基于灵敏度特性函数的特征提取与故障诊断

为解决电路故障诊断时故障可靠分类以及特征信息有效提取的问题,提出了一种基于灵敏度特性的故障样本分类和故障特征信息提取方法。基本思想是通过电路的特性分析和灵敏度的计算,进行故障样本的分类及优化,再根据灵敏度的计算结果提取相应特征信息。以此构造故障样本特征集,然后作为BP神经网络的输入对网络进行训练与诊断。对滤波器的仿真结果表明,该方法构造的样本集训练出来的神经网络,对模拟电路故障诊断的平均正确率为85%,优于传统方法。

Feature extraction and fault diagnosis based on sensitivity characteristic function

In the circuit fault diagnosis, in order to solve the fault reliable classification and feature information effective extraction, this paper presents a method for extracting fault classification and fault information based on the characteristic of sensitivity. The basic idea is through the circuit characterization and sensitivity calculations, fault classification and optimization of the sample, according to the sensitivity of lhe results to extract feature information accordingly,in order to construct fault samples feature set, and then as BP neural network input the network training and diagnosis. On the filter simulation results show that this method constructs a sample set trained neural network,the analog circuit fault diagnosis,the average accuracy was 85% ,better than the traditional methods.