基于不同退化阶段状态空间模型及粒子滤波的滚动轴承寿命预测  

Useful life prediction of rolling element bearings based on a particle filtering model and the state space model at different degradation stages

作　　者：马波[1] 翟斌[1] 彭琦[2] 王应全[3] 

机构地区：[1]北京化工大学机电工程学院故障诊断与自愈工程研究中心 [2]中国科学院高能物理研究所 [3]中国石油塔里木油田分公司

出　　处：《北京化工大学学报（自然科学版）》2017年第3期81-86,共6页Journal of Beijing University of Chemical Technology(Natural Science Edition)

基　　金：国家"863"计划(2014AA041806)

摘　　要：为解决滚动轴承全寿命数据有限及物理模型难以建立的问题,提出一种基于状态监测信息和滚动轴承退化物理模型的寿命预测方法。首先对Paris公式及Foreman公式进行改进,建立滚动轴承不同退化阶段状态空间模型,并将已知的滚动轴承物理数据及运行状态信息输入模型,利用最小二乘法对模型参数进行调整;然后利用粒子滤波算法对滚动轴承后期运行趋势进行递推预测;最后运用滚动轴承全寿命数据对所提方法进行验证,并将预测结果与单一状态空间模型及Gamma模型预测结果对比,结果表明该方法预测准确率更高,具有较强的实用性。In order to solve the problem of scant availability of whole life data for rolling bearings and the difficulty in establishing a physical model, we propose a method for the prediction of the useful life of rolling element bearings based on a particle filtering and the state space model at different degradation stages. The method improves the Paris formula and the Foreman formula, establishes the rolling element bearings state space model at different degrada- tion stages, puts the known physical data and the operating status information about the rolling bearing into the model, uses the least squares method to adjust model parameters, and finally utilizes the particle filter algorithm to predict the late-stage operating trends and realizes the rolling bearing life prediction. Rolling bearing full life experimental data are used to demonstrate the proposed method, and the predicted results are compared with the results using the single state space model and the Gamma model; the comparison results indicate that the method has higher prediction success with strong practicability.

关 键 词：不同退化阶段状态空间模型 滚动轴承 寿命预测 粒子滤波 

分 类 号：TH17[机械工程—机械制造及自动化]