基于线性回归的滚动轴承寿命预测起点的确定方法
A Method for Determining the Starting Point of Life Prediction of Rolling Bearings Based on Linear Regression
张海燕 1袁国栋1
作者信息
- 1. 甘肃建筑职业技术学院 环境市政与工程系,兰州 730050
- 折叠
摘要
针对滚动轴承剩余使用寿命(Remaining Useful Life,RUL)预测中开始预测时间(Start Prediction Time,SPT)的主观选择会导致预测信号噪声过大的问题,提出了一种基于线性回归的SPT点确定方法.采用滑动窗口线性回归方法,在均方根(Root Mean Square,RMS)范围内建立滑动窗口,得到RMS梯度域.设置RMS梯度的阈值,并使用连续触发阈值机制来确定SPT点.实验结果表明,基于线性回归的滚动轴承寿命预测起点的确定方法可以自适应地确定SPT点,提高了生命预测的精度.
Abstract
The subjective selection of Start Prediction Time(SPT)in rolling bearing Remaining Useful Life(RUL)Prediction will lead to the problem of excessive noise in Prediction signal,a method of SPT point determination based on linear regression is proposed.The sliding window is established in the Root Mean Square(RMS)range by using the sliding window linear regression method,and the RMS gradient domain is obtained.Thresholds for RMS gradients are set and a continuous triggering threshold mechanism is used to determine SPT points.The experimental results show that the SPT point determination method based on linear regression can adaptively determine SPT point and improve the precision of life prediction.
关键词
滚动轴承/剩余寿命预测(RUL)/开始预测时间(SPT)/线性回归Key words
rolling bearings/Remaining Useful Life(RUL)/Start Prediction Time(SPT)/linear regression引用本文复制引用
基金项目
甘肃省自然科学基金项目(22JR5RA814)
出版年
2024
NETL
NSTL
万方数据