Eichinger, Matthias, Heinlein, Alexander ORCID: 0000-0003-1578-8104 and Klawonn, Axel ORCID: 0000-0003-4765-7387 (2020). Surrogate Convolutional Neural Network Models for Steady Computational Fluid Dynamics Simulations. Technical Report.

[img]
Preview
PDF
CDS_TR-2020-12.pdf

Download (8MB) | Preview

Abstract

A convolution neural network (CNN)-based approach for the construction of reduced order surrogate models for computational fluid dynamics (CFD) simulations is introduced; it is inspired by the approach of Guo, Li, and Iori [X. Guo, W. Li, and F. Iorio, Convolutional neural networks for steady flow approximation, in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ’16, New York, USA, 2016, ACM, pp. 481–490]. In particular, the neural networks are trained in order to predict images of the flow field in a channel with varying obstacle based on an image of the geometry of the channel. A classical CNN with bottleneck structure and a U-Net are compared while varying the input format, the number of decoder paths, as well as the loss function used to train the networks. This approach yields very low prediction errors, in particular, when using the U-Net architecture. Furthermore, the models are also able to generalize to unseen geometries of the same type. A transfer learning approach enables the model to be trained to a new type of geometries with very low training cost. Finally, based on this transfer learning approach, a sequential learning strategy is introduced, which significantly reduces the amount of necessary training data.

Item Type: Preprints, Working Papers or Reports (Technical Report)
Creators:
CreatorsEmailORCIDORCID Put Code
Eichinger, Matthiaseichingm@smail.uni-koeln.deUNSPECIFIEDUNSPECIFIED
Heinlein, Alexanderalexander.heinlein@uni-koeln.deorcid.org/0000-0003-1578-8104UNSPECIFIED
Klawonn, Axelaxel.klawonn@uni-koeln.deorcid.org/0000-0003-4765-7387UNSPECIFIED
URN: urn:nbn:de:hbz:38-297602
Series Name at the University of Cologne: Technical report series. Center for Data and Simulation Science
Volume: 2020,12
Date: 14 December 2020
Language: English
Faculty: Central Institutions / Interdisciplinary Research Centers
Divisions: Weitere Institute, Arbeits- und Forschungsgruppen > Center for Data and Simulation Science (CDS)
Subjects: Natural sciences and mathematics
Mathematics
Technology (Applied sciences)
Uncontrolled Keywords:
KeywordsLanguage
Convolutional neural networksEnglish
computational fluid dynamicsEnglish
reduced order surrogate modelsEnglish
U-NetEnglish
transfer learningEnglish
sequential learningEnglish
Refereed: No
URI: http://kups.ub.uni-koeln.de/id/eprint/29760

Downloads

Downloads per month over past year

Export

Actions (login required)

View Item View Item