection members formed from one of the face sheets, i.e. without any addition of the core material [4].

CONCEPT DEVELOPMENT The motivation of the product development naturally is increased performance. In principle, two approaches exist to develop more efficient structures: either the application of new materials or the use of novel structural design – or a combination of these. Opencell™ idea itself introduces a large number of design variables for tailored properties. With Opencell™ one can build panels with balanced transversal and longitudinal stiffness properties and challenging panel shapes are possible (see Figure 1).

Within certain limits, panel height can be increased without mass penalty. This means increased bending stiffness offering potential for weight savings.

The initial concepts were not very efficient in terms of structural performance and the project team focused on the concept development. Different geometrical layouts

Fig. 1 - The Opencell™ structure can take the form of flat, single side curved and doubly side curved shapes.

(see Figure 2), providing a range of structural performance between two solid face sheets, such as straight webs (Iand various types of packing patterns, were developed, core), (rectangular) hollow sections (O-core), hat or simulated using FEA, and evaluated for the proper corrugated sheets (Vf/V-core), etc. [1]. understanding of their mechanical behavior.

In contrast to the previous ones, called calottes, formed The purpose of the work was to make general comparative indents to separate the panel face sheets provide more analyses using a design study, for which a reference isotropic properties. However the closed nature of calottes application was adapted from an earlier sandwich project limits their maximum depth (formability) [2, 3].

Another developing direction of all-metal sandwich panel technology is towards lattice truss core structures. This has evolved into a completely new approach for a metal panel structure concept called Opencell™.

This invention, described in patent application WO/2009/034226, includes a panel structure wherein the core structure is formed from face Fig. 2 - Some of the studied Opencell™ concepts. Three right-most solutions represent Opencell Delta™ concepts. sheets that are mechanically All unit cells are in mutual scale, i.e. height of the panel is constant.

Newsletter EnginSoft Year 8 n°1 -

Table 1. Panel dimensions in the design studies.

[5]. This consisted of a beam-like supported 54 mm high panel with a support span L of 2 m and a uniform pressure load of 4800 Pa. A deflection constraint was set to L/300.

The objective was to meet this constraint with 4 mm steel consumption divided between the two components.

39

In the optimization problem the objective was to maximize specific stiffness of the panel (inverse of the panel maximum deflection divided by the total thickness of the two sheets).

The optimization procedure was set up in modeFRONTIER, the design optimization and process integration software package of our choice. The design of experiments and optimization algorithms provided in modeFRONTIER were used to drive the Opencell™ towards our goals.

Thus modeFRONTIER created new designs by determining input values for the variables which were fed to a fully parametric FE model in ANSYS. Results for a single design study are presented in the 4-D image of Figure 3.

Displacement results are presented as a function of the three variables for the different designs.

For illustration, displacement results were selected because they are not as abstract as the specific stiffness. The unit cell width is on the horizontal axis. The color of the bubble indicates the total thickness of the face sheets.

For example, cyan bubble (SUM_30…) indicates that the total amount is 3.0 mm. The stiffest structure in terms of real displacements naturally comes with the highest amount of material. Therefore, red bubbles (5.0 mm) are on top (least negative values). In the study, the total Fig. 3 - Bottom left: optimum layout for curved panel applications. Right: maximum displacement results as a function of the unit cell width for a flat panel (H 75 mm, span 3 m). Top left: optimum amount of material was restricted to 5.0 layout for the flat panel. mm. The diameter of the bubble indicates the thickness of the plain sheet. The smaller the bubble is, the thinner the sheet. Therefore, for this application, the OPTIMIZATION stiffest structure is obtained by maximizing the thickness The next step in the project was to gather more of the cut sheet. This is the trend particularly for flat information about the type of applications for which the panels with long support spans. Increased thickness on Opencell Delta™ structure is appropriate and to determine the cut sheet moves the neutral plane closer to the the optimum cell configurations in the different geometrical mid-plane and therefore, panels work better applications. The design study was divided to flat and in bending. For long span applications, bigger unit cell curved panel applications. Three corresponding panel size is preferable as it increases the amount of material at heights were selected to represent different design groups the surfaces. For short span applications, which are out(Table 1). Each design group was further on divided to of-plane shear dominated, smaller unit cells are three sub-groups determined by the span of the beam-like preferable. It should be noted that the plain sheet should panel. In case of flat panels, two opposite edges were be rigid enough to avoid local deformation close to the simply supported. For curved panels, also the longitudinal supports or loading points. translations were constrained on both edges. In both applications, panels were loaded with a uniform pressure Curved panels behave quite differently. They act load of 4800 Pa. essentially the same way as pressure vessels, where loads are carried by membrane forces. Therefore, the in-plane After the conceptual design studies it could be concluded stiffness dominates the applications and the fewer cuts that to achieve the best performance, certain geometrical per unit area, the bigger the membrane area. The best measures need to be driven towards the minimum or specific stiffness is obtained with the combination of the maximum allowed value. As a result, free design variables thickest plain sheet and the thinnest cut sheet, and with were limited to the choice of the unit cell width and the the highest value of the unit cell width. Still, the cut thickness combination of the two face sheets. Allowed sheet should be thick enough so that local deformations face sheet thicknesses for the two applications are and stability are not a problem. presented in Table 1.

A great performance improvement was achieved when the Opencell Delta™ concept was established in which four delta-shaped legs form the core. Also, it provided high unit cell packing density and consequently, rather homogenous structure.

40 - Newsletter EnginSoft Year 8 n°1

Fig. 4 - The software solution provides panel key properties such as EI and GA. User can solve the panel deflection U under different boundary and loading conditions. Stress recovery is made for a single unit cell with a separate model.

SOFTWARE SOLUTION Opencell Delta™ simulation capabilities have been integrated as a separate module in ESAComp, a software tool for the analysis and design of layered composite structures. The software module consists of three components: geometry modeler, homogenization tool and simulation tool.

The geometry modeler creates an FE mesh for a single unit cell according to the user-defined parameters (e.g. panel height, unit cell width, angle of the legs, etc.) and provides basic functions for visualization of the geometry.

The homogenization tool calculates the basic properties of the Opencell Delta™ panel. In the conceptual design phase, designers would like to compare different alternatives with some key properties, like panel axial stiffness, bending stiffness, and shear stiffness, which are derived internally. The panel properties can be used in commercial FE codes to calculate deflections of flat and curved panels with arbitrary shapes, boundary and loading conditions. The homogenization approach is described in detailed in [6].

The simulation tool integrated in ESAComp software supports the analysis of flat panels (see Figure 4). The ESAComp solution relies on Elmer solver [7]. Two types of analyses are supported: static load response and analysis of details that covers, for example, failure analysis.

CONCLUSIONS The introduced Opencell Delta™ concept provides a brand new way to construct metal sandwich panels. The concept gives potential for cost savings in manufacturing due to the reduced number of components and simplified continuous manufacturing process. In specific applications, increased mechanical performance can be achieved even with less material when compared to traditional metal sandwich panels, as could be shown optimizing Opencell Delta™ panels with modeFRONTIER. Both aspects are highly valued especially in transportation applications. Opencell™ panels can be formed in challenging shapes and they provide internal space for wirings, piping and other equipment, which may be required in the product. The presented software

solution provides an efficient way to perform panel dimensioning. With a very limited effort one can reliably estimate if the Opencell Delta™ concept brings benefits in the specific application.

REFERENCES [1] Säynäjäkangas, J. and Taulavuori, T. 2004. A review in design and manufacturing of stainless steel sandwich panels. Stainless Steel World, October, 2004. pp. 55 – 59 [2] Lohtander, M. & Varis J. P. A novel manufacturing process for producing cell structures using a modern turrett punch press. 17th ICPR (International Conference of Production Research), 3 to 7 August, 2003. Blacksburg (VA), USA. 9 p.

[3] Larkiola, J., Martikainen, H., Pellikka, E. Simulations of the forming and loading conditions of calotte panel structures. Espoo, 2003. VTT Technical Research Centre of Finland, report BTUO35-031181. 13 p. + 2 p. app.

[4] Patent application WO/2009/034226. 2009. Panel structure. Outokumpu Oyj. Priority 11.09.2007, publ.

19.3.2009. 23 p.

[5] Gales, A., Sirén, M., Säynäjäkangas, J., Akdut, N., van Hoecke, D. and Sánchez, R. 2007. Development of lightweight trains and metro cars by using ultra-highstrength stainless steels. European Commission. Final report EUR 22837. 266 p.

[6] Katajisto, H., Valente, A. and Mönicke, A. Designoptimisation of the innovative, high-performance metal sandwich solution. 9th International Conference on Sandwich Structures ICSS 9. California Institute of Technology, Pasadena (CA), USA, 14 to 16 June, 2010.

[7] Elmer Models Manual, CSC - Scientific Computing Ltd., 2007, Elmer web site www.csc.fi/elmer

HARRI KATAJISTO, ANDRÉ MÖNICKE Componeering Inc., Itämerenkatu 8, FI-00180 Helsinki, Finland ANTONIO VALENTE PLY Engenharia, Lda, Largo dos Fornos 1, PT-2770-067 Paço de Arcos, Portugal

Newsletter EnginSoft Year 8 n°1 -

41

Componenti forgiati di qualità necessitano di un approccio CAE integrato – esperienze di simulazione di processo nel campo Energia e Nucleare

Acciaieria - processo di colata Il processo di colata di lingotti è caratterizzato da una serie di parametri molto complessi, che determinano la qualità del prodotto finale. La composizione della lega è uno degli aspetti più importanti, ma lo sono anche le modalità di colata (velocità di colata, utilizzo di polveri isolanti, …) ed i materiali utilizzati per la lingottiera. Tutti questi aspetti sono tenuti in conto dai software di simulazione, che possono dare delle utili informazioni su quello che succede nella fase di riempimento dello stampo e di solidificazione del metallo.

Software specifici in questo ambito (MAGMAsoft di Magma GmbH ad esempio) riescono a valutare, oltre ai ritiri in soli-

Fig. 1 – MAGMASOFT – simulazione del riempimento di un lingotto.

Fig. 2 – MAGMASOFT – simulazione della solidificazione di un lingotto.

Fig. 3 – MAGMASOFT – distribuzione di temperature in una lingottiera.

Fig. 4 – MAGMASOFT – Difetti: macro ritiri, micro ritiri, moti convettivi, segregazioni.

SIMULAZIONE DI PROCESSO NEL CAMPO ENERGIA E NUCLEARE

I processi di forgiatura a stampi aperti (open-die) e di laminazione circolare (ring-rolling) sono in grado di produrre pezzi di grosse dimensioni in acciaio, il più possibile privi di porosità e con la massima omogeneità nelle caratteristiche meccaniche. Tali specifiche sono richieste nei settori meccanico-siderurgico (alberi pignone, pignoni, ruote pignone, alberi eccentrici, terminali e manicotti, …), petrol-chimico (corpi valvola, tubi, B.O.P., …), navale (alberi intermedi, alberi pinna ed alberi timone, …) e dell’energia (alberi turbina, alberi ventilatore, alberi per eolico, generatori quadripolari, …). In particolare in questo ultimo ambito, i componenti necessari per la costruzione di centrali nucleari devono soddisfare delle normative molto restrittive in grado di garantire le prestazioni del manufatto dopo molti anni di utilizzo in condizioni severe di contatto con agenti aggressivi e sottoposti ad irraggiamento. Si cerca quindi di produrre particolari con meno discontinuità (porosità, inclusioni, segregazioni, …) possibile e che possano essere assemblati limitando al minimo le saldature. Lo sviluppo integrato (la design chain) parte della colata dell’acciaio in acciaieria in una forma opportuna (lingotti, barre, blumi, altre forme), per proseguire con il trasferimento in forgia, dove il lingotto viene controllato, trattato termicamente e quindi lavorato alla pressa e/o al laminatoio. Il trattamento termico e le lavorazioni meccaniche di sgrossatura e di finitura sono le fasi conclusive per l’ottenimento del pezzo finito. Per ognuna di queste fasi esistono degli strumenti di simulazione dedicati e verticalizzati che consentono di simulare accuratamente le condizioni al contorno della fase in esame allo scopo di prevedere e ottimizzare la qualità del componente, grazie ad una migliore comprensione dell’influenza dei parametri di processo. L’approccio viene applicato e descritto successivamente nel caso di un lingotto in acciaio.

42 - Newsletter EnginSoft Year 8 n°1 una uniformità di temperatura tra cuore e superficie.

La simulazione del riscaldo in forno (fig. 5) può essere effettuata con software come Forge, dove può essere specificata una curva dell’atmosfera del forno e le temperature del lingotto possono essere monitorate attraverso dei sensori. L’utilizzo dell’ottimizzatore integrato in Forge può essere utile per calibrare i tempi di permanenza in forno: grazie a questo strumento, nel caso specifico di un lingotto da Fig. 5 – Forge - riscaldo in forno lingotto da 40t – temperature a cuore ed a superficie.

40t, si è compreso come il tempo di permanenza alla rima temperatura deva essere aumentato da 16 a 24h, mentre alla massima temperatura si possono risparmiadificazione, anche la formazione di porosità, la segregazione re ben 12h. dei vari elementi, la presenza di cricche a caldo, … (Figure 1, 2, 3, 4).

Forgiatura – processi di deformazione open-die Al termine della prima fase del processo produttivo, ci si poIl processo di forgiatura è caratterizzato dalla presenza di ne come obiettivo quello di trasferire le proprietà microstrutdue elementi fondamentali: il manipolatore, che tiene il pezturali e gli eventuali difetti alla fase successiva come avviezo in posizione e ne guida gli spostamenti e le rotazioni, la ne nella sequenza reale. L’ormai pluriennale esperienza matupressa, che deforma il pezzo con diversi colpi e passate. La rata in EnginSoft sia nel CAE che nel processo manifatturiero forma finale viene ottenuta infatti attraverso una serie di deha permesso lo sviluppo di algoritmi specifici che agevolano formazioni localizzate sotto le mazze, con tempi di diversi il dialogo fra strumenti software commerciali di diversa natuminuti che comportano un raffreddamento dell’acciaio e la ra e storia. necessità di prevedere, soprattutto per pezzi di grosse dimensioni, delle ricalde in forno anche di diverse ore, per riForgiatura – processo di riscaldo del lingotto portare il pezzo alla temperatura di lavorazione. Le difficoltà Il lingotto prodotto in acciaieria viene trasportato alla forgia, principali nella simulazione di questo processo sono la necesdove viene inizialmente riscaldato in forno. Questo processo sità di automatizzare la sequenza degli afferraggi da parte non è così semplice come sembra, in quanto bisogna adottadei manipolatori, la sequenza dei colpi e delle passate, con re degli accorgimenti in modo da garantire un riscaldo unile corrette rotazioni e traslazioni del pezzo e/o delle mazze. forme cuore-superficie, guidando nel contempo le trasformaFondamentale è la corretta definizione del materiale in termini di curve di deformazione a caldo e dalla corretta definizione delle caratteristiche della pressa idraulica utilizzata. Il software Forge è stato sviluppato in questi termini grazie all’apporto dei molti utilizzatori francesi che producono nel Fig. 6 – Forge – possibili movimenti impostabili per ogni colpo/passata ed esempio blumatura. campo dell’energia: la sequenza di colpi e passate può essere definita con precisione zioni che avvengono nell’acciaio aumentando la temperatura. (fig. 6), indicando quando entrano in funzione i manipolatoLe dimensioni dei lingotti sono considerevoli, quindi bisogna ri. Per ciascuno di essi si può specificare la zona di presa o e modulare il forno in modo tale che l’acciaio venga portato l’eventuale rigidezza degli afferraggi, che possono arretrare gradualmente ad una temperatura attorno ai 700-800°C, che alla spinta del materiale. La geometria delle mazze e del lindeve essere mantenuta per qualche tempo in modo da rendegotto di partenza vengono infine importate da CAD (formati re omogeneo il cambio di fase e non creare eccessivi gradien.stl o .step). ti termici cuore - superficie, dopodiché si può procedere fino Durante la simulazione si può valutare come viene indotta la alla temperatura di forgia di 1150-1200°C, fino ad ottenere deformazione del materiale: quando la pressa esaurisce

Fig. 7 – Forge – ricalcatura, blumatura, stondatura spigoli, ricalcatura di testa e martellatura.

Newsletter EnginSoft Year 8 n°1 -

43

la posizione nel lingotto di partenza, come è mostrato in fig.

9. L’utilità di questo metodo consiste nella identificazione degli istanti in cui i difetti si generano e le relative cause per agevolare il tecnico nella ricerca della soluzione appropriata e efficace.

l’energia, la mazza si arresta e si passa al colpo successivo.

Se il materiale è troppo freddo non si raggiungono le altezze di ricalcatura/blumatura desiderate, perciò dall’analisi delle curve di discesa si può capire quando sia necessario sospendere la lavorazione per riportare in temperatura il pezzo.

Possono essere simulate pressoché tutte le lavorazioni che vengono effettuate in forgia: ricalcatura con mazza piana o con bocca, blumatura a stampi piani, curvi o sagomati, stondatura degli spigoli, ricalcature di testa (Fig. 7). Partendo da barre o blumi è possibile utilizzare solo due mazze verticali o macchine più complesse con 4 mazze, per il processo di martellatura rotante, o valutare in virtuale macchinari definiti solo sulla carta.

Fig. 9 – Forge – valutazione a ritroso posizione difetti attraverso l’uso di sensori.

La simulazione viene utilizzata anche per valutare se la deformazione è in grado di compattare a sufficienza il materiale, partendo eventualmente da una distribuzione di porosità proveniente dalla simulazione di colata del lingotto, come è mostrato in Fig. 8.

Gli eventuali difetti nel forgiato sono ovviamente oggetti di indagine e lo studio può essere effettuato anche a ritroso, ovvero definendo la posizione del difetto rilevato nel pezzo finito e seguendone il movimento “a marcia indietro” fino al-

Fig. 10 – processo di ottenimento di anelli a sezione rettangolare da laminare

In questo caso si effettuano delle analisi 2D molto rapide, simulando le fasi di ricalcatura, sagomatura, punzonatura e foratura, come mostrato nella fig. 11 per un anello sagomato.

La funzione di “chaining” consente impostare, calcolare ed analizzare in sequenza tutte le operazioni, trasferendo i risultati da una operazione alla successiva, fino all’ultima azione di tranciatura, nella quale si abilita la funzione di danneggiamento per seguire la separazione del fondo per effetto del punzone di tranciatura.

Volendo rimanere nel campo della produzione di componenti per il settore nucleare, un altro processo comunemente utilizzato per l’allargamento degli anelli è la bigornatura: l’anello ottenuto come sopra illustrato, di sezione rettangolare, viene caricato su un mandrino ed una mazza che ne provoca la deformazione localizzata. La rotazione del mandrino tra un colpo ed il successivo consente di ottenere un allargamento graduale dell’anello, preservandone la lunghezza (Figura 12).

Forgiatura – processi di laminazione circolare L’anello prodotto con le fasi sopra descritte può quindi essere laminato, per ottenere le dimensioni (allargamento) e/o la forma desiderata (sagomatura del profilo). I laminatoi, che possono essere di diverso tipo, sono costituiti generalmente da un rullo principale che induce la rotazione, da un mandrino, che spinge il materiale verso il rullo e da eventuali coni, che guidano l’altezza del profilo.

Tutti questi oggetti possono essere piani o sagomati. La simulazione di questo specifico processo risulta molto complessa per la limitata area di contatto tra mandrino, anello e rullo, nella quale si concentra la massima deformazione e per la cinematica guidata dall’allargamento dell’anello. L’elevato

SIMULAZIONE DI PROCESSO NEL CAMPO ENERGIA E NUCLEARE

Fig. 8 – Forge - chiusura delle porosità di colata con il processo di forgiatura.

Risulta quindi ovvia la possibilità di calibrazione virtuale delle fasi di forgiatura necessarie per l’ottenimento di anelli a sezione rettangolare o sagomati, che poi possono essere laminati (fig. 10).

44 - Newsletter EnginSoft Year 8 n°1

Fig. 11 – Forge – 2D ricalcatura, punzonatura e tranciatura fondo per un anello sagomato.

Fig. 12 – Forge – bigornatura anello.

i tipici difetti di forma di questo processo, presenti soprattutto per pezzi sagomati: fish-tailing, mancanze di profilo, rigetti di materiale.

Forgiatura – trattamento termico Per tutte le tipologie di pezzi sopra tratFig. 13 – Forge – simulazione processo di laminazione circolare anello rettangolare e sagomato. tate, dopo la fase di deformazione è sempre presente il trattamento termico per migliorare le caratteristiche meccaniche dell’acciaio. Il processo, che può essere anche molto articolato, si compone di un riscaldamento in forno alla temperatura di austenitizzazione ed una Fig. 14 – Forge – simulazione trattamento termico: temperatura, % martensite e durezza HV. successiva immersione in un bagno di tempra, che induce delle trasformazioni microstrutturali nell’acciaio, modificando le fasi presenti. A seconda della drasticità dello scambio termico, si vanno a formare ferrite, perlite, bainite e, nella zone dove massimo è il gradiente, martensite. Nello specifico, la trasformazione martensitica è una trasformazione esotermica ed induce una espansione Fig. 15 – Forge – simulazione fase di immersione dell’anello in bagno di tempra. della struttura cristallina, che può provocare una distorsione del pezzo. Anche numero di giri infine rende il numero di calcoli da effettuare per questa operazione è possibile utilizzare la simulazione, molto elevato con una mesh adattiva che si aggiorna in parcon il software Forge, per valutare, al variare del percorso di ticolare nelle zone di contatto. In tal caso le architetture tempra e della forma del pezzo la formazione delle varie fasi, hardware con calcolo parallelo multi-core o multi-processore in funzione del raffreddamento imposto alle varie zone, per (cluster) permettono di ottenere dei risultati sufficientemenconfronto con le curve TTT del materiale. Viene infatti effette accurati tempi ragionevoli. Nello specifico della definiziotuato un calcolo termico-meccanico-metallurgico accoppiato, ne delle cinematiche del mandrino e dei coni è possibile guigrazie al quale si ottengono le fasi e la conseguente durezza dare gli stampi mediante le stesse curve di laminazione che finale del pezzo (fig. 14). l’operatore imposta sul software del laminatoio, riproducendo quindi in virtuale il comportamento della macchina reale.

Recentemente il modello di calcolo è stato migliorato per tener conto di parametri di processo quali ad esempio la duraL’analisi del comportamento del materiale tra mandrino e rulta della fase di immersione nel bagno di tempra e l’effetto lo consente di modificare la curva di laminazione e risolvere sulla tempra del pezzo, come è mostrato in fig. 15.

Newsletter EnginSoft Year 8 n°1 -

45

Fig. 15 – Forge – simulazione fase di immersione dell’anello in bagno di tempra.

Una volta valutate tutte le singole fasi di produzione di un componente, dalla colata dell’acciaio, alla deformazione, alla tempra ed alle lavorazioni meccaniche, è possibile infine utilizzare i risultati ottenuti (distorsioni di forma, stress residui, proprietà meccaniche, difetti, …) per effettuare delle analisi strutturali o fluidodinamiche, nelle quali valutare le prestazioni in esercizio del componente. Nell’immagine seguente è mostrato come l’introduzione degli stress residui e delle proprietà meccaniche come condizioni iniziali per l’analisi strutturale modifichi in modo rilevante le caratteristiche meccaniche di un corpo valvola sollecitato in esercizio.

Gli strumenti adatti e le competenze adeguate determinano un binomio vincente per lo studio e la progettazione di componenti High Tech forgiati per il settore energetico e in particolare quella nucleare.

EnginSoft ha estese competenze nella simulazione di processo, derivanti da oltre 15 anni di esperienza a diretto contatto con questo tipo di problematiche del mondo industriale.

Conclusioni La presente panoramica ha evidenziato come oggi sia possibile simulare tutte le operazioni necessarie alla produzione di un particolare in acciaio di grosse dimensioni, partendo dalla colata del metallo nel lingotto, alla successiva lavorazione di forgiatura o di laminazione, al trattamento termico.

Aspetto saliente è la possibilità con questi diversi strumenti di valutare tutta la design-chain, in modo da essere in grado di comprendere le cause di un problema andando a ritroso lungo tutti i vari passaggi di produzione.

SOCIETÀ DELLE FUCINE – THISSEN KRUPP Società delle Fucine ha deciso di intraprendere la collaborazione con EnginSoft e dotarsi di strumenti di simulazione numerica del processo di fucinatura, scegliendo in particolare il software Forge nella versione parallela multiprocessore. La competenza e disponibilità dei tecnici di EnginSoft è risultata fondamentale per la rapida introduzione dei nostri parametri di processo e la taratura dei modelli numerici di fucinatura a stampi aperti che ha consentito di raggiungere simulazioni aderenti alla realtà in tempi molto rapidi. L'ing. Roberto Caldarelli, responsabile della preventivazione e della progettazione delle sequenze di produzione dichiara: “La scelta è caduta su questo programma grazie alla estrema flessibilità nella definizione delle cinematiche: tramite semplici istruzioni è possibile impostare le passate ed i singoli colpi, indicando il tempo di pausa tra un colpo ed il successivo e tutte le movimentazioni effettuate dal manipolatore per posizionare correttamente il pezzo sotto la pressa. Questi aspetti sono essenziali, assieme a risultati che abbiamo verificato essere molto precisi, per poter utilizzare Forge per prevenire possibili problemi di deformazione del pezzo sotto la pressa, adottando opportune modifiche dei cicli di stampaggio.“ “Prevediamo di utilizzare le simulazioni in modo via via sempre più sistematico per i nuovi pezzi prodotti e di estendere il suo utilizzo alle fasi di riscaldamento in forno, per prevedere tempi di riscaldo e dilatazioni e per il successivo processo di tempra, per valutare le deformazioni relative alla trasformazione martensitica, grazie alla possibilità di simulare le trasformazioni microstrutturali”. Per ultimo si cercherà un collegamento con i risultati ottenuti dalla simulazione della colata dei lingotti, in modo da tener conto delle caratteristiche proprie del lingotto di partenza e simulare tutto il processo produttivo.

SIMULAZIONE DI PROCESSO NEL CAMPO ENERGIA E NUCLEARE

FORGIATURA MAMÈ Abbiamo fondato il CRS (Centro di Ricerca e Sviluppo) con l'obiettivo di sviluppare il nostro know-how e migliorare le conoscenze sui nostri prodotti e sul processo produttivo.

L'acquisto di FORGE il software di simulazione del processo di forgiatura e trattamento termico ha lo scopo di aiutare il CRS nella creazione di know-how. Grazie a questo software è possibile quindi analizzare il processo nel dettaglio, ottimizzando le fasi di fabbricazione, la qualità dei prodotti e quindi la riduzione dei costi e dei tempi - ciclo. L'ambizione dell'azienda è quella di riuscire ad offrire ai propri clienti un concreto strumento di cooperazione nella fase di progettazione dei prodotti, analizzando e simulando le caratteristiche che più soddisfano i requisiti che il forgiato deve possedere in funzione della sua destinazione d'uso.

L'utilizzo di Forge rappresenta il punto più importante dell'attività del CRS: permetterà di acquisire una conoscenza oggettiva, di tipo scientifico-ingegneristico. Non più soltanto empirica e legata quindi solo all'esperienza personale di chi fa parte dell'azienda.

Per informazioni, rivolgersi a: ing. Marcello Gabrielli – EnginSoft info@enginsoft.it

46 - Newsletter EnginSoft Year 8 n°1

Landi Renzo: the global leader in the sector of components and LPG and CNG fuel systems Based in Cavriago (Reggio Emilia - Italy), with more than 50 years experience in the sector, Landi Renzo is distinguished by a sustained revenue growth, a listing in the STAR segment of the Italian stock exchange, and the extent of its international operations, with a presence in over 50 countries.

The Landi Renzo Company was established in 1954 when Renzo Landi and his wife Giovannina Domenichini founded Officine Meccaniche Renzo Landi, at the time the only manufacturer of mixers specifically designed for all kinds of vehicles.

Landi Renzo S.p.A. is now a global leader in the sector of components and LPG and CNG fuel systems for motor vehicles, serving more than 30% of the market of alternative automotive fuel systems and components. It is a preferred supplier by a growing number of worldwide brands like Daimler Chrysler, Fiat, Opel, PSA, Renault, Volkswagen, and more recently Toyota.

Landi Renzo S.p.A. Research and Development Centre is currently the only one in its field to use advanced technologies that allow creating and developing modern systems to convert vehicle fuel systems to LPG and CNG. Visit the website on: www.landi.it modeFRONTIER in LandiRenzo “The first project with modeFRONTIER®, a product of ESTECO srl, dates back to 2008, when we performed an optimization of the new Electronic Pressure Regulator (EPR) - says Ferdinando Ciardiello, Research & Development Modelling Manager at Landi. “Ercole Sangregorio, current EPR Project Manager, - continues Ciardello - built-up a two steps development: at first, leveraging on experimental test data available in our in-house facilities, modeFRONTIER calibrated a numerical model of the EPR. We obtained a very precise 1D model, able to predict well and quickly the system’s behavior, the steady-state and the transient in different possible configurations. Afterwards, modeFRONTIER was used as a process integrator and a multi-objective optimizer, connecting different software tools to build a truly and multi-disciplinary virtual bench, with mechanical, pneumatic and control system models, and finding overall optimal

configurations. In this way, we were able to minimize pressure oscillations in the control volume and to get an optimal and robust EPR configuration in just few weeks”. “Moreover, we expanded the concept to 3D fluid-dynamics design, particularly with the ANSYS Workbench direct node in modeFRONTIER, resulting in scheduled 3D simulation campaigns during night time and weekends. It proved to be a very efficient approach, based on the state-of-the-art Design Of Experiment available in modeFRONTIER.” Why modeFRONTIER and EnginSoft “Computer-Aided Engineering has always been a key success factor for our growth”, says Viliam Alberini, Leader of the Components Division, “and EnginSoft has been supporting our demands well for years. Adding modeFRONTIER to our software chain in 2008 has been a winning move for more than one reason: with modeFRONTIER our approach to product concept has become more systematic and now allows us to evaluate more alternatives and take into account the effects of more design variables. This translates into value to our customers: critical factors are understood and handled much earlier in the process and the design results more robust in a shorter time. modeFRONTIER has also improved the predictive power of our numerical models by feeding them with lab testing results. This philosophy has reduced development times and costs, and our team can cater to customer demands and discuss specifications with them more efficiently”.

Newsletter EnginSoft Year 8 n°1 -

47

The CAD-CAM Cooperation in Nissan Achieved by ASFALIS Manufacturing companies are increasingly tending to centralize the management of their 3D product data. This requires moving from segmented 3D CAD data conversion and communication paths to a single integrated system.

Such a system must be able to correctly translate product data from one CAD system to another and also be able to prepare the data for other uses, such as FEA and CAM. However, these tasks are not always done correctly or adequately due to the significant difficulties in handling mismatches between various software systems. Though this is a difficult undertaking, Nissan Motor Company – the well-known Japanese automobile manufacturer – has successfully built a reliable system for conversion and distribution of their 3D product data to be used companywide for all CAD-CAM operations.

At Nissan, the control and delivery systems had used many 3D tools in production technology, and the level of data quality had differed from tool to tool. This difference in quality frequently disrupted the accuracy of data translation. To solve this problem, Nissan launched a new project to shift to a totally new translation workflow, while changing its standard CAD system from Ideas to NX. On this project, Nissan chose ‘ASFALIS’, one of the products of Elysium - a Japanese provider of 3D interoperability solutions - to consolidate the company-wide data conversion system in order to achieve high accuracy and great stability of performance. ASFALIS helps users establish a large-scale and flexible system to automatically operate CAD-to-CAD conversion or other optimization. It has been introduced among many Japanese major automobile manufacturers for their CAD conversion.

“We are replacing all the 3D translators in production engineering with Elysium’s ASFALIS, which controls all the 3D data translation and distribution processes in Nissan,” said Katsuro Fujitani, senior manager of the manufacturing and SCM system department in global IS division (as of 2010). He continued, “With its preeminent translation performance, more than 99.9 percent of the

data can be converted without error.” ASFALIS adapters are ready for all the possible translation patterns, which allow the Nissan staff to utilize any type of CAD data. In fact, they translate 3D data between INCAM* and several CAD systems such as I-deas, NX and DELMIA. Even large amounts of data are automatically converted and delivered to predetermined destinations. It is also able to control concurrently running file translation processes. Because the ASFALIS-based system is integrated to the intranet and connected with ‘Teamcenter’, the master PDM, Nissan staff in domestic branches are able to access ASFALIS to execute translations. The results of translations are automatically delivered in a specified format to another branch.

Even though different paths are needed between approved data and data under consideration, users merely have to change settings. Once configured, ASFALIS automatically

translates data and distributes results, whose quality is admirable and stable. Elysium’s reliable 3D data translation and distribution system has improved the efficiency at every step of processes throughout the product lifecycle management (PLM) in Nissan. * INCAM is the in-house CAM system in Nissan.

For more information, please visit the ELYSIUM website: http://www.elysium-global.com

For information on Elysium products in Italy, please contact: Giorgio Buccilli at EnginSoft, info@enginsoft.it

48 - Newsletter EnginSoft Year 8 n°1

CAE-based tablet design It actually is an extremely important requirement today to develop tablets that are easy to swallow, both from a compliance and a usability point of view. Until recently, no quantitative research on the correlation between shape of tablet and ease of swallowing has been made. It was Mr. Hideaki Sato of ASAHI BREWERIES, LTD. who began to investigate the shape of tablets for ease of swallowing, using sensitivity engineering and optimization methods to develop the most suitable shape.

Mr. Hideaki Sato evaluated the “tablet shape/hardness” and the “pressure resistance” of the tablet machine pestle using the FEM simulation software ANSYS.

Supported by Economic growth has provided us with many rewards including a wealthy and comfortable society. At the same time though, we are facing problems that occur with an aging population and an increase in lifestyle diseases. Even in Japan, the country with the highest life expectancy, health problems linked to lifestyle and age are evident. While in Europe, Japanese cuisine has become quite popular in the last decade, also for supporting a healthy diet and for curing disorders linked to an excessive lifestyle, in Japan a trend of eating more “fast” and less Japanese food can be witnessed. The consequences are increases in bad dietary habits and lifestyle diseases. Both have led to more frequent and longer sicknesses and to shortened life expectancy. In a modern society like this, pharmaceuticals are more and more in demand to offer supplements that can effectively treat our increasing health disorders and thus improve the quality of our lives.

ASAHI BREWERIES GROUP is one of the largest food manufacturers in Japan. The Group mainly produces alcoholic beverages, but also focuses on the research and development of various supplements for the food production part of their business. Some of these supplements are designed for compensating the lack of a healthy nutrition, such as vitamins and minerals that tend to be insufficient in our modern diet.

Some of these supplements include beer yeast for which the ASAHI BREWERIES GROUP has become world-famous. The Group’s supplements are of the highest quality and hence are products that we can trust.

Pharmaceuticals and supplements are available in many forms, such as tablets, granulates, hard and soft capsules, jellies or syrups. Tablets are the most common today, and they come in many different shapes, colors and flavors. People sometimes find it difficult to swallow tablets because they must be taken without chewing and are usually washed down with hot or cold water (except for chewable tablets).

Study of the tablet’s shape for ease of swallowing To review the correlation between tablet diameter, radius of curvature, thickness and ease of swallowing of the most typical circular tablets, the following steps were carried out by using a sensory evaluation technology based on the experience of food development and response surface methods.

Step 1: Preparing tablets 36 different shapes of tablets made from microcrystalline cellulose and calcium stearate were prepared with all the possible combinations of tablet diameter (6mm, 7mm, 8mm and 9mm), radius of curvature (6mm, 9mm and 12mm) and thickness (3 values from 2.5mm to 6.5mm).

Fig. 1 - The shape of the circular tablet

Step 2: Sensory evaluation The participants were 10 healthy men and women who took tablets from Step 1 with a glass of water every 30 minutes in random order. With the sensory evaluation, the ease of swallowing was rated on a 5 level score.

Fig. 2 - The response surface of tablet diameter, radius of curvature and ease of swallowing

Newsletter EnginSoft Year 8 n°1 -

Step 3: Analysis of sensory evaluation result Based on the results of Step 2, the response surface of the tablet diameter, radius of curvature and ease of swallowing was established by spline interpolation connecting each data point.

This was done for actual ease of swallowing and for apparent ease of swallowing.

The result shows that it is easier to swallow when both the diameter and the radius of curvature are smaller with the smallest diameter and radius of curvature being the easiest when the thickness is 3.5mm. However, the best score for swallowing in cases where thickness exceeds 3.5mm is when the diameter is 7mm, not the lowest value 6mm. This way, it became clear that the smaller diameter is not always better for swallowing.

Moreover, it is important to consider the most appropriate diameter based on the radius of curvature and the thickness.

Regarding the apparent ease of swallowing, the result was different from the result of the actual ease of swallowing. The participants felt that smaller diameters and smaller radiuses of curvature are generally better. Additionally, the best shape for swallowing based on each volume was specialized. This research revealed the relation between tablet shape and ease of swallowing.

Considering the fact that it is indeed difficult to change pharmaceutical diameters in Japan, the conclusion was that the most appropriate solution would be to reduce the radius of curvature (i.e. to make the shape round).

However reducing the radius of curvature size entails the problem of decreased durability of the tablet and the die (a part of the tablet machine pestle). To overcome this, ANSYS was used for the stress simulation of the arbitrarily-shaped tablet and tablet machine pestle.

Stress simulation for the arbitrarily-shaped tablet Typically, in the pharmaceutical and food industries, the tablet strength is evaluated by a stress test to examine the fracture load under the unidirectional load of the tablet. This is called “Tablet Hardness”. To predict this as accurately as possible, CAE simulations were performed. We should mention here that this was the first time that a CAE approach has been applied in these industries. In fact, tablet strength evaluations are very challenging, as tablets are made of compressed formations of powdered substances and unlike mechanical structures, the shape and Young’s modulus are different and dependent on the pressure loads.

Step1: Simulation under the assumption of a constant Young moduleus As a first step, the reaction force of the tablet, a 1/8 symmetrical segment model, was calculated under enforced displacement. In this simulation, it was assumed that the tablet’s Young Modules was constant, and an adequate material property was defined for the model. It was done this way because 3 different shapes of tablets made by the same pressure load had almost the same volume (density), and the prediction was that the Young Modules would be constant. However, there was a divergence between the simulation results and the test

49

results under the same conditions. So, it became obvious that applicable accuracy cannot be expected under the assumption that the Young Modules is constant.

Fig. 3 - 1/8 model of the tablet and analysis condition

Step2: Simulation with the Young Module defined by the reaction force on the pestle This new approach was applied to gain realistic values for the tablet’s Young’s Modulus. The stress simulation of the tablet machine pestle was performed to obtain the reaction force on the pestle head when tableting, and then to verify the distribution of the reaction force as the distribution of the tablet’s Young Modulus by transcribing it to the tablet model.

Fig. 4 - The tablet machine pestle and the area of the simulation

Fig. 5 - 2D symmetrical model for the simulation

In order to determine the reaction force on the contact area, a stress simulation using the 2D symmetrical model shown in Fig.5, was performed. The model was made from chrome-nickel and a load of 20kN was applied to the upper region. The distribution of the reaction force was obtained by the simulation illustrated in Fig.6.

The reaction force was then transcribed to the tablet model for its own stress simulation. At the same time, the average reaction force for each divided region (2 parts or 4 parts) was transcribed

50 - Newsletter EnginSoft Year 8 n°1 Fig. 9 shows the 3 different contour plots of the stress simulation for the tablet with a diameter of 8mm and a radius of curvature of 15mm. These are the results when the Young Modulus is assumed to be constant, divided into 2 parts and divided into 4 parts, respectively from left to right.

The new approach to transcribe the reaction force on another model of the analysis model was applicable in this case. However further considerations regarding the applicable range (e.g. powder property, tablet machine type and tablet machining conditions) will become necessary in the future.

Fig. 6 - The reaction force distribution on the pestle head

Fig. 7 - Young Module transcription to the tablet model

Fig. 8 - Relation between tablet hardness and reaction force

to the similarly divided region of the tablet model (the transcription model of SATO-MIURA).

The result of the ANSYS simulation was consistent with the experimental result, and it led to suitable results with practical accuracy.

Stress simulation for the arbitrarily-shaped tablet machine pestle The tablet machine is in operation all day to compress powder instantaneously with hundreds or thousands of kgf of pressure.

A lot of stress occurs on the machine and sometimes this causes breakage. When reducing the tablet curvature size, the pestle head will be sharpened and the load capacity of the pestle will drop to a lower level. In the past, the load capacity of the pestle used to be based on the tablet machine manufacturerâ&#x20AC;&#x2122;s experimental rules. To predict the load capacity more accurately, an ANSYS simulation of the tablet machine was performed. A 2D axisymmetric model was prepared and the contact element between the pestle and the tablet was defined to represent the pestle sliding slightly on the tablet during the powder

Fig. 10 - Analysis condition

Young Module of the pestle: 166,100Mpa Poisson ratio of the pestle: 0.3 Acceptable stress value of metallic material of the pestle: 2,172Mpa Tablet diameter: 6.0mm and 8.0mm Land: 0.1mm Contact stiffness coefficient of the pestle and the tablet: 5.0

compression phase. The surface of the tablet was defined as rigid. The analysis condition is shown in Fig. 10. In this simulation, the calculation was repeated until the stress value inside the pestle reached the allowable stress value of 2,172Mpa in order to know the load capacity.

Fig. 11 shows the area which might break. This corresponds exactly with the tablet machine manufacturerâ&#x20AC;&#x2122;s experimental rule. Hence we can conclude that the CAE simulation for the tablet machine is valid.

Fig. 9 - the maximum principle stress of the tablet with 8mm tablet diameter and 15mm radius of curvature

The new approach of using CAE received a great response from industry Today, CAE is the standard tool of machine design manufacturers and many examples,

Newsletter EnginSoft Year 8 n°1 -

Fig. 11 - Contour plot of the equivalent stress

Comments from Mr. Sato of ASAHI BREWERIES, LTD. (Share the Kando.*) In today’s food industries, we only find a few examples for CAE- (and even fewer for FEM-) based product development. For the work described in this article, we applied ANSYS to evaluate our tablet design, and this attempt provided us with a lot of new and useful information. The response surface for the ease of swallowing has high prediction accuracy, this is why it is now used for product development in the ASAHI BREWERIES GROUP. Regarding the transcription model, the idea to consider the reaction force on the pestle as the tablet’s Young Module was a complete breakthrough.

For the use of the approach in the future, we would like to make a decision on the applicable range of the theoretical model based on the powdered material and the working conditions of the tablet machine.

We are no CAE specialists, so to us simulation is just a tool and not our main objective. It is necessary to cooperate with the CAE vendors for those simulation cases which cannot be solved by ourselves. In such situations, it is very important to deliver our analysis results and requirements to the CAE vendors as sufficient engineering knowledge and analytical thinking are indispensable. We expect from the CAE vendors that they don’t stick to their own technologies and simulation results, and that they provide flexible services. There are cases where – after sufficient communications and exchange of information - we find out that no use of CAE simulation is necessary.

I am very pleased that Mr. Miura of Cybernet Systems has always responded quickly to my requests, and that he is a reliable partner. I do believe that the best solutions come from human communication, not from automatic computational calculation. This is the spirit of “Share the KANDO.” *This is the corporate message of ASAHI BREWERIES, LTD. It means: Always creating new value moves people’s hearts and forms a strong bond. Always imagining a fresh tomorrow moves people’s hearts and helps them shine.

Sharing these emotional experiences with as many people as possible—this is the mission of the Asahi Breweries Group.

51

reports and testimonials come from these industries. The specialists in the design and development divisions are becoming more and more familiar with CAE. The various technologies are not too difficult to apply, even for beginners, as they usually have previous experience, and the manuals and guidelines are pretty clear. Still, CAE with stress simulation has not been used in the pharmaceutical and food industries worldwide until recently. Mr. Sato indeed has made a big step forward with his idea of using CAE for tablet design. The key success factor was the new approach of using the substituted condition for the target simulation in cases where the real physical condition cannot be determined. In 2010, these series of simulations were presented at different academic meetings and in various publications. Mr. Sato’s work and approach received a great response from the pharmaceutical and food industries and the CAE sector.

This article is based on the original case study by Mr. Hideaki Sato, Research Laboratories For Food Technology, ASAHI BREWERIES, LTD. and Mr. Takahiro Miura, Mechanical CAE Division, CYBERNET SYSTEMS CO.,LTD. Akiko Kondoh Consultant for EnginSoft in Japan Comments from Mr. Miura of CYBERNET SYSTEMS Co.,Ltd. CYBERNET SYSTEMS is a Japanese company offering computational engineering solutions, such as CAE software tools and all product-supporting services including seminars, support and consulting. ANSYS is one of our major business areas. We have a strong customer base in different industries, for example in automotive, electrical machinery, electrical devices, energy, aerospace and medical engineering. For the past 25 years since its establishment, CYBERNET SYSTEMS has been passionate about supporting MONOZUKURI in Japan as a CAE solution provider. Now, it also has some affiliated companies in Asia, North America and Europe, and has become a global player.

ASAHI BREWERIES GROUP is a pioneering Japanese food and beverage manufacturer. We are truly honored to be able to collaborate with Mr. Sato who is promoting cutting-edge research and development. Currently, CAE is not used extensively in the food and beverage industries in comparison with other industries. We believe that CAEbased engineering simulation will provide effective solutions for achieving “time savings”, “cost reduction”, “security assurance” and “environmental protection”, important topics also for companies in these industries.

We will endeavor to develop other examples for the engineers in the food and beverage industries, to encourage them to connect with and use CAE. We will also deepen the relations with our partners, like with Mr. Sato, and grow their passion for MONOZUKURI with our own passion.

52 - Newsletter EnginSoft Year 8 n°1

Tokyo a Metropolis

March 2011 Earthquake and Tsunami in Japan

Tokyo, the capital of Japan, is the world’s biggest mega city according to the United Nations’ 2010 report, with a population of 13 million and 36.6 million if we include its surrounding urban areas. Another report by PricewaterhouseCoopers (PwC) states that Tokyo has the highest GDP of any cities in the world.

Tokyo is also the heart of Japan with regards to politics, culture and education. When we think of Tokyo, images of typical cityscapes with high-rise buildings standing above busy crowds, elaborate train and subway systems, different varieties of academic and cultural facilities along with a rich entertainment heritage are conjured up. On the other hand, the city has many different aspects, such as numerous parks and green areas, waterways and finally the sea! There are many places where one can relax and unwind watching the changes of the seasons. From the many faces of Tokyo, I would like to introduce my favorite spots in this article.

This article was written before a terrible earthquake hit parts of Japan and its people.

If you want to help, please donate to:

Italian Red Cross: http://www.cri.it

The Japanese Red Cross Society: http://www.jrc.or.jp/english/

British Red Cross: http://www.redcross.org.uk/

German Red Cross: http://www.drk.de or to any other organization that helps Japan in the present crisis.

Thank you The Newsletter Editorial Team

The skyscrapers When visiting Tokyo, many of my European friends are surprised by the cluster of high-rise buildings in different areas and the endless expansion of crowded residential areas spread across the suburbs. In fact, many new buildings are constructed with incredible rapidity every year, and the landscape changes

Shinjyuku-Gyoen Park

The nightscape of Shinjyuku

constantly. The high-rise buildings of Tokyo not only overwhelm people at daytime, they also present amazing views after sunset.

Tokyo is known all over the world for its diversity of restaurants and gourmet places, in particular: Japanese, Italian and French cuisine are the people’s favorite. On the top floors of some tall buildings such as “Tokyo Midtown” in Roppongi and the “Marunouchi Building” near Tokyo station, we can enjoy the great bird’s eye view with a variety of gourmet food. This is truly a unique experience. After a busy day, a lot of people in Tokyo feel at home watching the illuminations and the slowly blinking lights floating into the night sky.

The green oases Surprisingly, there are many large parks with a lot of trees in Tokyo. In Shinjyuku, located nearby the Tokyo Metropolitan

Government offices, there is a park called Shinjyuku-Gyoen that I often visit with my family when the weather is fine. ShinjyukuGyoen is run by the Ministry of the Environment and covers an area of 580,000 m². This beautiful park invites visitors to enjoy gardens of three distinct styles: the French Formal Garden, the English Landscape Garden and the Traditional Japanese Garden.

In spring, the park’s 1300 cherry trees attract many visitors as one of the best cherry blossom-viewing spots in Japan. A short distance from Shinjyuku, there is another large green oasis called Meiji-Jingu, it is the home of a famous Shinto shrine and covers 700,000 m². Meiji-Jingu is surrounded by a very old manmade forest. As soon as you enter the area, you will feel a sublime atmosphere, far from the hustle and bustle of the city.

Once you have passed the wooden approach, you will reach the main hall enshrining a God, a treasure museum called Homotsuden and Shiseikan of martial arts. At week-ends, you might even be able to see a traditional Japanese wedding ceremony.

Asakusa – the old town of Tokyo Dwarfed by the modern buildings, Tokyo’s old towns welcome visitors with warmth and old world charm. The most famous town is Asakusa which is very popular among foreign travelers.

Newsletter EnginSoft Year 8 n°1 -

Senso-ji Temple

The Senso-ji temple can be found here, it is world-renowned for its Kaminari-mon. The existing main hall and five-story pagoda were reconstructed after having been burned down during the Second World War. The original buildings date back to more than a thousand years ago, and they are historical and symbolic temples of Tokyo. Along the approach from the Kaminari-mon to the Hozo-mon, there are many souvenir and snack shops lining up a street called Nakamise. Here we can enjoy shopping in a very Japanese atmosphere. The neighborhood is dotted with lots of classic restaurants, which will satisfy your appetite with very authentic Tokyoite dishes, such as Tenpura, Sukiyaki and Unagi.

On the west side, there is Kappabashi, a street devoted to kitchenware, which supplies most of the restaurants in Tokyo.

Recently, so-called ultra-realistic food models are sold here which have become very popular as souvenirs.

Returning to the east side of the Senso-ji temple, there is the Sumida river, where you can enjoy a nice walk and maybe a short trip on a cruise boat. Behind the red painted Azuma-bashi bridge, the headquarters of ASAHI BREWERIES, LTD. are located, the Japanese Group that we introduce in this Newsletter with the CAE tablet design case study. It is a major landmark because of the golden flame on top of the black building. On the left of the ASAHI BREWERIES buildings, the world’s tallest TV tower “Tokyo Sky Tree” appears. It is still under construction. In February 2011, its height has reached 574m. The construction will be completed at the end of 2011, the final height will be 634m. The number 634 was selected because of its pronunciation in Japanese which is MUSASHI, the old name of the Tokyo metropolitan area.

53

ODAIBA the bay area Odaiba is the bay area which did not exist in the old maps because it was constructed by massive landfills towards the end of the last century, and the current landscape only appeared after the 1990’s. Towards the end of the Edo Period (16031868), a number of forts were built here on the different islands in the bay, to protect Tokyo against possible attacks from the sea. More than a century later, Tokyo began a spectacular development project aimed to relieve the congestion in the city center, and it became a large business and residential district starting with the opening of the Rainbow Bridge. The beautiful scenery of neo-futuristic streets and the bay area, attracts many visitors from all parts of the country to Odaiba.

Odaiba is, at the same time, a favorite place for engineers. Many trade shows dedicated to the manufacturing and CAD/CAE industries are held at this “Tokyo Big Sight” which was opened in 1996. The hotels in this area are often chosen as venues for different Users’ Meetings of CAD/CAE software products.

Rainbow Bridge in Odaiba

Savor the Cuisine of Tokyo One of the major attractions of Tokyo is its cuisine and unique gastronomic variety. Here, locals and visitors from all parts of Japan and around the world, savor different kinds of Japanese food and the cuisines from many other cultures. In Tokyo, we can enjoy food of the highest qualities and standards at reasonable prices.

If you visit Tokyo and wish to look for a suitable restaurant, there are several handy search guides you can use, for example GourNavi. You can select from a wealth of information based on location, food type and price range.

If you search e.g., for a restaurant near Roppongi, about 2,000 restaurants will come up very quickly (100 on the English website). In case you are tired at some stage, from all the going out to restaurants, why not discover DepaChika, the department store's basement food floor in the station. Here you can buy your favorite dishes from a variety of choices and take them to a nearby park or wherever you are staying – this is very easy and convenient and what the locals do also!

Tokyo is one of the biggest and most modern metropolis in the world. At the same time, it is also a very interesting city that has always maintained its originality, a melting-pot of nature, people’s love for nature and traditional cultures…

There is so much more about Tokyo that we will bring to you in the next Japan Columns of the Newsletter!

The headquarters of ASAHI BREWERIES, LTD., and Tokyo Sky Tree

Akiko Kondoh, Consultant for EnginSoft in Japan

54 - Newsletter EnginSoft Year 8 n°1

President Obama Honors EnginSoft’s Partner with the Presidential Early Career Award for Scientists and Engineers Prof. Gianluca Iaccarino has been recognized last December with the Presidential Early Career Award for Scientists and Engineers (PECASE), the highest honor the U.S. government bestows on scientists and engineers in the early stages of their research careers, during an official ceremony at the White House in Washington D.C. Prof.

Iaccarino, one of 13 U.S. Department of Energy researchers named as recipients, was recognized for “his extensive and deep scientific contributions in the areas of turbulent flow and uncertainty quantifications for the National Nuclear Security Administration community,” according to a Department of Energy official. The award winners were honored for their research efforts in a variety of fields, from helping the nation achieve energy independence to exploring the realms of space to identify dark matter. These awardees are funded by the U.S. Department of Energy's Office of Science and the National Nuclear Security Administration. The winning DOE scientists are among 85 researchers supported by 10 federal departments and agencies who have received the award. In addition to a citation and a plaque, each PECASE winner will continue to receive DOE funding for up to five years to advance his or her research.

“Science and technology have long been at the core of America's economic strength and global leadership I am confident that these individuals, who have shown such tremendous promise so early in their careers, will go on to make breakthroughs and discoveries that will continue to move our nation forward in the years ahead” said President Obama.

“These gifted young scientists and engineers represent the best in our country. The awards recognize ingenuity, dedication, diligence and talent. I congratulate the PECASE awardees and wish them continued success towards new discoveries and advances in science, energy research, and national security” said Secretary Steven Chu. The Award Motivation “For his extensive and deep scientific contributions in the areas of turbulent flow and uncertainty quantifications and

quantified margins of uncertainty, which are amplified for the National Nuclear Security Administration (NNSA) community through his position of intellectual leadership at the NNSA Predictive Science Academic Alliance Program Center at Stanford” Nominated by Lawrence Livermore National Laboratory Prof. Gianluca Iaccarino Dr. Gianluca Iaccarino is an Assistant Professor at Stanford University with joint appointments in the Mechanical Engineering Department and the Institute for Computational Mathematical Engineering. He completed his graduate studies in Italy working on computational methods for fluid dynamics and worked as a Research Associate at the NASA Center for Turbulence Research before joining the Faculty at Stanford in 2007. He is the Deputy Director of the NNSA Predictive Science Academic Alliance Program (PSAAP) Center at Stanford and leads the effort on Quantification of Margins and Uncertainties, a decision-making computational framework aimed at managing risks associated to highconsequence systems. His research activities are focused on Computational Fluid Dynamics, in areas ranging from analysis of wind turbines, to hypersonic propulsion, to turbulence and transition modeling, to thermal management in batteries. In 2007 Dr. Iaccarino funded the Uncertainty Quantification Lab (http://uq.stanford.edu): a joint initiative between the School of Engineering and the Mathematics and Statistics Departments. The UQLab is supported by various grants from NNSA, DOE Office of Science, NSF, and industries and focuses on probabilistic algorithms for uncertainty analysis, stochastic inference and robust optimization. The research work ranges from the theoretical aspects of uncertainty representation, to algorithms for nondeterministic analysis, to large-scale applications leveraging massively parallel computers. Many of the current projects involve

Newsletter EnginSoft Year 8 n°1 -

active collaborations with Sandia, Lawrence Livermore and Los Alamos National Laboratories. Dr. Iaccarino is involved in various educational activities at Stanford and in the computational engineering community.

He organized Uncertainty Quantification tutorials, workshops and special sessions at major engineering conferences. He has published more than 50 papers in both engineering and mathematics journals and about 70 conference papers. He is also a Humboldt fellow at the University of Munich, Germany.

Dr. Iaccarino is also the Director of the Thermal and Fluid Sciences Affiliates and Sponsors Program (TFSA http://www.stanford.edu/group/tfsa/) which EnginSoft has recently joined and he is also one of the co-founders of Cascade Technologies Inc. (http://www.cascadetechnologies.com), EnginSoftâ&#x20AC;&#x2122;s Partner Company in Palo Alto (California) that develops, markets, and supports state of the art Computational Fluid Dynamics (CFD) analysis tools for engineering applications across industries.

About the Award The Presidential Early Career Award for Scientists and Engineers (PECASE) is the highest honor bestowed by the United States government on outstanding scientists and engineers in the early stages of their independent research careers. The White House, following recommendations from participating agencies, confers the awards annually. To be eligible for a Presidential Award, an individual must be a U.S.

55

citizen, national or permanent resident.

Winning scientists and engineers receive up to a five-year research grant.

History In February 1996, the National Science and Technology Council (NSTC), was commissioned by President Bill Clinton to create an award program that would honor and support the achievements of young professionals at the outset of their independent research careers in the fields of science and technology. The stated aim of the award is to help maintain the leadership position of the United States in science.

Originally, 60 recipients received the PECASE award per year. Due to increased participation by the Department of Defense, this has increase to 100 per year. The 2002 PECASE awards were not announced until May 2004 due to bureaucratic delays within the Bush administration.

Agencies The agencies participating in the PECASE Awards program are: Department of Agriculture, Department of Commerce, Department of Defense, Department of Energy, Department of Education, Department of Health and Human Services: National Institutes of Health, Department of Veterans Affairs, National Aeronautics and Space Administration, and the National Science Foundation.

From Wikipedia, the free encyclopedia: http://en.wikipedia.org/wiki/PECASE

56 - Newsletter EnginSoft Year 8 n°1

Formazione a distanza sugli elementi finiti EnginSoft ha messo a punto e sostiene, per conto di Consorzio TCN, l'iniziativa di formazione a distanza in ingegneria “improve.it”, con l’obiettivo di creare una risorsa di alta formazione continua, coerente con gli obiettivi di formazione del Consorzio.

Tramite il portale all’indirizzo http://www.improve.it è possibile accedere a corsi multimediali di auto-formazione sui temi della simulazione numerica e prototipazione virtuale e temi a questi complementari e affini.

Nel 2010 il portale improve.it è stato completamente rinnovato sia nella grafica che nei contenuti. Tra le nuove funzionalità offerte da improve.it sono disponibili il nuovo sistema di ricerca dei contenuti e la nuova procedura online per l’acquisto dei corsi. Il contenuto di tutti i corsi disponibili è inoltre stato migliorato e per la loro erogazione viene ora utilizzato un nuovo sistema multimediale di elevata qualità.

Istantanea di una delle lezioni multimediali che compongono il corso

zioni di forma, convergenza della soluzione, utilizzo di elementi di ordine superiore al primo.

Il catalogo dei corsi è costantemente aggiornato e riportato sul sito. Per ogni corso è disponibile il programma dettagliato. È inoltre consentito l'accesso gratuito ad alcuni corsi. Sono oggi disponibili on-line corsi a vari livelli relativi a metodo degli elementi finiti, analisi statica e dinamica, fluidodinamica computazionale, acustica computazionale, elettromagnetismo, progettazione, ottimizzazione multi-obiettivo, scienza dei materiali, processi produttivi, metallurgia…

Nuovo corso online 2011 di introduzione al FEM Il metodo degli elementi finiti: teoria e applicazioni meccanico-strutturali in campo elastico lineare Docente del corso: Prof. Leonardo Bertini, Dipartimento di Ingegneria Meccanica, Nucleare e della Produzione, Università di Pisa. Il corso, suddiviso in tre unità didattiche per un totale di 19 moduli multimediali, si propone di fornire gli strumenti teorici e applicativi per l'impiego corretto e ragionato del metodo degli elementi finiti per lo studio di strutture meccaniche in campo elastico lineare.

Prima unità didattica: le basi teoriche del metodo degli elementi finiti Vengono sviluppati otto moduli per introdurre la teoria del metodo degli elementi finiti, in modo semplificato e attraverso l'utilizzo di numerosi grafici ed esempi. Gli argomenti trattati includono: discretizzazione, campo di spostamenti, calcolo delle deformazioni, analisi agli elementi finiti, vincoli e carichi, fun-

Seconda unità didattica: applicazioni del metodo FEM alle principali classi di problemi strutturali in campo elastico lineare Nei successivi dieci moduli del corso vengono passate in rassegna le principali famiglie di elementi finiti e per ciascuna di esse vengono forniti esempi di applicazione e limiti di utilizzo. In particolare vengono affrontati i seguenti tipi di elemento: asta, trave, pipe, piani, di Fourier, gap, guscio assialsimmetrico, elementi guscio/piastra 3D, brick.

Terza unità didattica: analisi critica dei risultati di un modello FEM e criteri generali di modellazione con il metodo degli elementi finiti I due moduli conclusivi del corso affrontano i criteri generali di modellazione FEM e introducono l'analisi critica dei risultati attraverso esempi sui seguenti argomenti: singolarità dello stato di tensione, definizione e schematizzazione dei vincoli, utilizzo delle simmetrie.

Materiali aggiuntivi e questionario di autovalutazione Oltre al materiale multimediale della durata complessiva di 4 ore, il corso offre tracce di esercizi da svolgere tramite analisi agli elementi finiti, e un questionario di autovalutazione della comprensione degli argomenti trattati, composto da 20 domande a risposta multipla. Agli iscritti che affronteranno positivamente il questionario di vautazione verrà inviato l'attestato di partecipazione al corso. All'interno del corso è disponibile il forum privato per porre domande al docente. Agli iscritti al corso, oltre all'accesso ai materiali didattici, verranno inviate le dispense a colori con la riproduzione delle oltre 280 diapositive utilizzate dal Docente nelle lezioni.

Per informazioni e iscrizioni: http://www.improve.it

Newsletter EnginSoft Year 8 n°1 -

57

REACTIVE BUSINESS INTELLIGENCE From Data to Models to Insight usually dissected in books dedicated to different areas.

Brevity and attention to the essential ideas and methods were our design principles.

The new book by Roberto Battiti and Mauro Brunato is now available: ROBERTO BATTITI AND MAURO BRUNATO. Reactive Business Intelligence.

From Data to Models to Insight.

Reactive Search Srl, Italy, February 2011.

ISBN: 978-88-905795-0-9

Readers of the EnginSoft Newsletter deal with engineering simulations, models and designs and do not need many words to understand the powerful combination of models, simulators, and interactive visualizations.

We hope that this book will be useful to researchers and practitioners in widely different areas and business sectors.

Take the plunge into Reactive Business Intelligence!

Reactive Business Intelligence is much more than “pretty pictures”. It is about integrating data mining, modeling and interactive visualization, into an end-to-end discovery and continuous innovation process powered by human and automated learning.

The concept is illustrated in the figure RBI: reactive business intelligence. This holistic and unifying goal requires collecting and integrating topics which are The School of Athens, by the Renaissance artist Raphael, 1510.

Last but not least, using proper visualizations can provide us with aesthetic satisfaction and even artistic emotions, although maybe not to the same extent as Raffaello’s “The School of Athens”…

To buy this book, please visit the book’s web page:

http://reactivebusinessintelligence.com/ By inserting the coupon code ENGINSOFT-RBI, a special 20% discount will be applied (this offer ends on 31st May 2011).

58 - Newsletter EnginSoft Year 8 n°1

EnginSoft at the Optimization Day: Research and Applications

EnginSoft joins the Thermal and Fluid Sciences Affiliate Program of Stanford University and Sponsors a One Day Seminar on Optimization.

EnginSoft has recently strengthened its North American operations by means of expanding both the office’s space and the employed personnel at its Silicon Valley office, located at the Palo Alto Technology Center (San Francisco Bay Area California).

With the aim of stimulating the networking in the area as well as of providing continuous support and commitment to scientific research, EnginSoft has also joined the TFSA (Thermal and Fluid Sciences Affiliate) of Stanford University and has been welcomed as a new member during the 2011 TFSA conference which took place at the Munger Conference Center, inside the Stanford University Campus, in Palo Alto on February 2-4, 2011.

The conference is organized every year within the TFSA program and is aimed at presenting the latest research work of the Stanford Thermal and Fluid Sciences program. This year’s conference has been enriched and anticipated by a One-Day Seminar on Optimization (the “Optimization Day”) which has been held on February 1st, at Stanford Campus. The event, as part of TFSA program, has been organized by Prof. Gianluca Iaccarino and EnginSoft ’s staff in Palo Alto. EnginSoft has indeed sponsored and fully supported the event with the intent of stimulating the discussion on optimization as an effective and practical means for engineering practice, while bringing its

contribution in terms of leading technology (modeFRONTIER) and expertise in the field of MultiObjective Design Optimization, in particular with respect to Computational Fluid Dynamics (CFD) applications. The many applications of optimization that were presented

Newsletter EnginSoft Year 8 n°1 -

59

including affiliates representatives, sponsors of research, and special guests, and the technical program was made up of more than 50 oral presentations and a poster session. Several topics were covered with contributions in various areas of our research activities, including predictive science, aeroacustics and noise, fluid mechanics, LES, combustion (modeling and diagnostic) heat transfer, energy science, processing and engineering of materials, micro-scale flow and heat transfer and fuel cells, as well as issues regarding design Optimization under design uncertainties. Several new initiatives that are providing substantial growth in the research activities and new opportunities for industrial collaboration were described. During the last day of the conference, a lab tour was organized so as to illustrate the capabilities and state of the art of the Stanford labs and computing facilities. Finally, a dinner banquet at the Stanford Faculty Club was held; it was there that the best presentations and scientific papers were awarded (http://www.stanford.edu/group/tfsa/).

ranged from Rapid Product Development to Web searching, from Sophisticated Multidisciplinary Analysis to Robust Design Under Uncertainty. In each specific presentation the following questions were addressed: What are the Remaining Barriers for Optimization Algorithms? How are Present Computational Resources Changing the Paradigm of Engineering Design? Are Current Optimization Methods Sufficient to Drive Decision-Making? The objective of bringing together Stanford faculty and industrial representatives to discuss the current applications and Examples from Some Papers remaining bottlenecks to the adoption of Optimization TFSA Conference 2011 Algorithm was achieved “Large-Eddy Simulation of Active Flow Control”, by through this event with a Parviz Moin (Stanford Univ.) and Arvin Shmilovich great success of attendance.

(Boeing) Companies like Rolls Royce and GE illustrated their activities on aeroengines, while Ferrari brought its experience from its Formula 1 racing activities. Stanford faculty members illustrated their wide experience in aerodynamic design via control theory (Prof. A. Jameson) and gave talks on frontier topics such as optimal design under uncertainties (Prof. G. Iaccarino).

The following 3 days (2-4 February) the TFSA convened for its annual conference, covering advanced topics in CFD introduced by Professor Parviz Moin. The conference is the main event organized within the program every year in February. It was an exciting conference presenting the latest work of the Stanford Thermal and Fluid Sciences program. The 2011 conference had over 100 participants,

Presented at the LES of Supersonic Jets from Complex Nozzles“ by Joseph W. Nichols, Frank E. Ham, Yaser Khalighi, Sanjiva K. Lele, and Parviz Moin

60 - Newsletter EnginSoft Year 8 n°1

NAFEMS World Congress 2011 Preliminary Agenda Announced International Association for the Engineering Analysis Community Releases Line-up for Global Simulation & Analysis Conference GLASGOW, UK, FEBRUARY 18TH 2011 – NAFEMS, the International Association for the Engineering Analysis Community, has announced the preliminary agenda for its 2011 World Congress, being held in Boston, MA, USA between May 23rd and 26th. Including over 150 presentations in more than 40 sessions over three days, this represents the most comprehensive and wide-ranging collection of analysis and simulation specific papers available from one independent, international event.

Keynote speakers at the Congress will include; • Marc Halpern: Gartner, Inc., USA • Mike Hinton: QinetiQ, United Kingdom • Alexander Karl: Rolls-Royce Corporation, USA • Ronald Krüger: National Institute of Aerospace, USA • Wiley Larson: Stevens Institute of Technology, USA • Laura Michalske: Procter & Gamble, USA The full agenda is available to view, and to download, from the Congress website at www.nafems.org/congress.

Registration is also available here, as well as full details of the location, venue and exhibition opportunities.

About NAFEMS NAFEMS is a not for profit organization aimed at promoting best practices and fostering education and awareness in the engineering analysis community. In line with its objectives to promote the effective use of simulation technologies, NAFEMS is continually seeking to create awareness of new analysis methodologies, deliver education & training, and stimulate the adoption of best practices and standards by offering a platform for continuous professional development. For more information, visit www.nafems.org.

Further information and high-resolution images are available on request.

Tim Morris from NAFEMS is available for further comment by arrangement.

A number of short training courses and special workshops will also be available for delegates to attend, ensuring that their time in Boston is used to maximum effect.

As many as 6 parallel tracks will run over the three days of the Congress, covering topics including; • Optimization • Integration • Composites • Materials • CFD • Fatigue & Fracture • Geotechnics • MBS • Industrial Applications • Business Benefits • Dynamics & Testing • Education • Analysis Management • Simulation Data Management • Seismic Analysis • High Performance Computing • Business Benefits of Simulation • …and many more The NAFEMS World Congress is the only event dedicated to showcasing the state-of-the-art and state-of-practice in the simulation world in an impartial forum, open to everyone with an interest in how to get the most from their use of simulation.

Visit the Congress website at http://www.nafems.org/congress to find out more, and to register for the only independent, international conference dedicated to analysis and simulation technology.

Press Contact David Quinn (NAFEMS), +44 (0) 13 55 22 56 88 david.quinn@nafems.org

Newsletter EnginSoft Year 8 n°1 -

61

EnginSoft alla Fiera Made in Steel di Brescia Dal 23 al 25 marzo 2011 si è svolta presso il centro fiera Brixia Expo Fiera di Brescia la quarta edizione di Made in Steel, un evento biennale dedicato alla filiera dell’acciaio.

Made in Steel 2011 ha registrato un record di visite rispetto alle sue precedenti edizioni: 13.500 visitatori, provenienti da 46 nazioni, confrontati con i 12.000 della scorsa edizione nel 2009. Un aumento rilevante anche nel numero degli espositori, salito da 187 a 248, e di conseguenza nell’area espositiva (da 7.400 a 10.200 mq). Va ricordata inoltre la presenza delle delegazioni estere di Austria, Bielorussia e Cina.

EnginSoft ha partecipato a Made in Steel con uno stand, focalizzando l’attenzione sui software e le applicazioni specifiche per il settore metallurgico, in particolare quelle sostenute da: MAGMA (MAGMAsteel e MAGMAfrontier), Transvalor (FORGE) e Third Wave Systems (AdvantEdge FEM e Production).

La fiera è stata visitata prevalentemente da manager e il nostro stand ha registrato un buon afflusso di persone appartenenti ad importanti aziende con le quali sono stati organizzati degli incontri riservati dove sono stati discussi problemi specifici e ipotesi per future collaborazioni.

Il terzo giorno, venerdì 25 marzo, la nostra società ha organizzato presso la Sala Steel il seminario dal titolo: “Energia Nucleare di nuova generazione: engineering e design di processo e di prodotto”. I relatori sono stati Piero Parona, che ha illustrato la mission di EnginSoft e le proprie referenze, Marcello Gabrielli le applicazioni di FORGE dedicate ai forgiatori d’acciaio, Enrico Borsetto quelle di ADVANTEDGE per le lavorazioni meccaniche, Gianluca Quaglia quelle di MAGMA dedicate ai processi di fonderia e di Trattamento Termico e Massimo Galbiati quelle FEM di tipo fluidodinamico, termico e strutturale. Erano presenti una cinquantina di partecipanti provenienti da importanti aziende fra le quali ThyssenKrupp, il Gruppo Cividale, il Gruppo FOMAS e Hydromec, importante costruttore di presse bresciano col quale EnginSoft ha siglato un accordo di collaborazione tecnologica che porterà

Hydromec ad utilizzare FORGE per studiare nuove soluzioni di stampaggio e alla proposta congiunta di seminari tecnici per specifici settori dello stampaggio.

Nel complesso Made in Steel è stata un’esperienza positiva che ha dato modo ad EnginSoft di farsi conoscere maggiormente nel settore metallurgico, dove è presente da circa dieci anni con clienti molto importanti, ma che può essere ulteriormente sviluppato.

Hydromec srl ed Enginsoft hanno siglato un accordo di collaborazione tecnologica.

Brescia 1 Marzo 2011 – Hydromec srl, azienda di riferimento nella progettazione e costruzione di presse per lo stampaggio dei metalli ed Enginsoft, società italiana di maggior consistenza e tradizione nel settore della sperimentazione virtuale e del CAE, hanno siglato un accordo di collaborazione tecnica finalizzato all'integrazione e sviluppo delle proprie tecnologie.

Il processo di stampaggio dei metalli rappresenta oggi un settore manifatturiero di grande importanza e ad alto contenuto di sviluppo potenziale e, per questo motivo, le due Società hanno deciso di condividere know-how ed esperienze.

Da una parte la tecnologia meccanica, la qualità dei materiali e l'accuratezza di progettazione con sistemi innovativi di controllo e gestione di Hydromec srl, dall'altra software di simulazione, come Forge e Ansys, e di ottimizzazione dei parametri di processo come modeFRONTIER, rappresenteranno una base innovativa per risolvere le problematiche e le applicazioni tecnologicamente sempre più complesse che i processi di stampaggio oggi richiedono.

Hydromec srl Fondata nel 1980, Hydromec srl nasce come azienda per la revisione di macchine.

Ben presto Hydromec srl dirige i propri sforzi produttivi verso il settore dello stampaggio a caldo dell'ottone. Nascono così le presse della serie HF che si avvalgono di ben dieci brevetti tecnici. Successivamente Hydromec srl amplia la propria gamma di prodotti realizzando le presse oleodinamiche a quattro colonne della serie HSF utilizzate nel settore della forgiatura dell'acciaio a caldo, cui si aggiungono i laminatoi della serie LAR per la produzione di anelli, flange e sagomati in acciaio. Il Sistema Gestione Qualità di Hydromec è conforme alle norme UNI-EN ISO 9001:2008.

Per informazioni e contatti: www.hydromec.it

62 - Newsletter EnginSoft Year 8 n°1

EnginSoft Event Calendar ITALY EnginSoft is pleased to announce the next Seminars and Webinars. For more information, please contact: eventi@enginsoft.it Please visit www.enginsoft.com EnginSoft International Conference 2010 CAE Technologies for Industry.

To receive a copy of the The Conference Proceedings, please contact: eventi@enginsoft.it EnginSoft International Conference 2011 CAE Technologies for Industry Fiera di Verona - ITALY • 20-21 October 2011 Please stay tuned to www.caeconference.com for one of the major events for CAE Users in Europe!

FRANCE EnginSoft France 2011 Journées porte ouverte dans nos locaux à Paris et dans d’autres villes de France, en collaboration avec nos partenaires.

Pour plus d'information visitez: www.enginsoft-fr.com, contactez: info.fr@enginsoft.com Webinars Flowmaster: Introduction au logiciel Flowmaster • 31 March • 14 April 18 May - Etats Généraux Micado. France http://www.enginsoft-fr.com/events/index.html 15-16 June - Séminaire modeFRONTIER au CETIM Cetim Senlis. http://www.enginsoft-fr.com/events/index.html 20-26 June - Salon du Le Bourget - Paris Air Show Le Bourget, Paris.

Talk to our experts at the EnginSoft/ Flowmaster booth! http://www.paris-air-show.com/en 18-29 June – Teratec Conference. Ecole Polytechnique Palaiseau. Meet us at the EnginSoft / Flowmaster booth! http://www.enginsoft-fr.com/events/index.html 12 October - User Group Meeting modeFRONTIER France. Paris http://www.enginsoft-fr.com/events/index.html 13 October - User Group Meeting Flowmaster France. Paris http://www.enginsoft-fr.com/events/index.html

GERMANY Please stay tuned to: www.enginsoft-de.com Contact: info.de@enginsoft.com for more information. modeFRONTIER Seminars 2011. EnginSoft GmbH, Frankfurt am Main. Attend our regular Webinars and Seminars to learn more on how design optimization with modeFRONTIER. can enhance your product development processes 14-15 April - Efficient Design of Composite Structures – ESAComp Users' Meeting 2011. Technical University of Munich, Institute for Carbon Composites EnginSoft will be presenting: Optimization and robustness of composite structures: The whole design chain driven by modeFRONTIER In addition to presentations on simulation and design of composite structures, the latest advances in the ESAComp software will be presented. Three workshops will be held: the aerospace industry, - wind, marine energy and industrial applications, - optimizing composite structures.

EnginSoft is a sponsor of the Users’ Meeting http://www.enginsoft.com/events/esacomp_um.pdf www.esacomp.com Seminars Process Product Integration EnginSoft GmbH, Frankfurt am Main How to innovate and improve your production processes!

Seminars hosted by EnginSoft Germany and EnginSoft Italy SPAIN Programa de cursos de modeFRONTIER and other local events Please contact our partner, APERIO Tecnología: info@aperiotec.es Stay tuned to: www.aperiotec.es 10 March - National Instruments Day - Barcelona. Discover the latest trends in technology and new products from National Instruments during the Technology Forum on Graphic Design Systems.

AperioTec and ESTECO will be present to show the latest dedicated connection between modeFRONTIER and NI LabView.

Participation is free. http://www.ni.com/nidays/es/ 5-8 June - IDDRG 2011 International Conference. Bilbao (País Vasco). This year, in addition to stamping, material characterization, numerical simulation and tooling normally covered, the organizers would like to focus the conference on su-

Newsletter EnginSoft Year 8 n°1 -

stainability: of global concern, not only for industry but also for consumers, politicians and business leaders.

For more information, please visit: http://www.iddrg2011.eu/ SWEDEN 2011 Training Courses on modeFRONTIER - Drive your designs from good to GREAT EnginSoft Nordic offices in Lund, Sweden The Training Courses are focused on optimization, both multi- and single-objective, process automation and interpretation of results. Participants will learn different optimization strategies in order to complete a project within a specified time and simulation budget.

Other topics, such as design of experiments, metamodeling and robust design are introduced as well. The two day training consists of a mix of theoretical sessions and workshops.

• 7-8 April • 2-3 May • 7-8 June • 11-12 August • 5-6 September • 4-5 October • 2-3 November • 1-2 December To discuss your needs, for more information and to register, please contact EnginSoft Nordic, info@enginsoft.se UK The workshops are designed to give delegates a good appreciation of the functionality, application and benefits of modeFRONTIER. The workshops include an informal blend of presentation plus ‘hands-on’ examples with the objective of enabling delegates to be confident to evaluate modeFRONTIER for their applications using a trial license at no cost. modeFRONTIER Workshops Warwick Digital Laboratory, Warwick University • 12 April • 12 May • 21 June • 20 July • 17 August • 14 September • 13 October • 22 November • 14 December

• 16-17 May • 6-7 September For more information and to register, please visit www.enginsoft-uk.com. Contact: Bipin Patel, info@enginsoft.com 24-25 May - National Manufacturing Debate 2011. Vincent Building (Building 52), Cranfield campus, Cranfield University EnginSoft will be attending. www.cranfield.ac.uk/sas/manufacturingdebate 27-29 April - ESAFORM 2011. 14th International ESAFORM Conference on Material Forming. Belfast, Northern Ireland/UK. The purpose of this conference is to facilitate the communication between specialists in various fields of material forming sciences. Presentations concerning all the steps of material forming processes are welcome: from fundamental studies to applied aspects, from experimental to numerical research. Gino Duffett of AperioTec has been invited to give a plenary on advances and the future of simulation in the manufacturing industry www.qub.ac.uk/sites/ESAFORM2011/ GREECE 9 May - 5th PhilonNet CAE Conference. Athens. If you would like to present your work with ANSYS (including CFX, Fluent and Ansoft products), ANYBODY, DIFFPACK, ESACOMP, eta/DYNAFORM, eta/VPG, Flowmaster, FTI, LS-DYNA, modeFRONTIER, MOLDFLOW, SIMPLEWARE or ADVANTEDGE please send your abstract to: info@philonnet.gr For more information, please visit: www.philonnet.gr USA Courses on Design Optimization with modeFRONTIER Sunnyvale, CA For more information, please contact: training@ozeninc.com www.ozeninc.com JAPAN 17 June - CDAJ CAE Solution Conference 2011 modeFRONTIER Conference Day PAN PACIFIC Yokohama Bay Hotel Tokyu http://www.cdaj.co.jp/

modeFRONTIER Workshops with InfoWorks CS Warwick Digital Laboratory • 26 May • 9 November Please register for free on www.enginsoft-uk.com Multi-Disciplinary Optimization Training International Digital Lab, Warwick University

63

Course.

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: info@enginsoft.com

www.caeconference.com

速

ANSYS ITALIAN CONFERENCE 2011

Two major events coming together for the most significant occasion in the Italian CAE Calendar

APERS CALL FOR P EN IS NOW OP

CAE TECHNOLOGIES FOR INDUSTRY

ENGINSOFT INTERNATIONAL CONFERENCE 2011

VERONA -IT 20-21 OCTOBER