基于改进ABC优化RVM的模拟电路故障诊断

Fault diagnosis of analog circuit based on improved artificial bee colony optimize relevance vector machine

为提高模拟电路的故障诊断精度,提出了基于改进蜂群算法(IABC)优化相关向量机(RVM)的模拟电路故障诊断方法。该方法首先基于小波包能量提取故障特征集,然后将故障特征输入RVM进行故障诊断,同时利用IABC算法进行RVM参数的优化,避免参数选择的盲目性,提高故障诊断的精度。通过对Sallen-Key带通滤波器电路的单故障和复合故障诊断结果表明,该方法是有效的,相比与其他一些方法,可以获得更高的诊断精度,具有一定的优势。

In order to improve the accuracy of fault diagnosis in analog circuit fault diagnosis,an analog circuit fault diagnosis method based on improved artificial bee colony( IABC) optimize relevance vector machine( RVM) was proposed. The fault feature extractedbased on wallet packet energywas inputted into RVM to identify different fault. IABC was used to optimize RVM parameter in order toavoid blindness of parameters selection and improv the diagnostic accuracy. The single fault and composite fault diagnosis results of Sallen-Key bandpass filter circuit show that this method is effective. Compared with other existing methods,it is able to achieve higher fault diagnosis rate.