GENERIC APPROACH TO PLAUSIBILITY CHECKS FOR STRUCTURAL MECHANICS WITH DEEP LEARNING

Sprügel T, Schröppel T, Wartzack S (2017)

Publication Language: English

Publication Type: Book chapter / Article in edited volumes

Publication year: 2017

Publisher: Vancouver, Canada, 21.-25.08.2017

Edited Volumes: Proceedings of the 21st International Conference on Engineering Design (ICED17)

Book Volume: Vol. 1: Resource-Sensitive Design | Design Research Applications and Case Studies

Pages Range: 299 - 308

URI: https://www.archiv.mfk.tf.fau.de?file=pubmfk_59ba7ca14607d

Abstract

The simulation of product behavior is a vital part in virtual product development, but currently there is no tool or method available that can examine the quality of FE simulations and decide automatically on whether a simulation is plausible or non-plausible. In the paper a method is presented that enables automatic plausibility checks on basis of empirical simulation datasets. Nodal simulation data is transformed to numerical arrays, of fixed size, using virtual spherical detector surfaces. Afterwards the arrays are used to train a Deep Convolutional Neural Network (AlexNet). The Neural Network can then be used for plausibility checks of FE simulations (structural mechanics). In a first application a Deep Convolutional Neural Network is trained with simulation data of a demonstrator part, the rail of speed inline skates. After the GPU training of the Neural Network, further simulations are evaluated with the net. These simulations were not part of the training data and are used to calculate the prediction quality
of the Neural Network. This approach is to support development engineers during design accompanying FEA in virtual product development.

Authors with CRIS profile

Tobias Sprügel Chair of Engineering Design Tina Buker Chair of Engineering Design Sandro Wartzack Chair of Engineering Design

Related research project(s)

None - None Vorhersage von Produkteigenschaften mittels Data-Mining-Methoden (EFRE E|ASY OPT) Kompetenz- und Analyseprojekt für die „datengetriebene Prozess- und Produktionsoptimierung mittels Data Mining und Big Data“ Jan. 1, 2017 - Dec. 31, 2020

How to cite

APA:

Sprügel, T., Schröppel, T., & Wartzack, S. (2017). GENERIC APPROACH TO PLAUSIBILITY CHECKS FOR STRUCTURAL MECHANICS WITH DEEP LEARNING. In Proceedings of the 21st International Conference on Engineering Design (ICED17). (pp. 299 - 308). Vancouver, Canada, 21.-25.08.2017.

MLA:

Sprügel, Tobias, Tina Schröppel, and Sandro Wartzack. "GENERIC APPROACH TO PLAUSIBILITY CHECKS FOR STRUCTURAL MECHANICS WITH DEEP LEARNING." Proceedings of the 21st International Conference on Engineering Design (ICED17). Vancouver, Canada, 21.-25.08.2017, 2017. 299 - 308.

BibTeX: Download