Abstract
Inclusion of bypass diodes at the output terminal of the PV array mitigates the effect of partial shading (PS) but causes multiple peaks of power at the output. The conventional hill climbing and perturb and observe algorithms cannot track the optimal point during partial shading phenomena for multiple peaks corresponding to the different shading pattern on the Power-Voltage (P-V) curve. Fuzzy logic controller and artificial neural network-based methods for Maximum Power Point Tracking (MPPT) provide satisfactory results but at the cost of increased memory and computational burden. Recent work to incorporate exploration and exploitation phenomena of nature-inspired algorithms to track optimal power point have shown encouraging results by preventing convergence to local maxima and posing less burden on the processor. However, due to performance variation between different algorithms of this category newer algorithms with improved performances are still a requirement. In this paper, a novel most valuable player algorithm (MVPA) has been used to track the optimal operation point for extracting maximum power from a solar PV system. The algorithm's performance is compared with the commonly employed particle swarm optimization (PSO) and the recently proposed Jaya algorithm's modified form. It is observed that the proposed algorithm outperformed both the algorithms with a considerable improvement in terms of tracking speed, power tracking efficiency, robustness, faster decision for convergence after tracking the maximum power and lesser number of power fluctuations for different shading patterns.