{"title":"Multi-objective optimization design for anti-fatigue lightweight of dump truck carriage combined with machine learning","authors":"Kejun Lan, Wenyan Yu, Chengjie Huang, Yongjian Zhou, Zihang Li, Wei Huang","doi":"10.1177/16878132241269244","DOIUrl":null,"url":null,"abstract":"As urbanization continues to accelerate, dump trucks assume an increasingly important role in the transportation and construction of infrastructure. The carriage represents a critical structural assembly of dump trucks. One of the primary failure modes of the carriage is weld fatigue failure, which frequently gives rise to the problem of weld fatigue cracking during transportation. To increase the fatigue life of welds and enhance the degree of structural lightweight of a heavy dump truck carriage, a method for anti-fatigue lightweight design based on machine learning and multi-objective optimization is proposed. A high-fidelity finite element model of the carriage is established for static simulation analysis of the typical conditions. Based on the virtual reliability simulation test of the dump truck and the equivalent structural stress method, the fatigue life of the critical welds in the carriage is calculated. The important part thicknesses are selected as design variables through the comprehensive contribution analysis method. The maximum displacement and maximum stress under the dangerous condition are considered as constraints. The mass of the carriage and the minimum fatigue life of the critical welds are considered as optimization objectives. The GA-XGBoost machine learning approximation models (GA-XGBoost-MLAM) and NSGA-II algorithm are employed for multi-objective optimization design of the carriage. The entropy weighted TOPSIS method is utilized for multi-objective decision-making of Pareto solutions. The design after optimization and decision-making shows that, while satisfying the requirements of static structural performance, the minimum fatigue life mileage of the critical welds of the carriage is increased by 157,570 km, representing an increase of 36.58%. Additionally, the mass of the carriage is reduced by 295.69 kg, representing a decrease of 9.47%. Therefore, the proposed design method achieves a good effect in the anti-fatigue lightweight of dump truck carriage.","PeriodicalId":7357,"journal":{"name":"Advances in Mechanical Engineering","volume":null,"pages":null},"PeriodicalIF":2.1000,"publicationDate":"2024-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Advances in Mechanical Engineering","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1177/16878132241269244","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
As urbanization continues to accelerate, dump trucks assume an increasingly important role in the transportation and construction of infrastructure. The carriage represents a critical structural assembly of dump trucks. One of the primary failure modes of the carriage is weld fatigue failure, which frequently gives rise to the problem of weld fatigue cracking during transportation. To increase the fatigue life of welds and enhance the degree of structural lightweight of a heavy dump truck carriage, a method for anti-fatigue lightweight design based on machine learning and multi-objective optimization is proposed. A high-fidelity finite element model of the carriage is established for static simulation analysis of the typical conditions. Based on the virtual reliability simulation test of the dump truck and the equivalent structural stress method, the fatigue life of the critical welds in the carriage is calculated. The important part thicknesses are selected as design variables through the comprehensive contribution analysis method. The maximum displacement and maximum stress under the dangerous condition are considered as constraints. The mass of the carriage and the minimum fatigue life of the critical welds are considered as optimization objectives. The GA-XGBoost machine learning approximation models (GA-XGBoost-MLAM) and NSGA-II algorithm are employed for multi-objective optimization design of the carriage. The entropy weighted TOPSIS method is utilized for multi-objective decision-making of Pareto solutions. The design after optimization and decision-making shows that, while satisfying the requirements of static structural performance, the minimum fatigue life mileage of the critical welds of the carriage is increased by 157,570 km, representing an increase of 36.58%. Additionally, the mass of the carriage is reduced by 295.69 kg, representing a decrease of 9.47%. Therefore, the proposed design method achieves a good effect in the anti-fatigue lightweight of dump truck carriage.
期刊介绍:
Advances in Mechanical Engineering (AIME) is a JCR Ranked, peer-reviewed, open access journal which publishes a wide range of original research and review articles. The journal Editorial Board welcomes manuscripts in both fundamental and applied research areas, and encourages submissions which contribute novel and innovative insights to the field of mechanical engineering