基于DCCA-IWO-MKSVM的模拟电路故障诊断方法

为了提高模拟电路故障的诊断效果,提出基于DCCA-IWO-MKSVM的模拟电路故障诊断方法。采用DCCA算法对模拟电路的故障特征进行提取,构造新的融合特征。对支持向量机的核函数进行线性组合构造新的多核函数,并用IWO算法对其参数进行优化,以构建最优故障诊断模型,用于融合特征的学习分类。故障诊断实验结果表明:对于融合特征的故障诊断效率,该算法要优于单核函数的IWO-SVM算法,且整个故障诊断系统的诊断效果具有较高的准确率。

ANALOG CIRCUIT FAULT DIAGNOSIS METHOD BASED ON DCCA-IWO-MKSVM

In order to improve the diagnosis effect of analog circuit faults,this paper proposes a new method based on DCCA-IWO-MKSVM for fault diagnosis of analog circuits.The DCCA algorithm was used to extract the fault features of the analog circuit,and it constructed new fusion features.We constructed a new multi-core function by linear combination of kernel functions of SVM,and optimized its parameters with IWO algorithm to construct an optimal fault diagnosis model for learning classification of fusion features.Results of fault diagnosis experiment show that the algorithm is superior to the IWO-SVM algorithm of single kernel function for the efficiency of fault diagnosis of feature fusion rate,and the diagnosis effect of the whole fault diagnosis system has a high accuracy.