Physics-based machine learning for fatigue lifetime prediction under non-uniform loading scenarios

2025-03-07

AI Summary

Predicting fatigue life of structures under non-uniform cyclic loading remains challenging due to pronounced load sequence effects and prohibitive computational costs of high-fidelity simulations. Method: This study proposes a physics-informed feedforward neural network (φML-FFNN), which explicitly embeds the core evolution law of an anisotropic continuum damage model into its architecture—enabling physics-driven, small-data modeling. The model is trained on hybrid data from experimental calibration and numerical simulations. Contribution/Results: Validated on concrete cylinders, φML-FFNN significantly improves fatigue life prediction accuracy under multi-level variable-amplitude loading compared to purely data-driven models, aligning with state-of-the-art experimental trends. It enables cross-load-history damage superposition analysis, demonstrating strong generalizability and physical interpretability. The framework supports real-time fatigue assessment in digital twin applications.

Technology Category

Machine Learning: Life-Long and Continual LearningCognitive Modeling & Cognitive Systems: Conceptual Inference and ReasoningReasoning under Uncertainty: Relational Probabilistic Models

Application Category

Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applicationsGraph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphsWeb Mining and Content Analysis: Large pretrained models with web data

Abstract

Accurate lifetime prediction of structures subjected to cyclic loading is vital, especially in scenarios involving non-uniform loading histories where load sequencing critically influences structural durability. Addressing this complexity requires advanced modeling approaches capable of capturing the intricate relationship between loading sequences and fatigue lifetime. Traditional fatigue simulations are computationally prohibitive, necessitating more efficient methods. This study highlights the potential of physics-based machine learning ($phi$ML) to predict the fatigue lifetime of materials. Specifically, a FFNN is designed to embed physical constraints from experimental evidence directly into its architecture to enhance prediction accuracy. It is trained using numerical simulations generated by a physically based anisotropic continuum damage fatigue model. The model is calibrated and validated against experimental fatigue data of concrete cylinder specimens tested in uniaxial compression. The proposed approach demonstrates superior accuracy compared to purely data-driven neural networks, particularly in situations with limited training data, achieving realistic predictions of damage accumulation. Thus, a general algorithm is developed and successfully applied to predict fatigue lifetimes under complex loading scenarios with multiple loading ranges. Hereby, the $phi$ML model serves as a surrogate to capture damage evolution across load transitions. The $phi$ML based algorithm is subsequently employed to investigate the influence of multiple loading transitions on accumulated fatigue life, and its predictions align with trends observed in recent experimental studies. This work demonstrates $phi$ML as a promising technique for efficient and reliable fatigue life prediction in engineering structures, with possible integration into digital twin models for real-time assessment.

Problem

• Predict fatigue lifetime under non-uniform loading scenarios
• Enhance accuracy using physics-based machine learning
• Efficiently model damage accumulation in complex loading

Innovation

• Physics-based machine learning for fatigue prediction
• FFNN embeds physical constraints for accuracy
• Surrogate model captures damage evolution efficiently