论文部分内容阅读
为了改善标准的粒子群算法在模拟电路故障诊断中存在的不足,采用了自适应变异粒子群算法来优化BP神经网络的故障诊断方法。首先对待测电路的可测点的响应信号提取故障特征,并进行小波包分解和归一化从而构建样本集;然后利用粒子群改进算法来优化BP神经网络的权值和阈值,从而实现对待测电路的训练和测试。在针对某电路的故障诊断中发现了该方法的故障诊断时间和诊断率比改进之前有了明显的改善,并且在中心偏差范围为0.3时诊断率达到了99%。
In order to improve the standard PSO algorithm in the fault diagnosis of analog circuits, adaptive mutation particle swarm optimization (PSO) is used to optimize the BP neural network fault diagnosis method. Firstly, the fault signal is extracted from the response signal of the measurable point of the circuit to be measured, and the sample set is constructed by wavelet packet decomposition and normalization. Then the particle swarm optimization algorithm is used to optimize the weights and thresholds of the BP neural network, Circuit training and testing. In the fault diagnosis of a certain circuit, it was found that the diagnostic time and the diagnostic rate of this method have been significantly improved before the improvement, and the diagnostic rate reached 99% when the center deviation range was 0.3.