论文部分内容阅读
利用电力电子器件和仿真功能软件对我国电气机车整流器常用的控制电路(单相串联3重联结电路)进行了故障分析诊断;在Matlab/Simulink环境下对单相串联3重联结电路进行了电路分析和仿真,并在其基础上,提出了基于BP神经网络故障分析的方法;基于BP神经网络,利用电路分析产生的故障数据,采用归一化(或标准化)进行数据预处理,并提取故障特征及构造故障样本集,利用Matlab对神经网络进行学习训练,然后对所提出的故障特征进行了分析和辨识,最后通过仿真实验证明了该方法的正确性和可行性。
Fault analysis and diagnosis of common control circuit of electric locomotive rectifier (single-phase series 3-junction circuit) is carried out by using power electronics and simulation software. The circuit analysis of single-phase series 3-junction circuit in Matlab / Simulink And based on which, a method of fault analysis based on BP neural network is put forward. Based on BP neural network, the fault data generated by circuit analysis is used to normalize (or normalize) the data preprocessing and extract fault features And constructing the fault sample set. The neural network is trained by using Matlab. Then the fault features are analyzed and identified. At last, the simulation results show that the method is correct and feasible.