论文部分内容阅读
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。
Principal component analysis has the characteristics of data compression and feature extraction, while neural network has the advantages of non-linear mapping and learning reasoning. Combining the two, a fault diagnosis method of analog circuit based on principal component analysis and neural network is proposed. Through the principal component analysis of the step response characteristic parameters of the analog circuit, the main parameters are extracted, and then the neural network is used to classify and decide the eigenvectors of various states to realize the fault diagnosis of analog circuits. The simulation results of the standard circuit show that this method can detect and locate the fault quickly and has the characteristics of high accuracy.