农业装备与车辆工程2025，Vol.63Issue(4)：1-6,6.DOI:10.3969/j.issn.1673-3142.2025.04.001
基于SVR与混合Weibull分布融合模型的拖拉机可靠性评价方法研究
Research on tractor reliability evaluation method based on SVR and mixed Weibull distribution
摘要
Abstract
A tractor reliability evaluation method based on a fusion model of Support Vector Regression(SVR)and mixed Weibull distribution was proposed,aiming to address the issue of insufficient prediction accuracy of single models under complex operating conditions.The non-linear fitting capabilities of the SVR model were employed to conduct precise predictions of failure data.Simultaneously,the mixed Weibull distribution model was incorporated to describe the statistical characteristics of different failure modes,and a variable-weight fusion model was developed.The results showed that the variable-weight fusion model achieved high consistency between predicted reliability values and actual observations,outperforming single models in terms of Root Mean Square Error(RMSE)and Mean Absolute Percentage Error(MAPE).This method significantly improves the accuracy of reliability evaluation under complex operating conditions and provides a scientific basis for the design optimization and maintenance of tractors.关键词
拖拉机/可靠性评价/支持向量回归/混合Weibull分布/融合模型Key words
tractor/reliability evaluation/support vector regression/mixed Weibull distribution/fusion model分类
交通工程引用本文复制引用
张文贤,王蕾,吴孟,陈光阔,尹涛,韩兴昌,李光辉..基于SVR与混合Weibull分布融合模型的拖拉机可靠性评价方法研究[J].农业装备与车辆工程,2025,63(4):1-6,6.基金项目
工信部2024年质量工作重点项目"拖拉机可靠性在线验证、分析与评价" （）
山东省农业农村厅项目"大马力无级变速拖拉机研发制造推广应用"(NJYTHSD-202302) （NJYTHSD-202302）
