Nonlinear electro-elastic finite element analysis with neural network constitutive models
2024/03/28
In our recent work, we demonstrate the applicability of neural network constitutive models for complex electro-elastic finite element analysis.
To fully exploit the capabilities of electro-elastic metamaterials, efficient and accurate simulation tools are required. In our latest work, we demonstrate how this can be achieved by using physics-augmented neural network (PANN) constitutive models. We calibrated PANN constitutive models to different homogenized metamaterials, and applied them in complex nonlinear finite element simulations including large electrically induced deformations and instabilities.
Congratulations to our Ph.D. student Dominik Klein, and many thanks to Rogelio Ortigosa and Jesus Martinez-Frutos from Universidad Politecnica de Cartagena for this great collaboration!
The article “Nonlinear electro-elastic finite element analysis with neural network constitutive models” was just published in CMAME and is available at https://doi.org/10.1016/j.cma.2024.116910 (open access)
