量子遗传算法在变压器故障诊断模型中的应用  

Application of quantum genetic algorithm in transformer fault diagnosis model

作　　者：龚瑞昆 周国庆 GONG Ruikun;ZHOU Guoqing(North China University of Science and Technology,Tangshan 063000,China)

机构地区：[1]华北理工大学,河北唐山063000

出　　处：《现代电子技术》2018年第15期129-132,共4页Modern Electronics Technique

基　　金：国家自然科学基金项目(61271402)~~

摘　　要：传统的BP神经网络诊断模型容易陷入局部最优,且诊断正确率较低,因此,提出将量子遗传算法应用于RBF网络诊断模型。首先确定RBF神经网络的输入输出、建立RBF网络模型,然后把归一化后的数据送入RBF网络模型,利用量子遗传算法对RBF神经网络进行优化,得到最优诊断模型,最后输出诊断结果。用Matlab进行仿真,其结果表明该算法解决了系统容易陷入局部最优的问题,在训练48代后就快速获得最优解,加快了网络的收敛速度。同时RBF神经网络的泛化能力也得到很好的改善,故障诊断正确率达93%,远远高于传统神经网络模型。The traditional BP neural network diagnosis model is easy to fall into local optimum,and has low diagnostic accuracy.Therefore,the quantum genetic algorithm is applied to RBF network diagnosis model,with which the input and output of RBF neural network are determined to establish the RBF diagnosis model,the normalized data is sent to the RBF diagnosis model.The quantum genetic algorithm is used to optimize the RBF neural network to obtain the optimal diagnostic model and output the diagnosis result.The Matlab simulation results show that the algorithm can solve the problem that the system is easy to fall into local optimum,obtain the optimal solution after training 48 generations,and speed up the network convergence rate.The generalization ability of RBF neural network is improved greatly,the fault diagnosis accuracy can reach up to 93%,which is much higher than that of the tradition neural network model.

关 键 词：变压器 故障诊断 BP神经网络 量子计算 RBF神经网络 量子遗传算法 

分 类 号：TN98-34[电子电信—信息与通信工程] TP391.9[自动化与计算机技术—计算机应用技术]