摘要
为提高光伏并网发电的发电效率,提出了基于COOT优化算法和改进型重复PI控制的三相LCL型光伏并网系统。COOT优化算法用于前级MPPT算法,可以准确快速追踪到光伏发电系统全局最大输出功率;改进型重复PI控制用于后级逆变,通过阻尼因子设计LCL有源阻尼控制,在数学建模过程中把采样计算延时以及PI控制考虑进去,最后加入重复控制;为了减少奇次谐波,采用级联式重复PI控制结构;同时针对电网频率波动时重复控制策略控制效果降低的问题,采用一种基于拉格朗日插值原理的频率自适应级联式重复PI控制。基于所提出的控制策略,利用Simulink仿真平台对系统进行了建模仿真,仿真结果表明了所提控制策略的正确性和可行性。
In order to improve the efficiency of grid-connected photovoltaic power generation,this paper proposes a three-phase LCL-type grid-connected photovoltaic system based on COOT optimization algorithm and improved repetitive PI control.The improved repetitive PI control is used for the back-stage inverter,and the LCL active damping control is designed through the damping factor,taking the sampling calculation delay and PI control into account in the mathematical modeling process,and finally adding the repetitive control.In order to reduce the odd harmonics,a cascaded repetitive PI control structure is used.At the same time,to address the problem that the control effect of the repetitive control strategy decreases when the grid frequency fluctuates,a method based on the frequency adaptive cascade repetitive PI control based on the Lagrangian interpolation principle is used.Based on the proposed control strategy,the Simulink simulation platform is used to model and simulate the system,and the simulation results show the correctness and feasibility of the proposed control strategy.
作者
杨春辉
屈莉莉
廖慧
戴宏跃
YANG Chunhui;QU Lili;LIAO Hui;DAI Hongyue(School of Mechatronic Engineering and Automation,Foshan University,Foshan 528225,China;Guangdong Chuangdian Technology Co.,Ltd.,Foshan 528000,China;School of Intelligent Manufacturing,Guangzhou Vocational College of Technology&Business,Guangzhou 511442,China)
出处
《佛山科学技术学院学报(自然科学版)》
CAS
2024年第1期33-45,共13页
Journal of Foshan University(Natural Science Edition)
基金
广东省高校高端电源系统技术开发中心项目(2019GGCZX002)。
关键词
光伏并网系统
COOT优化算法
有源阻尼控制
级联式重复PI控制
拉格朗日插值原理
photovoltaic grid-connected systems
COOT optimization algorithms
active damping control
cascaded repetitive PI control
Lagrangian interpolation principle