基于故障特征系数的IGWO-VME方法及其在轮对轴承故障诊断中的应用  

IGWO-VME Method Based on Fault Feature Coefficient and Its Application in Wheelset Bearing Fault Diagnosis

作　　者：李翠省 刘永强[2,3] 廖英英 王俊锋[4] 刘文朋 LI Cuixing;LIU Yongqiang;LIAO Yingying;WANG Junfeng;LIU Wenpeng(College of Automotive Engineering,Hebei Vocational University of Industry and Technology,Shijiazhuang 050091,China;State Key Laboratory of Mechanical Behavior and System Safety of Traffic Engineering Structures,Shijiazhuang Tiedao University,Shijiazhuang 050043,China;School of Mechanical Engineering,Shijiazhuang Tiedao University,Shijiazhuang 050043,China;CRRC Tangshan Co.,Ltd.,Tangshan 063000,China)

机构地区：[1]河北工业职业技术大学汽车工程学院,河北石家庄050091 [2]石家庄铁道大学省部共建交通工程结构力学行为与系统安全国家重点实验室,河北石家庄050043 [3]石家庄铁道大学机械工程学院,河北石家庄050043 [4]中车唐山机车车辆有限公司,河北唐山063000

出　　处：《铁道学报》2024年第12期49-57,共9页Journal of the China Railway Society

基　　金：国家自然科学基金(12032017);河北省自然科学基金(A2024210004);石家庄科技局科研计划(241791307A)。

摘　　要：针对加速度传感器采集的信号中包含多种干扰成分,导致轴承早期故障特征难以准确获取问题,提出一种基于自适应变分模态提取(VME)的故障诊断方法。VME算法是一种从多分量信号中提取单一分量的有效技术方法,提取结果与惩罚因子和初始中心频率密切相关。为自适应获取这些关键参数,利用改进灰狼优化器(IGWO)优化VME。考虑故障特征系数(FFC)具有明确的指向性,可以准确评估信号中特定故障的含量,因此建立适应度函数引导IGWO进行全局寻优。通过一个数值模拟信号和两个轮对轴承实验信号的分析,验证IGWO-VME方法的有效性。与自适应变分模态分解(AVMD)和快速谱峭度图(FK)的对比分析表明,IGWO-VME方法具有更强的故障特征提取能力,是一种更为先进的轮对轴承故障诊断方法。Aiming at the difficulty to accurately obtain the early fault features of the bearing caused by multiple interference components in the signals collected by acceleration sensors,a fault diagnosis method was proposed based on adaptive variational mode extraction(VME).The VME algorithm is an effective technique to extract a single component from a multi-component signal,withthe extraction result being closely related to the penalty factor and the initial center frequency.To obtain these key parameters adaptively,the improved grey wolf optimizer(IGWO)was used to optimize the VME.As the fault feature coefficient(FFC),with clear directivity,can accurately evaluate the content of a specific fault in the signal,the corresponding fitness function was established to guide IGWO for global optimization.The effectiveness of the IGWO-VME method was verified by analyzing a numerical simulation signal and two wheelset bearing experimental signals.The comparison with adaptive variational mode decomposition(AVMD)and fast kurtogram(FK)shows that the IGWO-VME method,with stronger fault feature extraction ability,is a more advanced scheme for wheelset bearing fault diagnosis.

关 键 词：变分模态提取 惩罚因子 初始中心频率 改进灰狼优化器 故障特征系数 

分 类 号：U260.3312[机械工程—车辆工程] TH17[交通运输工程—载运工具运用工程]