摘要
针对模拟电路渐变性故障中的特征提取困难和故障信号无法进行有效分类的问题,提出利用免疫遗传算法(immune genetic algorithm,IGA)优化反向传播(back propagation,BP)神经网络中参数寻优过程,从而实现模拟电路故障诊断。首先,采用小波包分析(wavelet package analysis,WPA),对模拟电路输出响应进行4层小波分解和重构,完成特征向量的提取。然后,采用IGA优化BP神经网络进行训练及测试,实现对不同故障进行故障诊断。最后,通过两个模拟电路仿真实验对该方法进行实验验证。实验结果表明,与优化前的BP神经网络相比,所提方法提高故障诊断的准确率约15%。
In the view of the difficulty in feature extraction and failure signal classification in analog circuit with gradual change,an immune genetic algorithm(IGA)is proposed to optimize the parameter optimization process in back propagation(BP)neural network,so as to realize analog circuit fault diagnosis.Firstly,the wavelet package analysis(WPA)is used to decompose and reconstruct the output response of analog circuit in four layers,and the feature vector is extracted.Then,the IGA optimized BP neural network is used for training and testing to realize fault diagnosis of different faults.Finally,the two simulation methods are verified by simulation.The experimental results show that,compared with the BP neural network before optimization,the proposed method improves the accuracy of fault diagnosis by about 15%.
作者
王力
刘子奇
WANG Li;LIU Ziqi(Vocational Technical Institute,Civil Aviation University of China,Tianjin 300300,China;College of Electronic Information and Automation,Civil Aviation University of China,Tianjin 300300,China)
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2021年第4期1133-1143,共11页
Systems Engineering and Electronics
基金
国家自然科学基金(U1733119)资助课题。
关键词
反向传播神经网络
免疫遗传算法
模拟电路
特征提取
故障诊断
back propagation(BP)neural network
immune genetic algorithm(IGA)
analog circuit
feature extraction
fault diagnosis