2020/02/18 by Ruiyang Zhang, Yang Liu, Hao Sun · 1 citation
Computer Science · Engineering · Mathematics · Physics and Astronomy · #Artificial intelligence #Artificial neural network #Computer science #Deep learning #Extrapolation #Hydraulic and Pneumatic Systems #Machine learning #Mathematics #Metamodeling #Model Reduction and Neural Networks #Nonlinear system #Overfitting #Physics #Robustness (evolution) #Software engineering #Structural Health Monitoring Techniques #Theoretical computer science #cs.CE #eess.SP
paper · pdf · doi:10.1016/j.cma.2020.113226
21 pages, 13 figures
arxiv created 2020/02/18 · openalex publication_date 2020/06/27 · arxiv updated 2020/07/15 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
This paper introduces an innovative physics-informed deep learning framework for metamodeling of nonlinear structural systems with scarce data. The basic concept is to incorporate physics knowledge (e.g., laws of physics, scientific principles) into deep long short-term memory (LSTM) networks, which boosts the learning within a feasible solution space. The physics constraints are embedded in the loss function to enforce the model training which can accurately capture latent system nonlinearity even with very limited available training datasets. Specifically for dynamic structures, physical laws of equation of motion, state dependency and hysteretic constitutive relationship are considered to construct the physics loss. In particular, two physics-informed multi-LSTM network architectures are proposed for structural metamodeling. The satisfactory performance of the proposed framework is successfully demonstrated through two illustrative examples (e.g., nonlinear structures subjected to ground motion excitation). It turns out that the embedded physics can alleviate overfitting issues, reduce the need of big training datasets, and improve the robustness of the trained model for more reliable prediction. As a result, the physics-informed deep learning paradigm outperforms classical non-physics-guided data-driven neural networks.