The possibilities of saving computing time by using submodels are explored. For this purpose, optimization is performed in different levels in an interlocked manner. For example, the following levels are involved:
- Level 1: Finding the best possible concept
- Level 2: Optimal design of main system components
- Level 3: Detailed optimization of individual components
Different design variables and different objective and restriction functions are taken into account in the individual levels. The optimizations of the individual levels are coupled by appropriate interlocking algorithms. This enables the optimization of large systems. In the BMBF project "Development of software methods for efficient substitute model-based optimization for crash design in the vehicle development process (eEgO)", which is part of the KMUInnovativ funding program, the methods were implemented as prototypes together with partners.
Wielens, S., Benito Cia, L., Schumacher, A. (2022): Using automatically generated submodels in vehicle body crash optimization, International Journal of Crashworthiness, DOI: 10.1080/13588265.2022.2109873
Büttner, J., Schumacher, A. (2020): "Activities for Speed up the Optimization Processes in Car Body Design", Proceeding of the Automotive CAE Grand Challenge 2020, 29th and 30th of October 2020 in Hanau, Germany
Benito-Cia, L., Wielens, S., Schöne, C., Schumacher, A. (2020): "Automatic Submodel-based Multi-Level Optimization of Crash Structures", Proceeding of the Automotive CAE Grand Challenge 2020, 29th and 30th of October 2020 in Hanau, Germany
Mertler, S., Schumacher, A., Jansen, L. (2020): "Reduced Order Modeling for Correlation Analysis of Crash Structures ", Proceeding of the Automotive CAE Grand Challenge 2020, 29th and 30th of October 2020 in Hanau, Germany
Kracker, D., Garcke, J., Schumacher, A., Schwanitz, P. (2020): "Automatic analysis of crash simulations with dimensionality reduction algorithms such as PCA and t-SNE", 16th Intl. LS-DYNA Users Conference, June 10-11, 2020 www.dynalook.com/conferences/16th-international-ls-dyna-conference/automotive/t1-1-a-automotive-011.pdf/view
Schumacher, A., Singh, H., Wielens, S. (2019): "Submodel-based Multi-Level optimization of crash structures using statistically generated universal correlations of the different levels", Proceeding of the World Congress of Structural and Multidisciplinary Optimization, May 20-24, 2019, Beijing, China
Walser, A.F., Schumacher, A., Singh, H., Schöne, C., Benito Cia, L. (2018): „Verschachtelte Optimierung des Crashverhaltens von Fahrzeugen mit automatisch erzeugten und bewerteten Submodellen“, erscheint in den VDI-Berichten zum 19. VDI-Kongress SIMVEC – Simulation und Erprobung in der Fahrzeugentwicklung, 20.-21.11.2018 in Baden Baden
Falconi, C.J.; Walser, A.F.; Singh, H.; Schumacher, A. (2018): “Automatic generation, validation and correlation of the submodels for the use in the Optimization of crashworthy structures", in: Schumacher, A., Vietor, T., Fiebig, S., Bletzinger, K.-U., Maute, K. (Hrsg.): Advances in Structural and Multidisciplinary Optimization, Proceedings of the 12th World Congress of Structural and Multidisciplinary Optimization (WCSMO12), Springer Nature, 1558-1571, 2018 http://www.springer.com/de/book/9783319679877
Schumacher, A., Singh, H., Falconi, C.(2017): "Möglichkeiten der Entwicklung von Fahrzeugkomponenten mit optimierten Submodellen", Umdruck der VDI-Konferenz „Cabriolets“, 28.-29.06.2017 in Nürtingen, Deutschland
Singh, H., Schumacher, A., Falconi, C.J., Walser, A. F., Trentmann, S., Leyre Benito, C., Foussette, C., Krause, P. (2017): „Hierarchical Multi-Level-Optimization of crashworthy structures using automatic generated submodels”, Proc. of the 11th European LS-DYNA Conference 2017, 9. - 11. Mai 2017, Salzburg, Austria, 2017
Link, S., Singh, H., Schumacher, A. (2016): “Influence of submodel size and evaluated functions on the optimization process of crashworthiness structures", 14. LS-DYNA Forum 2016, Bamberg https://www.dynamore.de/de/download/papers/2016-ls-dyna-forum/Papers%202016/montag-10.10.16/crash/influence-of-submodel-size-and-evaluated-functions-on-the-optimization-process-of-crashworthiness-structures