贝叶斯框架支持向量机的模拟电路故障诊断(英文)

Fault Diagnosis for Analog Circuits Using SVM Within Bayesian Framework

基于贝叶斯证据框架下的最小二乘小波支持向量机,设计了一种新型模拟电路故障诊断方法。将贝叶斯证据框架应用于多类LS-WSVM分类器来选取正规化参数和核参数,并采用小波提升变换对从测试点得到的各种故障状态下输出电压信号进行分解获取多尺度的小波系数,对经处理的小波系数提取出故障特征量,以此作为样本训练多类LS-WSVM分类器来确定模拟电路故障诊断的模型。采用雷达系统模拟电路进行了仿真,结果表明,设计的模拟电路的故障诊断方法效果良好。

Based on least squares wavelet support vector machines （LS-WSVM） within the Bayesian evidence framework, a systematic method for fault diagnosis of analog circuits was proposed. The Bayesian evidence framework was applied to select the optimal values of the regularization and kernel parameters of multi-class LS-WSVM classifiers. Also output voltage signals under faulty conditions were obtained from analog circuits test points. Then wavelet coefficients of output voltage signals were gained by wavelet lifting decomposition, and faulty feature vectors were extracted from the coefficients. The faulty feature vectors were used to train the multi-class LS-WSVM classifiers, so the model of the circuit fault diagnosis system was built. The simulation result of scout radar shows that the fault diagnosis method of the analog circuits using LS-WSVM within the Bayesian evidence framework is effective.