文章摘要
高维士*,严运兵*,马强**,朱博文**,王晓东**.基于RBFNN分段在线优化的VSR无源控制[J].高技术通讯(中文),2020,30(11):1178~1188
基于RBFNN分段在线优化的VSR无源控制
  
DOI:10.3772/j.issn.1002-0470.2020.11.010
中文关键词: 径向基函数神经网络(RBFNN); 粒子群优化算法(PSO); 无源控制(PBC); LCL型滤波器; 整流器
英文关键词: radial basis function neural network(RBFNN), particle swarm optimization (PSO), passivity-based control (PBC), LCL type filter, rectifier
基金项目:
作者单位
高维士*  
严运兵*  
马强**  
朱博文**  
王晓东**  
摘要点击次数: 1984
全文下载次数: 1436
中文摘要:
      针对传统L型滤波器电压型脉冲宽度调制(PWM)整流器存在的入网电流波形总畸变率过高、稳定性差及控制精度低等问题,提出了基于径向基函数神经网络(RBFNN)分段在线优化的LCL型滤波器电压型PWM整流器无源控制策略,设计了LCL滤波电压型PWM整流器的内环无源控制器,和基于RBFNN的外环PID控制器。用粒子群优化算法(PSO)对初始注入阻尼及不同负载下的RBFNN学习率、动量因子及饱和函数的饱和值等参数进行离线优化,以负载电阻值作为RBFNN分段优化触发条件,根据负载变化使用PSO离线优化值对RBF-PID参数进行分段在线优化,实现最优动态调整。
英文摘要:
      Aiming at the shortcomings of the traditional L-filtered voltage-type pulse width medulation (PWM) rectifier with high total distortion rate, poor stability and low control precision, a voltage source PWM rectifier (VSR) passive control strategy based on radial basis function neural network (RBFNN) segmentation online optimization LCL filtering is proposed, the inner loop passive controller of LCL voltage type PWM rectifier and the outer loop PID controller based on RBFNN are designed. The particle swarm optimization algorithm (PSO) is used to optimize the initial injection damping, RBFNN learning rate, momentum factor and saturation function saturation value under different loads, and the load resistance value is used as the RBFNN segmentation optimization trigger condition. Using the PSO offline optimization value under different loads performs segmental online optimization of the RBF-PID parameters to achieve optimal dynamic adjustment.
查看全文   查看/发表评论  下载PDF阅读器
关闭

分享按钮