Multijunction solar cell parameter estimation based on metaheuristic algorithms

• The Newton-Raphson Method (NRM) hybrid with meta-heuristics for better starting points closer to the optimal solution. • The hybrid method leads to high accuracy and few iterations reaching convergence. • The hybrid method improves the robustness of the approach, introducing randomness and escaping local optima, thereby reducing the risk of NRM getting stuck in suboptimal regions. • Parameters estimation using meta-heuristics optimization algorithms based on modified model for CPV. • Optimization of conventional five, seven, and nine parameters identified for the SDM, DDM, and TDM and three additional parameters representing breakdown voltage. • The study compared the performance of different optimization algorithms: GWO, AHA, TSA, WSA, and MFO. • Extensive comparison of the five MAs based on RMSE, Absolute Error (AE), Mean Absolute Error (MAE), Relative Error (RE), and convergence time. Parameter estimation in MJSC achieved accurate energy output modeling under diverse environmental conditions. This issue maximizes energy yield and facilitates early fault detection by identifying performance deviations, enabling timely maintenance, and reliable energy production. MJSCs are highly nonlinear and require precise parameter selection. Several mathematical models exist for MJSC parameter estimation— Single Diode Model (SDM), Double Diode Model (DDM), and Triple Diode Model (TDM)—each offering a trade-off between complexity and accuracy, improved efficiency, and reliable operation. This paper evaluates three models—SDM, DDM, and TDM—focusing on MJSC parameter estimation. The models include five parameters for SDM, seven for DDM, and nine for TDM, with three additional terms for breakdown voltage alongside conventional parameters. In addition, The Newton-Raphson Method (NRM) hybrid with meta-heuristic algorithms, or MOAs, is used to improve solution estimates, accelerate convergence rates, reduce computational costs, and enhance robustness. The study compared the performance of five optimization algorithms: Grey Wolf Optimizer (GWO), Artificial Hummingbird Algorithm (AHA), Tunicate Swarm Algorithm (TSA), War Strategy Algorithm (WSA), and Moth flam optimizer (MFO). The results demonstrate the superior performance of the AHA algorithm, which has the lowest RMSE value (0.0582) in the SDM and DDM cases, while WSA has the lowest RMSE value (0.0488) in the TDM instance. The MFO algorithm outperforms the GWO algorithm in SDM and DDM scenarios, while the reverse is true for TDM. In terms of errors, and execution time, the TSA algorithm performs the poorest in all three circumstances with RMSE value greater than 0.16.