期刊
COMPUTERS AND GEOTECHNICS
卷 159, 期 -, 页码 -出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.compgeo.2023.105472
关键词
Physics-informed Neural Network (PINN); Artificial Neural Network; Deep learning; Constitutive modeling; Elasto-plasticity
In this work, a deep neural network architecture called Elasto-Plastic Neural Network (EPNN) is proposed to efficiently surrogate classical elasto-plastic constitutive relations. The EPNN incorporates physics aspects of classical elasto-plasticity, allowing for more efficient training with less data and better extrapolation capability. The architecture is model and material-independent and can be adapted to a wide range of elasto-plastic materials. The superiority of EPNN is demonstrated through predicting strain-controlled loading paths for sands with different initial densities.
In this work, we present a deep neural network architecture that can efficiently surrogate classical elasto-plastic constitutive relations. The network is enriched with crucial physics aspects of classical elasto-plasticity, including additive decomposition of strains into elastic and plastic parts, and nonlinear incremental elasticity. This leads to a Physics-Informed Neural Network (PINN) surrogate model named here as Elasto-Plastic Neural Network (EPNN). Detailed analyses show that embedding these physics into the architecture of the neural network facilitates a more efficient training of the network with less training data, while also enhancing the extrapolation capability for loading regimes outside the training data. The architecture of EPNN is model and material-independent; it can be adapted to a wide range of elasto-plastic material types, including geomaterials; and experimental data can potentially be directly used in training the network. To demonstrate the robustness of the proposed architecture, we adapt its general framework to the elasto-plastic behavior of sands. We use synthetic data generated from material point simulations based on a relatively advanced dilatancy-based constitutive model for granular materials to train the neural network. The superiority of EPNN over regular neural network architectures is demonstrated through predicting unseen strain-controlled loading paths for sands with different initial densities.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据