Rate-dependent multiaxial life prediction for polyamide-6 considering ratchetting: Semi-empirical and physics-informed machine learning models

Jingye Yang,Guozheng Kang,Qianhua Kan

DOI: https://doi.org/10.1016/j.ijfatigue.2022.107086

IF: 5.489

2022-06-30

International Journal of Fatigue

Abstract:The rate-dependent multiaxial fatigue life of PA6 is investigated by conducting a series of stress-controlled multiaxial fatigue tests involving the ratchetting. Based on the obtained experimental results and the critical plane approach, a semi-empirical life-prediction model is proposed, in which the effects of loading path, stress level, and stress rate are considered. To avoid the simplifications included in the semi-empirical model, a hybrid life-prediction model combing the critical plane approach and neural network is further developed for obtaining better accuracy and applicability. In addition, to compensate the insufficiency of data samples, two physical constraints (i.e., restricting the space of model parameters and generating domain knowledge-based data samples) are designed. The results demonstrate that: the hybrid model presents a better prediction accuracy than that of semi-empirical one; furthermore, by embedding the designed constraints into the hybrid model to steer the training of NN, the physically consistent solution is obtained and the extrapolation performance of modified hybrid model is significantly improved.

materials science, multidisciplinary,engineering, mechanical