A multiaxial low-cycle fatigue prediction method under irregular loading by ANN model with knowledge-based features

被引：23
作者：
Zhou, Tianguo [1 ,2 ]
Sun, Xingyue [1 ,2 ]
Chen, Xu [1 ,2 ,3 ]
机构：
[1] Tianjin Univ, Sch Chem Engn & Technol, Tianjin 300350, Peoples R China
[2] Haihe Lab Sustainable Chem Transformat, Tianjin 300192, Peoples R China
[3] Tianjin Key Lab Chem Proc Safety & Equipment Tech, Tianjin 300350, Peoples R China
来源：
INTERNATIONAL JOURNAL OF FATIGUE | 2023年 / 176卷
关键词：
Multiaxial fatigue; Irregular loading; Knowledge-based machine learning; Model interpretability; Feature selection; CRITICAL PLANE APPROACH; EQUIVALENT STRESS; LIFE PREDICTION; DAMAGE; CRITERION;
D O I：
10.1016/j.ijfatigue.2023.107868
中图分类号：
TH [机械、仪表工业];
学科分类号：
0802 ;
摘要：
An ANN model with knowledge-based features is proposed to predict multiaxial low-cycle fatigue life under irregular loading and verified on 304L stainless steel. Feature selection for knowledge-based features solves overfitting problem, improves model performance, and selects the genetic knowledge-based features. With help of genetic knowledge-based features, the proposed model combines physics knowledge and machine learning, being able to predict fatigue life of irregular cases through training only with regular cases. Most predicted results are located within 2-factor band. Besides, SHAP method interprets the ANN model, shows the contri-bution of features, and illustrates effectiveness of the model through the curves that fit facts.
引用
收藏
页数：12
相关论文
共 54 条
[1]   An equivalent stress process for fatigue life estimation under multiaxial loadings based on a new non linear damage model [J].
Aid, A. ;
Bendouba, M. ;
Aminallah, L. ;
Amrouche, A. ;
Benseddiq, N. ;
Benguediab, M. .
MATERIALS SCIENCE AND ENGINEERING A-STRUCTURAL MATERIALS PROPERTIES MICROSTRUCTURE AND PROCESSING, 2012, 538 :20-27
[2]   Using machine learning to predict lifetime under isothermal low-cycle fatigue and thermo-mechanical fatigue loading [J].
Bartosak, Michal .
INTERNATIONAL JOURNAL OF FATIGUE, 2022, 163
[3]   Network Dissection: Quantifying Interpretability of Deep Visual Representations [J].
Bau, David ;
Zhou, Bolei ;
Khosla, Aditya ;
Oliva, Aude ;
Torralba, Antonio .
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, :3319-3327
[4]   Some analytical expressions to measure the accuracy of the "equivalent von Mises stress" in vibration multiaxial fatigue [J].
Benasciutti, D. .
JOURNAL OF SOUND AND VIBRATION, 2014, 333 (18) :4326-4340
[5]   An analysis of mean stress in multiaxial random fatigue [J].
Brown, MW ;
Suker, DK ;
Wang, CH .
FATIGUE & FRACTURE OF ENGINEERING MATERIALS & STRUCTURES, 1996, 19 (2-3) :323-333
[6]   A multiaxial fatigue criterion for random loading [J].
Carpinteri, A ;
Spagnoli, A ;
Vantadori, S .
FATIGUE & FRACTURE OF ENGINEERING MATERIALS & STRUCTURES, 2003, 26 (06) :515-522
[7]   An on-line algorithm of fatigue damage evaluation under multiaxial random loading [J].
Chen, Hong ;
Shang, De-Guang .
INTERNATIONAL JOURNAL OF FATIGUE, 2011, 33 (02) :250-254
[8]   Fatigue modeling using neural networks: A comprehensive review [J].
Chen, Jie ;
Liu, Yongming .
FATIGUE & FRACTURE OF ENGINEERING MATERIALS & STRUCTURES, 2022, 45 (04) :945-979
[9]   A weight function-critical plane approach for low-cycle fatigue under variable amplitude multiaxial loading [J].
Chen, X ;
Jin, D ;
Kim, KS .
FATIGUE & FRACTURE OF ENGINEERING MATERIALS & STRUCTURES, 2006, 29 (04) :331-339
[10]   A critical plane-strain energy density criterion for multiaxial low-cycle fatigue life under non-proportional loading [J].
Chen, X ;
Xu, S ;
Huang, D .
FATIGUE & FRACTURE OF ENGINEERING MATERIALS & STRUCTURES, 1999, 22 (08) :679-686
123456