摘要
为了提高感应电机系统参数估计与状态监测的准确性和效率,针对感应电机非线性、强耦合、参数易时变的特性,引入带外部输入的非线性自回归(NARX)神经网络时序预测模型。针对传统NARX神经网络初值依赖和收敛速度慢的问题,利用天牛须搜索算法(BAS)对神经网络预测模型进行参数寻优,提高神经网络的收敛速度和预测精度。实验结果表明,该方法能够以较简单的网络结构高效、准确、稳定地预测估计电机参数。
In order to improve the accuracy and efficiency of parameter estimation and condition monitoring of the induction motor system,a time series prediction model based on nonlinear autoregressive with exogenous input(NARX)is introduced in consideration of the characteristics of nonlinearity,strong coupling and time⁃varying parameters of the induction motor.To solve the problem of initial value dependence and slow convergence speed of traditional NARX neural network,the beetle antennae search(BAS)algorithm is adopted to optimize the parameters of the neural network prediction model and improve convergence rate and prediction accuracy of the neural network.The results of experiments show that the proposed method can predict and estimate the motor parameters efficiently,accurately and steadily with a simple network structure.
作者
金逸珲
吴滨
顾晓峰
张晓昕
JIN Yihui;WU Bin;GU Xiaofeng;ZHANG Xiaoxin(Engineering Research Center of IoT Technology Applications(Ministry of Education),Wuxi 214122,China;Department of Electronic Engineering,Jiangnan University,Wuxi 214122,China)
出处
《现代电子技术》
北大核心
2020年第3期112-115,120,共5页
Modern Electronics Technique
基金
江苏省研究生科研与实践创新计划项目（KYCX18_1856）
中央高校基本科研业务费专项资金（JUSRP51510）
