Transfer Project T7

Adaptation and application of non-deterministic simulation and design strategies in the engineering practice

Director

Prof. Dr.-Ing. Wolfgang Graf; federführend
Prof. Dr.-Ing. habil. Michael Kaliske
Institute of Statics and Dynamics of Structures

Staff

Dipl.-Ing. Stephan Pannier

Partner Institution

BMW AG
Contact persons: Dr. sc. ETH Zürich Kathrin Grossenbacher, Dipl.-Math. Markus Ganser

DYNAmore GmbH
Contact persons: Dr.-Ing. Heiner Müllerschön, Dr.-Ing. Martin Liebscher

Objectives

The industrial engineering practice is dominated by applying deterministic numerical simulation methods and iterative design strategies. These methods are applicable to simulate the structural behavior of reinforced concrete structures or alternatively to simulate complex technical processes like metal forming processes, crash analysis of vehicles, etc. However, frequently discrepancies between numerical predicted and real observed technical parameters arise. These discrepancies are caused by information deficits/ data uncertainty in specifying parameters of strucures and processes. Data uncertainty results amongst others from incomplete, fluctuating or imprecise data, measure inaccuracies, or linguistic assessments of technical parameters. A frequently used scientific classification distinguishes between aleatoric and epistemic uncertainty. The uncertainty models fuzziness, randomness and fuzzy randomness trace back to this classification. Data are represented non-deterministically with the aid of those uncertainty models.

The current research already provide simulation methods and design strategies, which incorporate data uncertainty. The aim of this project is the adaptation and application of non-deterministic simulation and design strategies, developed at the institute for structural analysis and supported by CRC 528. The consideration of data uncertainty substantially contributes to obtain more reliable numerical prognosis and designs.

In the automotive manufacturing construction and process parameters are vague, imprecise and partially unknown in an early design stage. This information deficit can be modelled with the introduced uncertainty models above.In the transferproject the non-deterministic simulation- and design strategies are proven in the industrial engineering practice of the automotive manufacturing. The application of non-deterministic simualtion strategies yields uncertain result quantities. They reflect the fluctuation margin of the parameters, which are the basis for technical reasonable decisions. The design process is performed by means of inverse solution or fuzzy optimization methods.

In the scope of the transferproject the provided solutions, developed at the TU Dresden, should be adapted to enable reliable designs under consideration of uncertain data bases for the process chain metal forming - crash. This requires the development of appropriate interfaces and visualisation tools for diverse processes.

By reason of feedback provided by the cooperation partners the adaptation and application of non-deterministic simulation and design strategies in the engineering practice will advance impulses for further research activities.

Methods

The work in the transferproject is based on scientific methods for research and development of numerical simulation and design strategies. The origin of the simulation and design strategies is the theory of fuzzy random numbers, fuzzy set theory, possibility theory, stochastic and methods of explorative data analysis. In application are concepts, which bases on generalized theoretical approaches for capturing uncertainty of space and time dependent parameters (fuzzy stochastic finite element method).

The algorithms are integrated in a software, which is a requirement for numerical simulation and design with uncertain data.

Numerical simulation methods presuppose the availability of data and the specification of uncertainty as well as the modelling of those data with fuzzy quantities or fuzzy random quantities. Those quantities may be extended to fuzzy functions and fuzzy random functions. The assessment of obtained results requires the intensive dialog with colleagues from the industrial engineering practice to reach consensus on aspired design measures. The cooperating partners enable the professional dialog with structural designers.

Publications

2011

Sickert, J.-U.; Graf, W.; Pannier, S.: Numerical design approaches of textile reinforced concrete strengthening under consideration of imprecise probability Structure and Infrastructure Engineering 7 (2011), pp. 163–176 – doi:10.1080/15732471003588700

2010

Graf, W.; Kaliske, M.; Sickert, J.-U.; Pannier, S.; Freitag, S.: Neural Networks and Imprecise Probability Concepts for the Design of Industry-Sized Structures. In: Khalili, N.; Valliappan, S.; Li, Q.; Russell, A. (Eds.): Proceedings of the 9th World Congress on Computational Mechanics (WCCM), Sydney, 2010, pp. 72-73
Sickert, J.-U.; Pannier, S.; Graf, W.; Kaliske, M.: Efficient Numerical Reliability Based Design of TRC Strengthening Layers for Complex Structures. In: Brameshuber, W. (ed.): Proceedings of the International RILEM Conference on Material Science (MatSci) - Volume I, 2nd International Conference of Textile Reinforced Concrete (ICTRC), Aachen, RILEM Publications S.A.R.L., Bagneux, 2010, pp. 307-318
Pannier, S.; Sickert, J.-U.; Graf, W.; Kaliske, M.: Tailored metamodels for fuzzy reliability based optimization task. In: Rodrigues, H. et al. (eds.): Proceedings of the 2nd International Conference on Engineering Optimization (EngOpt), Lisboa, 2010, pp. 264-265, Volltext (10 S.) CD-ROM
Graf, W.; Sickert, J.-U.; Pannier, S.; Kaliske, M.: Robust design with uncertain data and response surface approximation. In: Beer, M.; Muhanna, R.L.; Mullen, R.L. (eds.): Proceedings of the 4th International Workshop on Reliable Engineering Computing (REC 2010), Singapore, 2010. Singapore : Research Publishing Services, 2010, pp. 554-574 – doi:10.3850/978-981-08-5118-7_027
Pannier, S.; Graf, W.: Sectional sensitivity measures with artificial neural networks. In: 9. LS-DYNA Forum 2010, Bamberg, Kap. I-I, pp. 21-34
Pannier, S.; Grossenbacher, K.; Liebscher, M.; Ganser, M.; Graf, W.; Müllerschön, H.; Lipp, A.; Kaliske, M.: Increasing reliability of metal forming processes in early design stages. In: 9. LS-DYNA Forum 2010, Bamberg, Kap. C-II, pp. 17-27
Pannier, S.; Graf, W.; Kaliske, M.; Grossenbacher, K.; Ganser, M.; Lipp, A.; Liebscher, M.; Müllerschön, H.: Affecting reliability of deep drawing processes in early design stages. In: Dudeck, F. et al. (eds.): 8th ASMO UK/ISSMO Conference Engineering Design Optimization, Proceedings, London, 2010, pp. 57-58
Pannier, S.; Graf, W.: Assessment of structural sensitivity on the basis of artificial neural networks. Proceedings in Applied Mathematics and Mechanics 10 (2010), pp. 205-206 – doi:10.1002/pamm.201010095
Graf, W.; Kaliske, M.; Sickert, J.-U.; Pannier, S.: Konzepte zur Beurteilung der Robustheit von Tragwerken. 14. Dresdner Baustatik-Seminar, TU Dresden, Institut für Statik und Dynamik der Tragwerke, 2010, S. 69-91

2009

Sickert, J.-U.; Graf, W.; Pannier, S.: Entwurf von Textilbetonverstärkungen - computerorientierte Methoden mit verallgemeinerten Unschärfemodellen. In: Curbach, M. (Hrsg.), Jesse, F. (Hrsg.): Textile Reinforced Structures : Proceedings of the 4th Colloquium on Textile Reinforced Structures (CTRS4) und zur 1. Anwendertagung, Dresden, 3.-5.6.2009. SFB 528, Technische Universität Dresden, D–01062 Dresden : Eigenverlag, 2009, S. 313-324 URN: urn:nbn:de:bsz:14-ds-1244047293129-54264
Piotrov, A.; Liebscher, M.; Pannier, S.; Graf, W.: Grouping detection of uncertain structural processes by means of cluster analysis. In: 7. European LS-DYNA Conference, Salzburg, 2009. Proceedings, S. 128 and CD-ROM (11 S.)
Sickert, J.-U.; Pannier, S.; Graf, W.; Jenkel, C.: Robustness assessment of structures incorporating generalized uncertainty models. In: Furuta, H.; Frangopol, D.M.; Shinozuka, M. (eds.): Safety, Reliability and Risk of Structures, Infrastructures and Engineering Systems, Proceedings of the 10th International Conference on Structural Safety and Reliability (ICOSSAR2009). London : Taylor & Francis, p. 111, Volltext (8 S.), CD-ROM
Pannier, S.; Sickert, J.-U.; Graf, W.: Patchwork approximation scheme for reliability assessment and optimization. In: Furuta, H.; Frangopol, D.M.; Shinozuka, M. (eds.): Safety, Reliability and Risk of Structures, Infrastructures and Engineering Systems, Proceedings of the 10th International Conference on Structural Safety and Reliability (ICOSSAR2009). London : Taylor & Francis, p. 531, Volltext (8 S.), CD-ROM
Möller, B.; Liebscher, M.; Pannier, S.; Graf, W.; Sickert J.-U.: An inverse solution of the lifetime-oriented design problem. In: Structure and Infrastructure Engineering: Maintenance, Management, Life-Cycle Design and Performance (2009), online available – doi:10.1080/15732470802658946
Kröger, T.; Pannier, S.; Kaliske, M.; Altrogge, I.; Graf, W.; Preusser, T.: Optimale applicator placement in hepatic radiofrequency ablation on the basis of rare data.Computer Methods in Biomechanics and Biomedical Engineering (2009), online available – doi:10.1080/10255840903317394