Physics-Informed Neural Network for Nonlinear Structural System Identification

Tong Liu, Hadi Meidani

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Structural system identification is a critical task in resilience assessments, especially following a natural hazard. In this paper, we propose PIDynNet, a novel physics-informed approach that produces an ordinary differential equation-constrained neural network model for structural system identification. PIDynNet improves the estimation of structural parameters of nonlinear structural systems by embedding an auxiliary physics-based loss term into the overall loss function as well as a supervised data-driven loss term. The proposed framework has the generalization capability to predict nonlinear structural response given unseen ground excitations. Two nonlinear numerical experiments are conducted to demonstrate the advantage of PIDynNet over other identification methods in problems with or without latent variables.

Original languageEnglish (US)
Title of host publicationStructural Health Monitoring 2023
Subtitle of host publicationDesigning SHM for Sustainability, Maintainability, and Reliability - Proceedings of the 14th International Workshop on Structural Health Monitoring
EditorsSaman Farhangdoust, Alfredo Guemes, Fu-Kuo Chang
PublisherDEStech Publications
Pages3001-3011
Number of pages11
ISBN (Electronic)9781605956930
StatePublished - 2023
Event14th International Workshop on Structural Health Monitoring: Designing SHM for Sustainability, Maintainability, and Reliability, IWSHM 2023 - Stanford, United States
Duration: Sep 12 2023Sep 14 2023

Publication series

NameStructural Health Monitoring 2023: Designing SHM for Sustainability, Maintainability, and Reliability - Proceedings of the 14th International Workshop on Structural Health Monitoring

Conference

Conference14th International Workshop on Structural Health Monitoring: Designing SHM for Sustainability, Maintainability, and Reliability, IWSHM 2023
Country/TerritoryUnited States
CityStanford
Period9/12/239/14/23

ASJC Scopus subject areas

  • Computer Science Applications
  • Civil and Structural Engineering
  • Safety, Risk, Reliability and Quality
  • Building and Construction

Fingerprint

Dive into the research topics of 'Physics-Informed Neural Network for Nonlinear Structural System Identification'. Together they form a unique fingerprint.

Cite this