基于ISSA-VMD与加权集合峭度的轴承故障诊断  

Bearing Fault Diagnosis Based on ISSA-VMD and Weighted Ensemble Kurtosis

作　　者：魏晓鹏 高丙朋[1] WEI Xiao-peng;GAO Bing-peng(School of Electrical Engineering,Xinjiang University,Urumqi 830017,China)

机构地区：[1]新疆大学电气工程学院,乌鲁木齐830017

出　　处：《组合机床与自动化加工技术》2022年第12期67-71,共5页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：新疆维吾尔自治区自然科学基金项目(2019D01C079)。

摘　　要：针对滚动轴承故障信号易被噪声所掩盖,故障特征不易提取及故障类型无法确定的问题,提出一种基于改进麻雀搜索算法(ISSA)优化变分模态分解(VMD)与加权集合峭度的故障诊断方法。首先,构建一种综合函数作为目标函数,通过ISSA优化VMD实现故障信号的自适应分解;其次,建立加权集合峭度选取最优分量(IMF)并重构;最后,使用改进阈值对重构信号进一步去噪,通过1.5维能量谱对去噪信号分析,准确判断轴承故障类型。仿真和实际工程数据验证了所提方法的有效性,结果表明与其他智能算法优化VMD和包络谱分析对比,所提方法效果更优。Aiming at the problem that rolling bearing fault signals are easily disturbed by noise,fault features are not easily extracted and fault types cannot be determined,a fault diagnosis method based on improved sparrow search algorithm(ISSA)optimized variational mode decomposition(VMD)and weighted ensemble kurtosis was proposed.Firstly,a synthetic function as the objective function was constructed,VMD was optimized using ISSA to realize the adaptive decomposition of fault signals;Then,IMF was selected by the weighted ensemble kurtosis,and signal reconstruction was performed;Finally,the improved threshold denoising was performed on the reconstructed signal,the denoised signal was analyzed using the 1.5-dimensional energy spectrum to realize fault feature extraction and achieve to judge the type of faults accurately.The analysis of simulated and actual engineering signals verifies the effectiveness of the proposed method.Compared with other intelligent algorithms optimized VMD and envelope spectral analysis,the results show that the proposed method is more effective.

关 键 词：改进麻雀搜索算法 变分模态分解 加权集合峭度 改进阈值去噪 故障诊断 

分 类 号：TH133.3[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]