考虑环境改变的部件核密度剩余寿命预测  

Remaining useful life prediction of component kernel density estimation considering environmental changes

作　　者：赵彬 李佳娟 石慧[1] 任谦力 康辉 ZHAO Bin;LI Jiajuan;SHI Hui;REN Qianli;KANG Hui(School of Electronic and Information engineering,Taiyuan University of Science and Technology,Taiyuan 030024,China)

机构地区：[1]太原科技大学电子信息工程学院,太原030024

出　　处：《振动与冲击》2024年第8期145-154,共10页Journal of Vibration and Shock

基　　金：国家青年科学基金项目(61703297,72071183);山西省重点研发计划项目(202202100401002,202202090301011,202202150401005);山西省科技重大专项计划“揭榜挂帅”项目(202201090301013);山西省基础研究计划(自由探索类)面上项目(20210302123206,202203021211205);山西省回国留学人员科研资助项目(2021-135,2021-134);山西省留学人员科技活动择优资助项目(20220029)。

摘　　要：针对风机关键部件运行过程中剩余寿命预测精度受环境变化影响的问题,提出一种考虑环境冲击影响的核密度估计剩余寿命预测模型。首先采用自适应核密度的非参数估计方法对部件的连续自然退化过程建模;其次假设恶劣环境对部件的影响为部件受到的随机冲击,在考虑改变运行环境条件下建立其部件所受冲击与连续退化随机相关的剩余寿命预测模型,并分析其可靠性,最后通过风电机组退化数据及齿轮箱磨损实测数据进行了试验分析,验证了所提模型的准确性和有效性。To address the problem that the remaining useful life prediction accuracy of a component is affected by environmental changes during operation,a kernel density estimation remaining useful life prediction model considering the impact of environmental shocks wasproposed.Firstly,the non-parametric estimation method of adaptive kernel density was used to model the continuous natural degradation process of the component;secondly,assuming that the impact of the harsh environment on the component is therandom shocks on the component,the remaining useful life prediction model was established by introducing a virtual age function and a variable sudden failure threshold under the consideration of changing the operating environment conditions to correlate the impact on the component with the continuous degradation process stochastically,and its reliability was analyzed.Finally,the accuracy and validity of the proposed model were verified by experimental simulation analysis of wind turbine degradation data and gearbox wear measurement data.

关 键 词：核密度估计 环境改变 随机相关 剩余寿命预测 

分 类 号：TH17[机械工程—机械制造及自动化]