基于IVMD算法的动车组滚动轴承故障特征提取方法研究
Study on the Rolling Bearing Fault Feature Extraction Method of Multiple Units Based on IVMD Algorithm
安国平1
作者信息
- 1. 国家铁路局 安全技术中心,北京 100160
- 折叠
摘要
动车组转向架滚动轴承的运行安全一直是影响列车安全运行的重要环节,目前国内外有多套监测体系进行了车载应用,但误报、漏报等现象时有发生,滚动轴承故障特征提取方法的准确性是该领域研究的重点之一.本文提出了一种基于改进的变分模态分解算法(Improved Variational Mode Decomposition,IVMD)的故障特征提取算法,采用混合蛙跳算法(Shuffled Frog Leaping Algorithm,SFLA)对模态数K和带宽控制参数α进行最优自适应选择.建立了基于包络熵、峭度和相关系数的多目标评价函数来选择最优模态分量.利用功效系数法将多目标优化问题转化为单目标优化问题,用频谱分析法对最优模态分量进行重构和处理.最后利用全实物电机轴承滚动实验台数据进行了方法的测试,有效验证了提出的改进方法分解故障信号及提取故障特征频率的准确性.
Abstract
The operation safety of the rolling bearings of bogies for multiple units is an important link influencing the safe operation of train.At present,there're multiple monitoring systems applied on board home and abroad,but false alarms,omissions and other faults happen from time to time.The accuracy of the rolling bearing fault feature extraction method is one of the keys of study in this field.This article proposes a fault feature extraction algorithm based on the Improved Variational Mode Decomposition(IVMD),using the Shuffled Frog Leaping Algorithm(SFLA)to make the optimal adaptive selection of the mode number K and the bandwidth control parameter α.A multi-objective evaluation function is established based on envelope entropy,kurtosis and correlation coefficient to select the optimal modal components.The efficiency coefficient method is used to transform a multi-objective optimization problem into a single-objective optimization problem,and the spectral analysis is made to reconstruct and process the optimal modal components.Finally,the method is tested using the data from a full physical motor bearing rolling test bench,effectively verifying the accuracy of the proposed improved method in decomposing fault signals and extracting fault feature frequencies.
关键词
动车组/滚动轴承/故障诊断/变分模态分解/混合蛙跳算法Key words
multiple units/rolling bearing/fault diagnosis/Variational Mode Decomposition(VMD)/Shuffled Frog Leaping Algorithm(SFLA)引用本文复制引用
出版年
2024
NETL
NSTL
维普
万方数据