Adaptive Hybrid PO and Drone Squadron Optimization MPPT Tecnique for Solar PV Systems Under Diverse Conditions

Abstract

Recent studies on solar photovoltaic (PV) system aim to make the system more efficient and generally, the performance of the PV arrays. A key aim is to reduce energy losses that arise due to variable weather patterns and irregular load demands. Traditional approaches include Perturb and Observe (PO) which are easy and work well when there is a stable setting, often fail under partial shading conditions (PSC). This paper proposes a hybrid MPPT approach that integrates Perturb and Observe (PO) with Drone Squadron Optimization (DSO) to improve the performance of solar PV systems under varying environmental conditions. The proposed hybrid PO–DSO method combines the fast response of PO with the global search capability of DSO, allowing the system to avoid local optima and accurately track the Global Maximum Power Point (GMPP). A comprehensive simulation study was conducted in different scenarios as constant irradiance, temperature variation, and partial shading. At constant irradiance case, the proposed algorithm achieved a PV power output of 249.5 W and delivered 243.68 W to the load. Under temperature variation, it maintained a high output of 249.3 W,
demonstrating thermal stability. During PSC, it successfully tracked the GMPP with PV output of 128.25 W & 128.3 W and a load power of 124.31 W & 124.2 W. Comparative analysis with particle swarm optimization (PSO), cuckoo search algorithm (CS) and flower pollination algorithm (FPA) showed that the proposed method achieved the highest tracking efficiency of up to 99.8 %, with faster convergence and fewer oscillations. These results highlight the hybrid method’s robustness, making it a promising MPPT solution for real-world PV systems operating in unpredictable and dynamic weather conditions. Lastly, the viability and validation of the use of the approach and its applicability in practice were tested through real-time simulation on the OPAL-RT platform.