PROBABILISTIC FATIGUE DATA ANALYSIS USING PHYSICS-GUIDED MIXTURE DENSITY NETWORKS

Jie Chen, Yongming Liu

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

Abstract

A Physics-guided Mixture Density Network (PgMDN) model is proposed for uncertainty quantification of fatigue data analysis in this paper. It integrates a Mixture Density Network for probabilistic modeling and physics knowledge as regularizations. This model can handle arbitrary distribution of data (e.g., strongly non-Gaussian, multi-mode, and truncated distributions). The physics knowledge from parameters and their partial derivatives is used as equality/ inequality constraints. The training of the physics-guided machine learning is formulated as a constrained optimization problem. To train the neural network with the commonly used backpropagation algorithm, the constrained optimization problem is transformed to an unconstrained one using a dynamic penalty function algorithm. With the physics constraints, the required training data sized can be reduced and the overfitting problem can be mitigated. The PgMDN is applied for fatigue stress-life curve estimation for multiple data sets. Some discussions are given to illustrate the effectiveness of incorporating the physics knowledge, the improvement of the dynamic penalty function method compared with the static method, and the benefits achieved from the distribution mixture compared with a single Gaussian distribution.

Original languageEnglish (US)
Title of host publicationProceedings of ASME 2023 Aerospace Structures, Structural Dynamics, and Materials Conference, SSDM 2023
PublisherAmerican Society of Mechanical Engineers
ISBN (Electronic)9780791887141
DOIs
StatePublished - 2023
EventASME 2023 Aerospace Structures, Structural Dynamics, and Materials Conference, SSDM 2023 - San Diego, United States
Duration: Jun 19 2023Jun 21 2023

Publication series

NameProceedings of ASME 2023 Aerospace Structures, Structural Dynamics, and Materials Conference, SSDM 2023

Conference

ConferenceASME 2023 Aerospace Structures, Structural Dynamics, and Materials Conference, SSDM 2023
Country/TerritoryUnited States
CitySan Diego
Period6/19/236/21/23

Keywords

  • constrained optimization
  • fatigue
  • neural network
  • physics-guided machine learning
  • probabilistic
  • uncertainty quantification

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Building and Construction
  • Mechanics of Materials
  • Architecture

Cite this