A Generative Adversarial Network Approach to EstimateFinite Element Displacement
The finite element analysis(FEA) usually requires complex procedures to set up, and depends on constitutive relation to obtain final simulation results. In order to explore solutions other than the finite element method, the displacement response is considered as a picture generation process with gi...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Office of Control and Information Technology
2019-01-01
|
| Series: | Kongzhi Yu Xinxi Jishu |
| Subjects: | |
| Online Access: | http://ctet.csrzic.com/thesisDetails#10.13889/j.issn.2096-5427.2019.06.300 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The finite element analysis(FEA) usually requires complex procedures to set up, and depends on constitutive relation to obtain final simulation results. In order to explore solutions other than the finite element method, the displacement response is considered as a picture generation process with given conditions, bypassing the physical method. Based on generative adversarial network(GAN), and combined with DCGAN and CGAN, a deep learning model was proposed to directly obtain the displacement solution of two-dimensional plane instead of the finite element method. The proposed model is trained to make the generated displacement distribution close to FEA. The results show that the approximate distribution of displacement can be obtained by this model, and the calculation time is also lower than that of the FEA, which verifies the feasibility of solving displacement response by GAN. |
|---|---|
| ISSN: | 2096-5427 |