基于GRA-PLS的滚动轴承故障诊断

Research on Rolling Bearing Fault Diagnosis Based on Grey Relational Analysis and Partial Least Squares

针对仅用时域和频域指标无法准确诊断滚动轴承故障的问题,提出一种基于灰色关联度(GRA)与偏最小二乘(PLS)的故障诊断算法。首先,对原始振动信号进行灰色关联度分析,提取关联度较高的振动信号作为样本信号;其次,通过时域分析和频域分析获得故障特征集,利用基于遗传算法(GA)和Elman神经网络的组合算法(GA-ENN)对故障特征进行提取;最后,利用PLS算法对滚动轴承的故障类别进行识别。实验结果表明,所提方法能有效剔除原始振动信号中无信息变量,并且实现时、频域指标下滚动轴承故障的准确诊断。

Aiming at the problem that only time domain and frequency domain index cannot accurately diagnose rolling bearing faults,this paper proposes a fault diagnosis algorithm based on partial least squares(PLS)and gray correlation analysis methods.Firstly,the gray correlation analysis(GRA)was used to get the sample vibration signal with higher correlation degree.Then the time domain analysis and frequency domain analysis was used to obtain the fault characteristics.Secondly,the fault characteristics by the combination algorithm based on genetic algorithm(GA)and Elman neural network(GA-ENN).Finally,the PLS algorithm was used to identify the fault category of the rolling bearing.The experimental results show that the method proposed in this paper can effectively eliminate the uninformed variables in the original vibration signal and realize the accurate diagnosis of rolling fault under time and frequency domain index.