A hybrid metaheuristic framework for robust photovoltaic model parameter estimation based on crested porcupine and differential evolution algorithms.

Accurate parameter estimation is essential for the efficient design and performance evaluation of solar photovoltaic (PV) systems. However, the strong nonlinearity of PV equivalent circuit models often causes conventional optimization algorithms to become trapped in local optima. To overcome this challenge, this paper proposes a novel hybrid optimization framework, termed Crested Porcupine Differential Evolution (CPO-DE), which combines the global exploration capability of the Crested Porcupine Optimizer with the robust mutation mechanism of Differential Evolution. The proposed CPO-DE algorithm is first validated mathematically using 23 benchmark functions alongside three classical engineering design problems. It is subsequently applied to the Single Diode Model (SDM), Double Diode Model (DDM), and Triple Diode Model (TDM) of PV systems. In addition, a novel two-stage evaluation strategy is introduced, where computationally efficient algebraic equations are employed during the iterative optimization process to reduce computational burden, followed by an explicit Newton-Raphson numerical solver to ensure high physical accuracy of the estimated parameters. The effectiveness of the proposed approach is evaluated using experimental data from the RTC France solar cell over 300 independent runs. The results demonstrate that CPO-DE achieves a 100% convergence success rate with near-zero standard deviations, indicating excellent robustness and stability. Moreover, the proposed method outperforms eleven state-of-the-art optimization algorithms and consistently attains the top rank according to Friedman statistical tests. These results confirm that CPO-DE is a highly reliable, stable, and efficient engineering tool for advanced photovoltaic model parameter estimation.