基于模糊神经网络的机械轴承故障诊断方法研究

Research on Mechanical Bearing Fault Diagnosis Method Based on Fuzzy Neural Network

针对机械轴承智能化故障诊断的需求,提出了一种融合模糊逻辑和神经网络的故障诊断方法。利用EMD-AR谱提取机械故障振动信号特征,将提取的特征向量作为训练样本库和检验样本库,运用模糊神经网络实现故障诊断。最后设计机械轴承故障诊断专家系统,并通过轴承故障诊断实例,验证了智能诊断技术在机械故障诊断领域可以较好地满足诊断需求。

Aiming at the requirement of intelligent fault diagnosis of mechanical bearing,a fault diagnosis method based on fuzzy logic and neural network is proposed in this paper.The feature of mechanical fault vibration signal was extracted by using EMD-AR spectrum,and the extracted feature vector is used as training sample library and test sample library.Fuzzy neural network is used to realize fault diagnosis.Finally,the expert system of mechanical bearing fault diagnosis is designed,and the bearing fault diagnosis example is used to verify that the intelligent diagnosis technology could meet the requirements of mechanical fault diagnosis.