【摘要】 主元分析具有数据压缩及特征提取的特性,而神经网络具有非线性映射和学习推理的优点。将二者结合起来,提出基于主元分析与神经网络的模拟电路故障诊断方法。通过对模拟电路的阶跃响应特征参数进行主元分析,提取主要参数,然后利用神经网络对各种状态下的特征向量进行分类决策,实现模拟电路的故障诊断。对标准电路仿真结果表明:该方法能够实现快速故障检测与定位,具有准确率高的特点。更多还原

【Abstract】 Combining the extracting feature vectors and compressing data characteristics of principal component analysis(PCA) with the nonlinear mapping and generalizing of neural networks(NN),a method of fault diagnosis in analog circuits is proposed.The step response feature parameters are preprocessed by PCA to generate major ones.Feature vectors under certain states can be classified using NN,and fault diagnosis is realized.Simulation results on benchmark circuits show that this scheme is feasible and has many powerful features,such as diagnosing and locating faults quickly and exactly.更多还原