基于蝙蝠算法优化VMD参数的滚动轴承复合故障分离方法  

A compound fault feature separation method of rolling bearings based on VMD optimized by the bat algorithm

作　　者：张伟 李军霞[1,2] 陈维望 ZHANG Wei;LI Junxia;CHEN Weiwang(College of Mechanical and Vehicle Engineering,Taiyuan University of Technology,Taiyuan 030024,China;National-Joint Engineering Laboratory of Mining Fluid Control,Taiyuan University of Technology,Taiyuan 030024,China)

机构地区：[1]太原理工大学机械与运载工程学院,太原030024 [2]太原理工大学矿山流体控制国家地方联合工程实验室,太原030024

出　　处：《振动与冲击》2022年第20期133-141,共9页Journal of Vibration and Shock

基　　金：中央引导地方科技发展资金项目(YDZJSX2021A023);国家自然科学基金(52174147);晋中市科技重点研发计划(Y211017);山西省重点研发计划项目(201903D121062)。

摘　　要：针对强背景噪声干扰下的滚动轴承复合故障特征难以提取且变分模态分解(variational modal decomposition,VMD)中参数需要提前确定的问题,提出一种基于蝙蝠算法优化VMD参数的滚动轴承复合故障分离方法。首先,提出一种新的复合影响指数,将其与现有指标进行比较,结果表明,所提指标对故障信号的敏感性提高了29.6%。然后,将最小平均复合影响指数作为目标函数,利用蝙蝠算法自适应搜索VMD的最优参数进行变分模态分解。最后,对分解后的模态分量进行包络解调分析,通过包络谱判断轴承的故障类型。仿真及试验结果表明,该方法能够从噪声干扰下的复合故障信号中有效分离出单一故障信息,实现轴承故障类型的确定,从而验证了该方法的有效性。Aiming at the problems that the composite fault feature information of a rolling bearing under the interference of strong background noise is difficult to extract,and the parameters in the variational modal decomposition(VMD)need to be determined in advance,a composite fault separation method for a rolling bearing based on VMD optimized by the bat algorithm was proposed.First,a new comprehensive impact index was proposed and compared with existing indicators.The results show that it has increased by 29.6%to the sensitivity of the fault signal.Then,the minimum average comprehensive influence index was proposed as the objective function,the bat algorithm was used to adaptively search the optimal parameters of VMD,and the signal was decomposed by VMD.Finally,the decomposed modal components were subjected to envelope demodulation analysis,and the fault type of the bearing was judged through the envelope spectrum.The simulation and experimental results show that the method can effectively separate the information of a single fault from the composite fault signal under noise interference,and determine the type of bearing fault,thus the effectiveness of the method is verified.

关 键 词：蝙蝠算法 变分模态分解(VMD) 复合影响指数 复合故障 

分 类 号：TH212[机械工程—机械制造及自动化] TH213.3