Performance assessment of a solar PV array using a new dynamic reconfiguration based on the Cuckoo Search optimizer

Partial shading is a major challenge in photovoltaic (PV) systems, as it causes significant power losses and leads to multiple local maximum power points (MPPs) on the power-voltage (P-V) curve, reducing overall system efficiency. To address this issue, we proposed a novel reconfiguration approach for shaded PV arrays using the Cuckoo Search (CS) optimization algorithm. With the proposed method, we aimed to identify an optimal switching matrix structure that minimizes current mismatch among rows and maximizes power output under shading conditions. The methodology involved implementing a 9 × 9 PV array model in MATLAB/Simulink and evaluating its performance under five distinct shading patterns. The performance of the CS-based reconfiguration was compared against four established techniques: Total Cross-Tied (TCT), standard Sudoku, Optimal Sudoku, and the Multi-objective Grey Wolf Optimizer (MOGWO). The evaluation was based on key statistical and performance metrics, including on power output, system reliability, and convergence behavior. Testing results demonstrated the superiority of the CS method, achieving global maximum power point (GMPP) values of 23.6071 kW, 23.0057 kW, 22.1083 kW, 22.6669 kW, and 22.4937 kW across the five tested shading scenarios. These results validated the effectiveness of the CS algorithm in enhancing the power output of PV arrays under partial shading and highlight its potential applicability in real-world energy systems.