摘要
电动机轴承故障产生的冲击性信号具有非平稳、非线性的特点,且极易被随机噪声信号干扰,导致轴承故障特征信号的提取极具挑战性。为此提出一种基于连续变分模态分解(SVMD)和波形因子筛选准则的轴承故障信号提取方法。首先,考虑到传感器的零点漂移问题,对原始振动信号去趋势;其次,运用平滑噪声稳健差分器进行降噪;然后,运用SVMD算法分解降噪后的信号,分解过程无需事先知道信号模态数目,避免了模态数难以确定的问题;接下来,使用波形因子准则筛选出合适的模态分量描述轴承运动特征信号;最后,运用包络谱分析提取故障特征频率。仿真和实际轴承故障数据实验结果表明,与变分模态分解类方法相比,所提方法避免了事先预估模态数量,能够准确提取电动机轴承故障的特征信号。
The impact signal from motor bearing fault is non-stationary and nonlinear,and it is prone to be contaminated by random noise signal.Therefore,it is a challenge for extracting the features of the bearing fault signal.To adress this issue,an extraction method of the bearing fault feature signal based on successive variational mode decomposition(SVMD)and waveform factor screening criteria is proposed.Firstly,the original vibration signal is de-trended to relieve the zero drift of the sensor.Secondly,the smooth noise-robust differentiator is utilized to suppress noise.Then,the signal after noise reduction is decomposed by SVMD without the number of signal modes,which avoids the difficult problem of determining the number of modes.Next,the waveform factor criterion is employed to select the appropriate modal components for representing the bearing motion feature signal.Finally,the fault feature frequency is extracted by using envelope spectrum analysis.The experimental results of simulation and measured bearing fault data demonstrate that,compared to variational modal decomposition methods,the proposed method can effectively extract the feature signal of motor bearing fault under different working conditions without requiring any information about the number of signal modes.
作者
解春维
余美仪
Xie Chunwei;Yu Meiyi(Guangzhou Electromechanical Technician College,Guangzhou 510435,China;Foshan University,Foshan,Guangdong 528225,China)
出处
《机电工程技术》
2024年第1期238-242,共5页
Mechanical & Electrical Engineering Technology
基金
广东省职业技术教育学会科研规划项目(202103Z152)。
关键词
连续变分模态分解
轴承故障检测
波形因子
故障诊断
successive variational mode decomposition
bearing fault detection
waveform factor
fault diagnosis
