摘要
为提高基于支持向量机(SVM)模拟电路故障诊断的准确率和优化效率,在灰狼优化(GWO)算法的基础上,通过引入非线性收敛因子、动态权重和边界变异策略,提出了一种改进灰狼优化(IGWO)算法优化SVM参数(IGWO-SVM)的改进型分类器.首先,在Sallen-Key带通滤波器和四运放双二次高通滤波器电路中采集故障样本,并对故障样本进行小波包特征提取;然后,将特征提取后的样本划分为训练样本和测试样本,利用IGWO算法来优化SVM中的惩罚参数C和核参数γ,得到最优的SVM分类器模型;最后,与其他算法优化的SVM分类器进行对比,结果表明IGWO-SVM分类器可以防止种群陷入局部最优,同时收敛速度有了较大提升.
In order to improve the accuracy and optimization efficiency of analog circuit fault diagnosis based on support vector machine(SVM),on the basis of gray wolf optimization(GWO)algorithm,this paper proposes a modified classifier that uses the improved gray wolf optimization(IGWO)algorithm to optimize the parameter of SVM(IGSA-SVM)by introducing the nonlinear convergence factor,dynamic weight and boundary variation strategy.Firstly,the fault samples are collected in the Sallen-Key bandpass filter circuit and four opamp biquad highpass filter circuit,and wavelet packet feature extraction is applied to fault samples.Then,feature-extracted samples are divided into training samples and test samples.The IGWO algorithm is used to optimize the penalty parameter C and the kernel parametersγin SVM to obtain the optimal SVM classifier model.Finally,compared with SVM classifiers optimized by other algorithms,the results show that the IGWO-SVM classifier can prevent the population from falling into a local optimum,and the convergence speed has been greatly improved.
引文
[1]毛先柏.基于支持向量机的模拟电路故障诊断研究[D].武汉:华中科技大学,2009.
[2]刘美容,张立玮.基于小波分解和模糊聚类的模拟电路软故障诊断[J].微电子学与计算机,2014,31(12):140-143.
[3] Yuan L,He Y,Huang J,et al.A new neural-network-based fault diagnosis approach for analog circuits by using kurtosis and entropy as a preprocessor[J].IEEE Transactions on Instrumentation&Measurement,2010,59(3):586-595.
[4] Long B,Xian W,Li M,et al.Improved diagnostics for the incipient faults in analog circuits using LSSVM based on PSO algorithm with mahalanobis distance[J].Neurocomputing,2014,133(10):237-248.
[5] Chen P,Yuan L,He Y,et al.An improved SVM classifier based on double chains quantum genetic algorithm and its application in analogue circuit diagnosis[J].Neurocomputing,2016(211):202-211.
[6]梅恒荣,殷礼胜,刘冬梅,等.改进粒子群算法优化的SVM模拟电路故障诊断[J].电子测量与仪器学报,2017,31(8):1239-1246.
[7] Mirjalili S,Mirjalili S M,Lewis A.Grey wolf optimizer[J].Advances in Engineering Software,2014,69(3):46-61.
[8]郭振洲,刘然,拱长青,等.基于灰狼算法的改进研究[J].计算机应用研究,2017,33(12):3603-3606.
[9]邓乃扬,田英杰.支持向量机:理论、算法与拓展[M].北京:科学出版社,2009.
[10]姜媛媛,韩振云,崔江.基于DAGSVC的模拟电路故障字典法[J].传感器与微系统,2011,30(4):12-16.