关键词: 局部阴影条件, 光伏阵列, 最大功率点跟踪, 差分进化法, 自适应梯度下降法

Abstract: Under the partial shading condition, the PV curve of the PV array will show the characteristics of multiple power peak points, while the traditional methods for PV array can only track the single power peak point. As it is fail to use these traditional methods to track the maximum power point in all of the multiple peak points under the partial shading condition, the existing research works are mainly based on particle swarm optimization algorithm and differential evolution algorithm. These algorithms have good global optimization ability, but the convergence stagnation problem easily arises in their local search process, therefore an improved algorithm based on adaptive gradient descent and differential evolution algorithm is proposed, which uses the adaptive gradient descent method to perform local search optimization at the later stage searching of differential evolution algorithm. The simulation results show that the proposed algorithm can find the maximum power point accurately and solve the convergence stagnation problem effectively. Compared with particle swarm optimization algorithm and difference evolution algorithm, the average total optimization time of the proposed algorithm is reduced by 52.67 % and 40.05 % respectively, and the convergence rate is further improved.

Key words: Partial shading condition, photovoltaic array; maximum power point tracking, difference evolution, adaptive gradient descent