摘要
提出了基于优化BP神经网络的电机故障识别模型.针对以往隐节点数依靠经验选取,缺乏理论指导的问题,利用隐节点输出的相关系数对网络隐层优化,简化模型结构.实验结果表明,该优化方法可行、所建立的故障识别模型有效.
A fault recognition model of motor is proposed based on optimized BP neural network to solve the problem that the number of hidden nodes is decided by experience and lacking of theoretical guidance.The hidden layer optimization is implemented by adopting correlation coefficients of hidden layer output.The test results show that the optimization is feasible,and the fault recognition model proposed is effective.
出处
《大连交通大学学报》
CAS
2014年第2期89-92,共4页
Journal of Dalian Jiaotong University
基金
国家863计划资助项目(2012AA040912)
辽宁省教育厅高等学校科研计划资助项目(L2011077
L2012159)
关键词
电机
故障识别
模型优化
BP网络
motor
fault recognition
model optimization
BP neural network
