基于WOA-VMD和快速谱峭度的轴承故障诊断  

Bearing fault diagnosis based on WOA-VMD and fast spectral kurtosis

作　　者：杨栋 黄民[1] 马超[1] YANG Dong;HUANG Min;MA Chao(Mechanical Electrical Engineering School,Beijing Information Science&Technology University,Beijing 100192,China)

机构地区：[1]北京信息科技大学机电工程学院,北京100192

出　　处：《北京信息科技大学学报（自然科学版）》2023年第2期16-22,共7页Journal of Beijing Information Science and Technology University

基　　金：国家重点研发计划课题(2020YFB1713203)。

摘　　要：针对轴承故障特征提取技术不能充分准确地提取特征信息、需手动设置分解算法参数导致故障识别精度不高、普适性差的问题,提出了基于鲸鱼算法(whale optimization algorithm, WOA)优化变分模态分解(variational mode decomposition, VMD)和快速谱峭度的轴承故障诊断方法。首先,通过快速谱峭度对原始信号计算分析,选择包含故障频率的滤波范围,进行带通滤波,减少原始振动信号的噪声影响;其次,利用WOA对VMD的分解模态数k与惩罚因子α进行寻优,选取信号分解最优参数,并对滤波后信号进行分解,获得本征模态分量(intrinsic mode functions, IMF),通过时域特征指标选取包含故障信息最丰富的IMF;最后,对最优IMF包络谱解调分析,得到准确的故障特征频率。实验结果表明,提出的故障诊断方法能够有效地识别滚动轴承内圈故障。This research focuses on the problems that the bearing fault feature extraction technology cannot fully and accurately extract the feature information,and the parameters of the decomposition algorithm need to be manually set,resulting in low fault identification accuracy and poor universality.A bearing fault diagnosis method based on whale optimization algorithm(WOA)-variational mode decomposition(VMD)and fast spectral kurtosis was proposed.Firstly,the filtering range containing the fault frequency was selected through the calculation and analysis of the original signal by fast spectral kurtosis,and the band-pass filtering was carried out to reduce the noise impact of the original vibration signal.Secondly,the number of decomposition modes k and the penalty factorαwere optimized;the optimal parameters for signal decomposition were selected;the filtered signal was decomposed to obtain the intrinsic mode functions(IMF),and the most abundant fault information was selected by the time domain characteristic index.Finally,the optimal IMF envelope spectrum was demodulated and analyzed,and the accurate fault characteristic frequency was obtained.The experimental results show that the proposed fault diagnosis method can effectively diagnose the inner ring faults of rolling bearings.

关 键 词：轴承故障 快速谱峭度 鲸鱼优化算法 变分模态分解 

分 类 号：TH17[机械工程—机械制造及自动化] TH133.33