A multi-fidelity physics-informed machine learning framework for probabilistic low-cycle fatigue life prediction of shot-peened materials
Authors:
- Zhichun ZhouZZZhichun Zhou
- Dianyin HuDHDianyin Hu
- Jianxing MaoJMJianxing Mao
- Huanhuan ChenHCHuanhuan Chen
- Liu XiLXLiu Xi
- Sanfeng XinSXSanfeng Xin
- Liucheng ZhouLiucheng Zhou
- Xiao SuXSXiao Su
- Rongqiao WangRWRongqiao Wang
Preprint
- 5 March 2026
- preprint
- Published by Elsevier in SSRN Electronic Journal
Abstract
Inherent randomness in the shot peening (SP) process induces significant dispersion in structural integrity (SI) parameters. Due to the limitation of high-fidelity fatigue test data, severe challenge exists in constructing high-precision probabilistic low-cycle fatigue (LCF) life prediction for turbine disks processed by shot peening. In this study, a multi-fidelity physics-informed machine learning (MF-PIML) framework integrating physical mechanisms with multi-source data is proposed. First, through LCF tests and SI characterization across four typical surface conditions, the governing mechanisms of surface roughness and residual stress distribution on the scatter of life are revealed. Accordingly, SI correction terms are introduced to modify the classical Smith-Watson-Topper (SWT) model, establishing a physical baseline with predictions falling within the scatter band of 2.8. Subsequently, a mean-variance dual-output neural network based on transfer learning is constructed. Abundant low-fidelity finite element simulation data are utilized for pre-training to capture underlying physical mapping laws, followed by fine-tuning with sparse high-fidelity experimental data. Concurrently, physical constraints and negative log-likelihood terms are embedded into the loss function to ensure physical consistency and enable uncertainty quantification. Validation results demonstrate that the MF-PIML model successfully narrows the life prediction accuracy to within the scatter band of 1.8. Furthermore, SHapley Additive exPlanations (SHAP) analysis is employed to quantify the contributions of key SI parameters, significantly enhancing model transparency and interpretability. This study provides an efficient and reliable predictive tool for the design and process optimization of high-performance turbine disks.Keywords
This publication has 22 references indexed in Scilit:
- Fretting wear simulation based on dynamic behavior of debris layer considering surface morphology influenceTribology International, 2025
- A convolutional neural network-based deep learning approach for predicting surface chloride concentration of concrete in marine tidal zonesScientific Reports, 2025
- Fatigue life predictor: predicting fatigue life of metallic material using LSTM with a contextual attention modelRSC Advances, 2025
- Model reduction for fatigue life estimation of a welded joint driven by machine learningHeliyon, 2024
- Probabilistic physics-guided machine learning for fatigue data analysisExpert Systems with Applications, 2021
- Simulation of the shot peening process with variable shot diameters and impacting velocitiesAdvances in Engineering Software, 2017
- Bayesian Reliability and Performance Assessment for Multi-State SystemsIEEE Transactions on Reliability, 2014
- Uncertainty Quantification in Gear Remaining Useful Life Prediction Through an Integrated Prognostics MethodIEEE Transactions on Reliability, 2013
- On the potential applications of a 3D random finite element model for the simulation of shot peeningAdvances in Engineering Software, 2009
- Bayesian reliability analysis for fuzzy lifetime dataFuzzy Sets and Systems, 2006