Photovoltaic Array Multi-Peak MPPT Control Strategy Based on Fused NSSA-PSO and Enhanced PVS-PO Algorithm

In complex environments, photovoltaic (PV) arrays frequently exhibit multiple power peaks in their output characteristics, a phenomenon often attributed to partial shading. Traditional Maximum Power Point Tracking (MPPT) algorithms tend to converge to the Local Maximum Power Point (LMPP), whereas more advanced intelligent algorithms can identify the Global Maximum Power Point (GMPP) but suffer from slow tracking speeds and substantial power oscillations. To mitigate these limitations, we propose a novel fusion algorithm that integrates the Natural Selection Strategy Adaptive Particle Swarm Optimization (NSSA-PSO) with the Partition Variable Step Perturb and Observe (PVS-PO) method for MPPT. This algorithm dynamically adjusts the inertia weight and learning factors throughout the optimization process and employs a natural selection strategy within each iteration. These enhancements facilitate faster convergence and minimize power oscillations during the tracking phase, thereby expediting the approach to the GMPP. Subsequently, the algorithm employs the PVS-PO method for the precise determination of the GMPP. The efficacy of this algorithm is demonstrated through simulations using a PV model under partial shading conditions, confirming its practical utility.