基于模糊RBF神经网络的滚动轴承故障诊断  

Fault Diagnosis of Rolling Bearing Based on Fuzzy RBF Artificial Neural Network

作　　者：孙旺旺[1] 任传胜[1] 朱春伟[1] 

机构地区：[1]中国科学技术大学工程科学学院,安徽合肥230026

出　　处：《机械研究与应用》2013年第2期13-14,17,共3页Mechanical Research & Application

摘　　要：针对现有滚动轴承故障诊断方法存在人为因素影响、表达模糊信息能力弱等问题。结合模糊评价和RBF神经网络的优点,选取3层小波包分解方法以获取评价因子,并使用正态分布型隶属度函数,构建了模糊RBF神经网络滚动轴承故障诊断模型。网络测试结果表明,该模型客观准确,诊断结果与实际情况一致。To solve the problems in fault diagnosis methods of rolling bearing such as the impact of human factors, and poor a-bility of expressing fuzzy information, by the advantages of the fuzzy recognition and the RBF artificial neural network, in this paper the three-layer wavelet packet decomposition method is chosen to get the evaluation factors, using the normal distribu-tion membership function , a fuzzy RBF artificial neural network model is established for the fault diagnosis of rolling bearing. The test results of the network show that the model is objective and accurate, the diagnosis results is identical with the actual state completely.

关 键 词：模糊RBF神经网络 小波包分解 滚动轴承 故障诊断 

分 类 号：TH133.33[机械工程—机械制造及自动化] TP183[自动化与计算机技术—控制理论与控制工程]