A physics-informed U-Net-LSTM network for nonlinear structural response under seismic excitation

TL;DR

This paper introduces a physics-informed deep learning model combining U-Net and LSTM architectures to accurately and efficiently predict nonlinear structural seismic responses, addressing limitations of traditional methods.

Contribution

The paper presents a novel physics-informed U-Net-LSTM framework that integrates physical laws into deep learning for improved seismic response prediction.

Findings

Enhanced predictive accuracy over conventional ML models

Reduced computational cost compared to FEM

Effective integration of physics constraints into deep learning

Abstract

Accurate and efficient seismic response prediction is essential for the design of resilient structures. While the Finite Element Method (FEM) remains the standard for nonlinear seismic analysis, its high computational demands limit its scalability and real-time applicability. Recent developments in deep learning - particularly Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Long Short-Term Memory (LSTM) models - have shown promise in reducing the computational cost of the nonlinear seismic analysis of structures. However, these data-driven models often struggle to generalize and capture the underlying physics, leading to reduced reliability. We propose a novel Physics-Informed U-Net-LSTM framework that integrates physical laws with deep learning to enhance both accuracy and efficiency. The proposed 1D U-Net captures the underlying latent features of the long-term input sequences. By embedding domain-specific constraints into the learning process, the proposed model achieves improved predictive performance over conventional Machine Learning (ML) architectures. This approach bridges the gap between purely data-driven methods and physics-based modeling, offering a robust and computationally efficient alternative for predicting the seismic response of structures.