首页|资源下载
登录|注册
热门搜索：fpga51单片机protel99se机器人linux 单片机 dsp armProteusmatlab

您现在的位置是：电子研发网 > 资源下载 > 基于多测试点多特征信息的模拟电路故障诊断

基于多测试点多特征信息的模拟电路故障诊断

资 源 简 介

为了提高模拟电路故障诊断准确性，要采集尽可能多的故障样本信息作为BP神经网络的输入；提出了一种基于多测试点多特征信息构造原始样本集的方法；运用这种方法构造原始故障特征集，然后作为BP神经网络的输入对网络进行训练，能提高模拟电路故障诊断正确率；对Sallen--Key二阶带通滤波器的仿真结果表明，该方法构造的样本集训练出来的网络对模拟电路故障的识别正确率为85％，优于传统方法，为模拟电路的故障诊断提供了一种可行的方法。

In order to improve the diagnostic accuracy of analog circuit fault, to capture as much fault samples as the input of BP neural network, put forward a new method to construct the original sample set based on multi--points and raulti--feature information. Using this method to construct the feature set of the original failure, and then as the input of BP neural network to train the network, can improve the correct rate of analog circuit fault diagnosis. Sallen--Key second order band pass filter simulation results show that, the network training with sample set constructed by this method used in fault diagnosis of analog circuits is better accuracy than traditional methods, fault recogni- tion correct rate is 85%, provides a feasibl...

每月领取下载积分

相 关 资 源

电子研发网版权所有 京ICP备2021023401号-2