一种滚动轴承故障诊断方法  

A method of fault diagnosis of rolling bearing

作　　者：梁志成 王芳[1] 徐皞昊 LIANG Zhicheng;WANG Fang;XU Haohao(School of Electrical Engineering, Shanghai Dianji University, Shanghai 200240, China)

机构地区：[1]上海电机学院电气学院,上海200240

出　　处：《上海电机学院学报》2020年第2期99-105,共7页Journal of Shanghai Dianji University

摘　　要：滚动轴承作为旋转设备的重要组成部分,在各领域有着广泛的应用。在风力发电机组中,滚动轴承作为较易损坏的部件之一,其故障会导致风力发电机组整机工作异常。为了及时发现滚动轴承的缺陷和损伤,提出了一种基于小波和改进粒子群优化径向基函数(PSO-RBF)神经网络的滚动轴承故障诊断方法。对原始振动信号进行小波降噪、小波包分解,然后进行相空间重构,提取特征向量。利用特征向量作为输入,相对应的轴承状态作为输出,进行改进PSO-RBF神经网络的训练,得到神经网络故障诊断模型。对于待诊断的滚动轴承信号,对其进行处理后,输入到神经网络故障诊断模型中,即可得到滚动轴承的诊断结果。As an important part of a rotating equipment,the rolling bearing has been widely used in various fields.In wind turbines,the rolling bearing is one of the more vulnerable parts,and the failure will cause the wind turbines to work abnormally.In order to find out the defects and damage of the rolling bearing in time,the present paper proposes a fault diagnosis method for the rolling bearing based on the wavelet and improved particle swarm optimization-radial basis function(PSO-RBF)neural network.The original vibration signal is denoised by the wavelet and decomposed by the wavelet packet.The phase space is reconstructed,and the eigenvectors are extracted.With the eigenvectors as the inputs and the corresponding bearing state as the output,the training of the improved PSO-RBF neural network is carried out,and the neural network fault diagnosis model is obtained.For the signals of the rolling bearing to be diagnosed,the present paper can obtain the diagnosis results of the rolling bearing by processing the signals and inputting them into the fault diagnosis model of the neural network.

关 键 词：滚动轴承 故障诊断 风力发电机组 小波包分解 神经网络 粒子群算法 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]