基于二元混合随机过程的轴承剩余寿命预测  

Binary hybrid stochastic process-based approach for the estimation of bearing remaining useful life

作　　者：金晓航 李建华 郭远晶[4] 贾虹[1,2] Jin Xiaohang;Li Jianhua;Guo Yuanjing;Jia Hong(Key Laboratory of Special Purpose Equipment and Advanced Processing Technology of Ministry of Education,Zhejiang University of Technology,Hangzhou 310023;College of Mechanical Engineering,Zhejiang University of Technology,Hangzhou 310023;Ninghai ZJUT Academy of Science and Technology,Ninghai 315600;Zhijiang College,Zhejiang University of Technology,Shaoxing 312030)

机构地区：[1]特种装备制造与先进加工技术教育部重点实验室(浙江工业大学),杭州310023 [2]浙江工业大学机械工程学院,杭州310023 [3]宁海县浙工大科学技术研究院,宁海315600 [4]浙江工业大学之江学院,绍兴312030

出　　处：《高技术通讯》2020年第12期1284-1291,共8页Chinese High Technology Letters

基　　金：国家自然科学基金(51505424,51675484);宁波市自然科学基金(2018A610045);特种装备制造与先进加工技术教育部/浙江省重点实验室(浙江工业大学)开放基金(EM2019120104,EM2017120101);浙江省自然科学基金(LQ17E050006,LY15E050019)资助项目。

摘　　要：针对轴承退化问题,基于两个不同变化特点的性能指标,提出了基于二元混合随机过程的轴承剩余寿命(RUL)预测方法。首先,分析两个特点不同的性能指标,选择合适的随机过程(伽马过程或维纳过程)分别构建基于不同性能指标的退化模型;其次,利用Copula函数分析两个性能指标间的相关特性并构建剩余寿命的联合概率密度函数;然后采用分步极大似然估计法在线更新模型参数,预测未来时刻的剩余寿命;最后,通过仿真和轴承实验数据对所提方法进行验证分析。结果显示所提方法能有效地预测轴承的剩余寿命,通过与基于一元随机过程的剩余寿命预测方法的对比分析,发现所提方法具有更好的预测精度。A binary hybrid stochastic process-based approach is proposed to estimate bearing’s remaining useful life(RUL).Firstly,two different health indices are constructed to analyze the characteristic of the degradation process,appropriate stochastic process(Gamma process or Wiener process)is selected to construct the degradation model;Secondly,a Copula function is used to analyze the correlation between these two health indices,and then a joint probability density function is built.The maximum likelihood estimation algorithm is used to estimate and update model’s parameters.Finally,the RUL is predicted.The proposed method is verified by a simulation data and an experimental life data of bearing.The results show that the proposed method can predict bearing’s RUL effectively.Compared with the unary stochastic process-based approach in RUL estimation,the proposed approach has better performance.

关 键 词：二元混合随机过程 COPULA函数 剩余寿命(RUL)预测 极大似然估计 

分 类 号：G63[文化科学—教育学]