基于二次型单神经元PID的MPPT控制  

MPPT control based on quadratic single neuron PID

作　　者：高金辉[1] 刘永[1] 朱元培[1] 

机构地区：[1]河南师范大学物理与信息工程学院,河南新乡453007

出　　处：《电源技术》2015年第5期1068-1069,1121,共3页Chinese Journal of Power Sources

基　　金：河南省重点科技攻关项目(102102310041)

摘　　要：传统型PID控制是光伏最大功率跟踪系统(maximum power point tracking,MPPT)最常用的控制方法,但该方法存在许多的不足,如自适应能力较弱、调节时间长、动态性能较差等缺点。结合光伏发电系统的特点,针对目前控制方法的不足,提出一种新的最大功率点跟踪优化算法,即使用二次型单神经元与模糊控制理论相结合的方法进行控制。使用Matlab/Simulink软件建立通用型光伏矩阵模型,并使用二次型单神经元PID进行控制,模拟光照强度、环境温度等参数变化时的光伏阵列工作状态。仿真和实验结果表明,基于二次型单神经元的控制算法能够迅速达到稳定状态,并且在最大功率附近震动较小,控制效果良好,效率有所提高。The traditional PID control is the most common method for photovoltaic maximum power tracking system, however, this approach has many deficiencies, such as weak adaptive capacity, long adjust time, poor dynamic performance and so on. Combined with the characteristics of photovoltaic power generation system, a new maximum power tracking method was proposed in view of the deficiencies of current control method, using quadratic single neuron with fuzzy control theory for control. Matlab/Simulink simulation software was used to create a general-purpose PV matrix model, and quadratic single neuron PID was used to control. The photovoltaic array working state was simulated with changing light intensity, temperature and other parameters. The results show that the control algorithm using the quadratic single neuron can rapidly reach a steady state. It＇s stable near the maximum power point and works well.

关 键 词：二次性能指标 神经网络 PID MPPT 

分 类 号：TM914[电气工程—电力电子与电力传动]