通过分区变步长和过采样技术提高MPPT算法的控制性能

Improving the Control Performance of the MPPT Algorithm with the Partition Variable Step Size and Oversampling Technique

扰动观察法由于具有控制成本低且易于实现的特点,因此被普遍应用于实际光伏系统MPPT控制。光照条件迅速变化时,扰动观察法可能会导致误判,增大功率损失,降低算法的跟踪速度,为此提出了一种基于过采样技术的分区变步长的MPPT算法,过采样技术是通过增加一次电压、电流采样,利用平滑算法来计算下一光照条件的功率,以此解决光照突变时出现的的误判问题;分区变步长是基于曲线分区,根据功率变化率的大小,将整个电压区域分为3个部分,不同的部分采用不同的占空比进行扰动,并且将二级步长与变步长相结合,以解决扰动观察法存在的跟踪精度和速度的矛盾问题,仿真结果验证了算法的有效性。

Perturb and observe （P＆O） method was widely used because of its low-cost and ease of implementation. When irradiance is changing rapidly, this method fails to track MPP with rapid speed and may lead to misjudgment and makes power come down. In this paper a partition variable step MPPT algorithm based on oversampling technique was proposed, and a voltage and current sampling was added and the power of the next light condition was calculated with smoothing algorithm, which could solve misjudgment when irradiance changed rapidly. The partition variable step size was based on curve partition. According to the size of the power rate, the voltage region was divided into three parts adopting different duty ratio in different regions. This new method could also improve the precision and tracking speed of system＇s state by combining two step-sizes with variable step size. Simulation results are presented to demonstrate the effectiveness and feasibility of the proposed algorithm.