基于输出频谱和支持向量机的模拟电路故障诊断方法

摘要

针对模拟电路的特点和其故障诊断中存在的问题,提出了一种基于输出频谱和支持向量机的新型模拟电路故障诊断方法（FMSVM）。该方法采用多音信号作为模拟电路的激励信号,并用其输出频谱中的有限个频率点的幅值作为故障特征,采用多分类支持向量机（MSVM）进行故障模式判别,实现了模拟电路的故障诊断。实验结果表明,该故障诊断方法具有速度快、准确率高的特点,具有重要的实用价值。%On the basis of the characteristics of analog circuits and the problems of analog circuits＇ fault diagnosis,a new fault diagnosis method is proposed based on the Output Frequency and Multiple Support Vector Machine（FMSVM）.Multitone signal is used to stimulate analog circuits in this method,and the amplitude of limited frequency points is used as fault tags.Analog circuits＇ fault diagnosis is realized by using Multiple Support Vector Machine（MSVM） to identify different pattern of faults.This method characterized by rapid calculation and high accuracy is proved to have important practical value by simulation.