A robust hybrid Marine Predator-Puma Optimizer for enhanced parameter estimation of photovoltaic models

Accurate parameter estimation of photovoltaic (PV) models is critical for simulation, performance prediction, and fault diagnosis in PV systems. However, the strong nonlinearity of the current–voltage characteristics makes this a challenging optimization problem. To address this challenge, this paper presents a hybrid optimization algorithm called the Marine Predator-Puma Optimizer (MPAPO) for accurate and efficient parameter estimation of photovoltaic (PV) models. The parameter extraction problem is formulated as a nonlinear optimization task that minimizes the root mean square error (RMSE) between the measured and simulated current–voltage characteristics of the PV model. The MPAPO combines the global exploration ability of the Marine Predator Algorithm with the adaptive hunting-based exploitation of the Puma Optimizer to maintain a balanced and efficient search process. The algorithm is applied to single-, double-, and triple-diode PV models using experimental data from the RTC France solar cell and Photowatt-PWP201 module. It achieves the best RMSE of 7.937 × 1 0 − 4 , 9.1939 × 1 0 − 4 , and 7.8602 × 1 0 − 4 with the single-, double-, and triple-diode models, respectively. Comparative analysis shows that MPAPO achieves higher accuracy, faster convergence, and greater numerical stability than recent optimization algorithms. The results confirm that the proposed hybrid approach provides a reliable and computationally efficient framework for modeling the nonlinear behavior of photovoltaic systems.