粒子群和PIDNN控制器在VSC-HVDC中的应用.pdfVIP

第 31 卷 第 3 期 中 国 电 机 工 程 学 报 Vol.31 No.3 Jan.25, 2011 8 2011 年 1 月 25 日 Proceedings of the CSEE ©2011 Chin.Soc.for Elec.Eng. 文章编号：0258-8013 (2011) 03-0008-06 中图分类号：TM 721 文献标志码：A 学科分类号：470·40 粒子群与 PIDNN 控制器在 VSC-HVDC 中的应用 王国强,王志新 (上海交通大学电子信息与电气工程学院，上海市 闵行区 200240) Application of PSO and PIDNN Controller for VSC-HVDC WANG Guoqiang, WANG Zhixin (School of Electronic, Information and Electrical Engineering, Shanghai Jiaotong University, Minhang District, Shanghai 200240, China) ABSTRACT: The double closed-loop PI regulator is widely 摘要：海上风电场并网柔性直流输电系统中，双闭环 PI 调 used for the control of wind farm side and grid side converters 节器常用来控制风场侧和电网侧变流器,该方法较为成熟, in the VSC-HVDC system for offshore wind farms. But there 但存在采用的调节器过多、且参数整定困难等不足。文中以 are too many PI regulators, so it is difficult to adjust all the 神经网络中间层至输出层的权值作为粒子群寻优参数,采用 parameters. In this paper we designed PID neural network 粒子群(particle swarm optimization，PSO)算法设计 PID 神经 controller combined with particle swarm optimization (PSO). 网络(PID neural network，PIDNN)控制器，并将该控制器用 And it was used to control the converters of VSC-HVDC. First 于控制海上风电场柔性直流输电变流器。根据 PIDNN 的结 simple improvement was carried out based on PIDNN structure 构特点，经过简单改进，即将输入层至中间层的权值设定为 characteristic. We set the linking weights of the first two layers 定值，这时粒子群只需优化中间层至输出层权值，能够明显 as constant values. Only the interlayer and output layer linking 减少粒子维数，并提高训练速度。用训练获得的 PIDNN 控 weights of PID neural network (PIDNN) were used as the 制器代替传统 PI 调节器，建立变流器控制系统的传递函数， optical parameters of PSO. It re