跳到主要导航 跳到搜索 跳到主要内容

Predictive current regulator using linear neural networks for three-phase voltage source PWM-inverter

  • GD Chen*
  • , YZ Shen
  • , MH Huang
  • *此作品的通讯作者

科研成果: 书/报告/会议事项章节 › 会议稿件 › 同行评审

2 引用 (Scopus)

摘要

This paper presents a predictive control strategy based on linear neural networks for the current control of a three-phase voltage source PWM-inverter. The base of this regulator is a conventional deadbeat control loop which ensures the relatively good dynamic performance of system. However, because of the inherent characteristic of deadbeat control, existing time delay between system reference input and actual output, therefore causes the steady state error when the input is sinusoidal. To eliminate this error, a novel predictive control method is presented in this paper. Based on the deadbeat controller, two sinusoidal predictors implemented by linear neural networks are further introduced for the predictions of reference current and the AC side voltage respectively. As a result, the time delay and steady state error are minimized to almost zero. Simulation results are presented to verify the effectiveness of the proposed algorithm.
源语言英语
主期刊名PROCEEDINGS OF THE 2005 INTERNATIONAL CONFERENCE ON NEURAL NETWORKS AND BRAIN, VOLS 1-3
出版商IEEE; IEEE Circuits and Systems Society (CAS); IEEE Communications Society (ComSoc); IEEE Computer Society; IEEE Signal Processing Society
页793-798
页数6
ISBN(印刷版)0-7803-9422-4
出版状态已出版 - 13 10月 2005

丛书

姓名Predictive current regulator using linear neural networks for three-phase voltage source PWM-inverter

关键词

  • neural networks
  • predictive control
  • current regulator
  • PWM-inverter

收录

  • CPCI-S

学术指纹

探究 'Predictive current regulator using linear neural networks for three-phase voltage source PWM-inverter' 的科研主题。它们共同构成独一无二的学术指纹。

引用此