摘要
针对玻璃生产线退火窑辊道轴承振动信号存在强噪声污染、故障诊断准确率低、效率差等问题,提出一种基于灰关联熵分析和敏感特征评估的辊道轴承故障智能诊断方法。首先,将原故障信号用经验模态分解(Empirical Mode Decomposition,EMD)为多个本征模式分量(Intrinsic Mode Function,IMF),采用灰关联熵分析法筛选IMF分量并进行小波阈值降噪,重构故障信号。其次,选择时域和频域特征,定义基于故障特征类间、类内距离的敏感特征评估因子,筛选出敏感特征集。最后,使用径向基函数(Radial Basis Function,RBF)神经网络对故障特征集进行识别。滚动轴承故障测试实验结果表明,该方法能够有效提升故障轴承振动信号的信噪比,并评估筛选出敏感特征,从而实现对滚动轴承的智能诊断。
The roller bearing vibration signal of the annealing furnace in the glass production line is disturbed by the strong noise,which will lead to low accuracy and poor efficiency for fault diagnosis.In this paper,an intelligent fault diagnosis method for roller bearings is proposed based on grey relational entropy analysis and sensitive feature evaluation.First of all,the original fault signal is decomposed into several intrinsic mode function(IMF)components by using empirical mode decomposition(EMD).Then,the gray relation entropy analysis method is used to select the IMF components and perform wavelet threshold denoising.Afterwards,the fault signal is reconstructed.Then,several features of time domain and frequency domain are selected,and the sensitive feature evaluation factors is defined based on the distance between the fault feature classes and the distance within the fault feature classes to select out the sensitive feature set.Finally,the radial basis function(RBF)neural network is used to identify the fault feature set.The experimental results of the rolling bearing failure test demonstrate that the proposed method can effectively improve the signal-to-noise ratio of the vibration signals of the failed bearing,and evaluate the sensitive features to realize intelligent diagnosis of the rolling bearings.
作者
周康渠
阚志群
辛玉
吴雪明
张朝武
ZHOU Kangqu;KAN Zhiqun;XIN Yu;WU Xueming;ZHANG Chaowu(College of Mechanical Engineering,Chongqing University of Technology,Chongqing 400054,China;ChongqingWansheng Float Glass Co.,Ltd.,Chongqing 400800,China)
出处
《噪声与振动控制》
CSCD
北大核心
2021年第5期147-154,160,共9页
Noise and Vibration Control
基金
重庆市智能制造领域技术创新与应用示范专项重点示范资助项目(cstc2018jxzx-cyzdx0169)。
关键词
故障诊断
滚动轴承
灰关联熵分析
特征评估
RBF神经网络
fault diagnosis
rolling bearing
grey correlation entropy analysis
feature assessment
RBF neural network