A novel chaotic differential variation snake optimization algorithm for parameters estimation of photovoltaic cell and module models

With the growing demand for solar energy, accurate, stable, and efficient parameters estimation of critical photovoltaic (PV) cell and module models has become highly important and challenging since it is crucial for multiple aspects of PV systems, such as simulation, design, control, optimization, evalua-tion, efficiency calculations, and maximum power point tracking. To address this issue, more effective attempts and algorithms with better performance are highly desired. This paper presents a novel chaotic differential variation snake optimization (CDVSO) algorithm to efficiently and effectively estimate PV model parameters (PVMPs). In CDVSO, firstly, an improved circle chaotic mapping (ICCM) strategy is proposed to improve the diversity and quality of population in the initialization phase. Secondly, an improved differential variation operator is used to perform the differential variation updating with three randomly selected solutions in the new solution generated phase of each generation, thus enhancing the global searching capability and quitting from local optima more easily. Thirdly, another improved differential variation operator is proposed to generate the new better variation solution with the current best solution and current solution in each generation, achieving a better balance between exploration and exploitation during the optimization process. To evaluate the effectiveness and efficiency of proposed CDVSO, parameters estimation on five PV models is conducted, and the experimental results show that the CDVSO with competitive computation efficiency exhibits the superiority including accuracy, convergence speed, and stability, among the compared algorithms. In conclusion, the proposed CDVSO can provide an effective solution for PV model parameters estimation and possess wide potential applications. • CDVSO algorithm is proposed for photovoltaic model parameters estimation. • Improved Circle chaotic mapping is proposed to improve the initial population diversity and quality. • Differential operator 1 improves the global search capability. • Differential operator 2 balances global exploration and local exploitation. • Experimental results on five photovoltaic models demonstrate the excellent performance of CDVSO.