Accurate photovoltaic (PV) cell modeling is critical to the analysis, diagnostics, and design of solar energy systems. At its center is the extraction of unknown parameters of the single-diode or double-diode models, which are nonlinear in nature and have high sensitivity to initial guesses. In this paper, we propose an efficient and powerful optimization method using Simulated Annealing (SA) for PV parameter extraction at standard test conditions. SA, a probabilistic metaheuristic inspired by the annealing process in metallurgy, is employed in minimizing the deviation of the experimental current–voltage (I-V) data from the model-generated curves.
The algorithm is tested on benchmark PV modules with real I-V characteristics, and its performance is verified in terms of Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and R² score. A comparison with other optimization techniques such as Particle Swarm Optimization (PSO) and Genetic Algorithms (GAs) confirms the effectiveness and reliability of the SA-based approach. The results show that the SA algorithm leads to fast convergence, avoids local minima, and provides highly accurate parameter estimates with very good agreement with the experimental data.
This research demonstrates the potential of SA optimization as a flexible and reliable approach to PV modeling, especially for researchers and engineers who need precision and strength in solar energy system simulations.