基于模糊神经网络的故障诊断及其在无刷直流电机系统中的应用

Fault Diagnosis Based on Fuzzy Neural Network and its Application in Brushless DC Motor System

本文针对故障诊断知识具有模糊性的特点，提出了一种基于模糊神经网络的故障诊断方法，将模糊理论与神经网络融合在一起，实现了对故障的模糊诊断。文中给出了模糊神经网络的结构和学习方法，并提出了一种阈值向量故障判别方法，使故障判别更具灵活性。并以无刷直流电机系统为对象，将该方法应用于系统的故障诊断，仿真结果表明，该方法是行之有效的。

In view of the fuzzy property of the fault diagnosis (FD) knowledge,a FD approach based on a fuzzy neural network (FNN) is presented in this paper which combines fuzzy theory with neural network to implement the fault fuzzy diagnosis. The structure of the FNN is given,and the learning method is presented. Meanwhile a fault discriminant method for threshold vector is employed,which makes fault discrimination even more flexible.Moreover,it has been applied to the FD of a brushless DC motor system. The simulation shows that it is useful and effective.