摘要
针对三相PWM整流器的d-q控制算法对网侧电流瞬时值进行高精度采样的要求,提出一种基于小波神经网络的网侧电流数字化检测算法。该算法结合了神经网络的函数逼近能力和小波变换的良好局部特性及多分辨率特性,使网络能根据数据的分布情况以不同的分辨率进行学习,从而使网络具有更灵活有效的函数逼近能力。仿真和实验结果表明,采用该方案进行网侧电流检测,结合d-q控制算法,使得三相PWM整流器系统稳压精度高,有较快的动态响应,网侧电流谐波畸变率小。
According to the high performance demand of grid-connected current,the paper presents a sampling scheme based on wavelet and neural network theory.The method composes the function approximation capability of the neural network and local and multi-resolution characteristics of wavelet transform.For this reason the network has more flexi-ble and efficient function approximation capability.The simulation and experiment verify the scheme can work well.Highly accurate steady output,fast response and less distorted waveform are gained.
出处
《电力电子技术》
CSCD
北大核心
2011年第4期61-63,共3页
Power Electronics
关键词
整流器
网侧电流采样
小波神经网络
rectifier
grid current sample
wavelet and neural network