粗糙集和主元分析在基于神经网络的模拟电路故障诊断中的应用

Fault diagnosis in analog circuits based on PCA、rough set and neural networks

针对基于神经网络的模拟电路故障诊断中,故障特征集维数过高带来的诊断难点,提出了利用粗糙集和主元分析法对故障特征集进行预处理。粗糙集对故障诊断决策表进行属性约简,主元分析进行数据压缩及特征提取。试验仿真表明,对预处理后的数据进行识别,简化了神经网络结构,可有效提高网络的训练速度与诊断效率。

Directed toward fault diagnosis in analog circuits based on neural networks, feature parameters make its fault diagnesstic difficult. In order to solve this problem, we present a method of neural network fault diagnosis is based on PCA, Rough Sets and Neural Networks.And Rough Sets theory is used to eliminate unnessary attributes from the decision table, Feature parameters of the step response are compressed using principal component analysis （PCA）. The simulation experiments shows that it has many good properties, such as simplifying the structure of NN, improving the training speed and fault coverage, which obviously quickens Iraining speed and decreases training time, and the application effect is notable.