TY - GEN

T1 - PROBABILISTIC FATIGUE DATA ANALYSIS USING PHYSICS-GUIDED MIXTURE DENSITY NETWORKS

AU - Chen, Jie

AU - Liu, Yongming

N1 - Publisher Copyright: © 2023 by ASME.

PY - 2023

Y1 - 2023

N2 - 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.

AB - 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.

KW - constrained optimization

KW - fatigue

KW - neural network

KW - physics-guided machine learning

KW - probabilistic

KW - uncertainty quantification

UR - http://www.scopus.com/inward/record.url?scp=85176812220&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85176812220&partnerID=8YFLogxK

U2 - 10.1115/ssdm2023-104976

DO - 10.1115/ssdm2023-104976

M3 - Conference contribution

T3 - Proceedings of ASME 2023 Aerospace Structures, Structural Dynamics, and Materials Conference, SSDM 2023

BT - Proceedings of ASME 2023 Aerospace Structures, Structural Dynamics, and Materials Conference, SSDM 2023

PB - American Society of Mechanical Engineers

T2 - ASME 2023 Aerospace Structures, Structural Dynamics, and Materials Conference, SSDM 2023

Y2 - 19 June 2023 through 21 June 2023

ER -