Photovoltaic MPPT Control Strategy Based on Tuna Swarm Optimization TSO Algorithm

Abstract

In photovoltaic systems, multiple power peaks imitated by the partial shading led to trapping of conventional maximum power point tracking MPPT algorithms such as Perturb and Observe (P&O) at local maxima resulting in about 30–50% power loss. A hybrid Tuna Swarm Optimization-Perturb and Observe (TSO-P&O) algorithm is proposed in this paper where the location of global maximum power point (MPP) estimated by using TSO and local refinement provided through P&O based on a sudden change from environmental with an adaptive restart mechanism. TSO-P&O outperforms P&O, Particle Swarm Optimization (PSO), Genetic Algorithm (GA) and Grey Wolf Optimizer (GWO) validated by MATLAB/Simulink. It reaches a maximum static shading efficiency of 99.06% and dynamic shading convergence in as fast as 0.071s (98.52%) with the same equipment, respectively; It also displays a minimized steady-state oscillation (0.42% standard deviation) and superior statistical performance at p < 0.05 as well. The algorithm is very suitable for real-time embedded implementation, with a runtime of 2.5 ms per iteration and a memory footprint of only 5 KB. Experimental validation of the proposed algorithm against a hardware-prototype, extension to larger PV arrays experiencing more complex shading patterns, and hybridizations with learning-based methods could yield additional opportunities for enhanced performance in future work