MATLAB code for RL‑SFO‑based parameter extraction of single‑, double‑ and triple‑diode PV models

This software package offers a complete MATLAB implementation of the reinforcement-learning-enhanced Starfish Optimization (RL-SFO) framework developed in this study for extracting parameters of photovoltaic (PV) models. It includes twelve optimization algorithms (RL-SFO and eleven advanced metaheuristics) designed for single-, double-, and triple-diode PV models, along with shared model functions, objective functions, and scripts for loading experimental IV datasets measured in Suez, conducting multi-run optimization experiments, and generating convergence plots, IV/PV curves, and error metric tables reported in the associated manuscript. The main driver scripts are organized by PV model (SDM, DDM, TDM) and by optimization algorithm, allowing users to reproduce all numerical results and easily adapt the framework to additional PV datasets or optimization methods. The code was developed by Nader M. A. Ibrahim and Bassam A. Hemade (2026) and is archived on Zenodo under DOI: 10.5281/zenodo.18881250. This software accompanies a manuscript currently under review at Scientific Reports (Springer Nature); the citation will be updated once the article is published.