基于小波神经网络的电力电子电路故障模式识别

提出了基于两种不同小波神经网络的电力电子电路故障模式识别方法。针对电力电子电路故障,构造了激活函数型和权值型两种不同的三层小波神经网络,给出了相应的数学模型和学习算法。以三相整流桥电路为例,建立了小波神经网络的输出与故障元之间的对应关系,实现了电路故障的模式识别,并与用普通BP网络识别的结果进行了比较。仿真结果验证了两种故障识别方法的正确性和较好的准确性。

Fault pattern recognition of power electronic circuits based on wavelet neural network

A fault pattern recognition method of power electronic circuits based on two wavelet neural networks is presented. Two different three-layer wavelet neural networks are constructed for fault pattern recognition of power electronic circuits, and the mathematical model and learning arithmetic are given. Taking three-phase rectifier circuit as an example, the relationship between the output of wavelet neural network and fault component is constructed, the fault recognition of circuits achieved, and the recognition results by two networks compared with that by general BP network. The validity and veracity of two methods are verified by simulation.