关键词: 光伏系统, 最大功率跟踪, 差分策略, 樽海鞘群算法, 帕累托分布函数, 混沌映射

Abstract: Partial shading condition causes multiple peaks in power-voltage （P-V） curve of photovoltaic systems. And conventional maximum power point tracking （MPPT） algorithms are prone to be trapped at a local maximum power point. An improved salp swarm algorithm is proposed by expanding the zoom factor and introducing a difference strategy. The Pareto distribution and chaotic maps are added to improve the global search performance in the leader position update process; the difference strategy is introduced to improve the local search ability in the local search process. The improved method is applied to the MPPT of multi-peak photovoltaic system. The experimental results demonstrate that the improved salp swarm algorithm （ISSA） can achieve the fastest and most stable global MPPT under Partial shading in comparison to the conventional incremental conductance method （INC）, the perturbation observation method （P&O）, and the salp swarm algorithm （SSA）. It can be seen that this method has strong adaptability and stability in photovoltaic MPPT.

Key words: photovoltaic systems maximum power point tracking, difference strategy, salp swarm algorithm, Pareto distribution function, chaotic maps