Enhanced MPPT Using Hybrid Smell Agent and Particle Swarm Optimization under Partial Shading in PV Arrays

Abstract

In this study, comparative evaluation of Maximum Power Point Tracking (MPPT) performance in photovoltaic (PV) systems was conducted using three meta-heuristic approaches: Smell Agent Optimization (SAO), Particle Swarm Optimization (PSO), and a hybrid PSO–SAO algorithm. The analysis is performed under four irradiance scenarios—uniform, partial, low, and severe shading—to capture a broad range of practical operating conditions. Simulation results show that SAO demonstrates strong global search capability, enabling effective avoidance of local optima, but it suffers from slower convergence and noticeable oscillations, particularly in partial and high-irradiance cases. PSO achieves faster convergence and efficient local exploitation, though it exhibits slight instability under severe shading. The hybrid PSO–SAO algorithm successfully combines the strengths of both methods: SAO’s broad exploration ensures thorough coverage of the search space, while PSO’s rapid convergence accelerates fine-tuning toward the optimal solution. Convergence plots show that the hybrid method consistently reaches the global maximum power point (GMPP) within a few iterations across all irradiance patterns, yielding the highest power outputs and maintaining minimal post-convergence oscillations. Under severe shading, the hybrid approach achieves a peak output of 70.7074 W—surpassing standalone PSO (70.5985 W) and SAO (70.5984 W). These findings confirm that the proposed hybrid method provides a robust, precise, and efficient MPPT strategy, particularly suited to PV systems operating in challenging and dynamically changing shading conditions.