摘要
为了提高异步电机故障诊断的准确性,引入了一种基于模拟退火粒子群算法优化BP神经网络(SAPSO-BP算法)的故障诊断方法.根据电机转子振动频谱中所提取的特征参数与故障类型之间的关系数据,利用模拟退火粒子群算法来优化BP神经网络的权、阈值参数,再由优化好的BP网络进行故障诊断.实验结果表明,该方法具有较好的故障模式的识别效果,明显提高了异步电机故障诊断的准确性.
In order to improve the accuracy of the asynchronous motor fault diagnosis, introduces a fault diagnosis method based on simulated annealing particle swarm optimization BP neural network (SAPSO-BP algorithm). According to the relationship data extracted between the motor rotor vibration spectrum characteristic parameters and fault types, using simulated annealing particle swarm algorithm to optimize BP neural network rights, the threshold parameter, and then by the optimized BP network troubleshooting. The experimental results show that this method has better failure modes identification, and significantly improve the accuracy of asynchronous motor fault diagnosis.
出处
《红河学院学报》
2014年第2期36-38,共3页
Journal of Honghe University
关键词
模拟退火
粒子群算法
神经网络
异步电机
故障诊断
simulated annealing
particle swarm optimization algorithm
neural network
asynchronous motor
fault diagnosis