摘要
针对利用领域覆盖算法(Neighborhood Covering Algorithm,NCA)解决模拟电路故障诊断过程中出现的故障诊断率不高的问题,文章采用一种改进的领域覆盖算法称之为点对主分量分析算法(Double Points Principal Component Analysis Algorithm,DPCAA)进行模拟电路故障诊断,首先通过对待诊断的模拟电路的可测点采用幅频特性技术进行故障特征提取,然后构建点对主分量分析算法的三层神经网络进行模拟电路故障诊断。为了验证该方法的可行性,本文最后对某一个带通滤波电路进行模拟电路故障诊断,对该电路的故障诊断率提高了2.22个百分点。
To solve the problem of covering algorithm of neural network about fault diagnosis in analog circuit fault diagnosis process,this paper uses double points principal component analysis algorithm （DPCAA） to solve fault diagnosis in analog circuit. Amplitude-Frequency characteristic is usedas a tool for extracting feature .Then, after training the double points principal component analysis algorithm , the model of the circuit with thefault diagnosis system is built .Simulation results show that the method is more effectiveandfauh diagnostic rate is increased by 2.22% .
出处
《电子设计工程》
2017年第7期126-129,共4页
Electronic Design Engineering
关键词
模拟电路
故障诊断
覆盖算法
点对主分量分析算法
fault diagnosis analog circuit
fault diagnosis
covering algorithm
double points principal component analysis algorithm
