基于粒子群优化的神经网络电机故障诊断系统

提取电机定予电流信号及转于振动信号,构成用于电机故障诊断网络的训练及测试样本.用BP神经网络建立诊断输入征兆与故障输出间的映射关系,引入改进粒子群优化的策略,对神经网络权值和阀值进行优化,提高了网络系统诊断的可靠性.仿真对比研究表明,经粒子群优化后的BP网络收敛速度显著提高,更适合于电机类故障诊断的要求.

Neural Network Fault Diagnosis System Based on Particle Swarm Optimization for Motor

The training and testing samples of neural network were made by extracting the signals of stator current and rotor oscillation,which was used to diagnose motor fault.The mapping relationship between input phenomenon and fault output was established by using BP neural network,and the improved particle swarm optimization was adopted to optimize the neural network weights and threshold,to improve the diagnosis reliability of system.The simulation results show that the convergence speed of BP network optimized by particle swarm increases significantly,more suitable for the fault diagnosis of motor.