Adolescent identity search algorithm for parameter extraction in photovoltaic solar cells and modules

• Published in: Journal of computational electronics Vol. 21; no. 4; pp. 859 - 881
• Main Authors: Lekouaghet, Badis, Khelifa, Mohammed Amin, Boukabou, Abdelkrim
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.08.2022; Springer Nature B.V
• Subjects: Alternative energy; Design optimization; Efficiency; Electrical Engineering; Energy resources; Engineering; Exploitation; Foraging behavior; Heuristic methods; Mathematical and Computational Engineering; Mathematical and Computational Physics; Mathematical models; Mechanical Engineering; Mutation; Optical and Electronic Materials; Optimization algorithms; Parameter estimation; Parameter identification; Photovoltaic cells; Search algorithms; Solar cells; Theoretical; Parameter estimation; Metaheuristics; Adolescent identity search algorithm; Photovoltaic models; Optimization
• ISSN: 1569-8025; 1572-8137
• DOI: 10.1007/s10825-022-01881-1

Analysis and modeling of photovoltaic (PV) solar cells and modules based on experimentally measured data are critical for optimizing their design. The need for new algorithms to optimize the PV parameters, many of which owe their inspiration to the metaheuristic search concepts, is still a principal subject of interest and discussion. In this paper, an optimization algorithm that simulates the identity formation behavior of adolescents in the peer group, namely the adolescent identity search algorithm (AISA), was applied to identify the unknown parameters of PV models. In AISA, the updating process proceeds in the exploitation and exploration stages as follows. First, the new best position is generated by identifying and imitating the best identity features of a selected peer from the group to accelerate the exploitation process and produce better performance using a dynamic selection strategy. Second, any locally optimal solution is avoided in the exploration stage for the global optimal solution by adopting the negative/undesirable identity features observed in the peer group. In this context, AISA is applied to identify the unknown parameters of various benchmark test PV models, i.e., single-diode, double-diode, and PV module models. Obtained results showed that this algorithm performed very accurately since lower values of root mean square errors (RMSE) are achieved ( 9.8602 × 10 - 4 , 2.4251 × 10 - 3 , 1.7298 × 10 - 3 , 1.6212 × 10 - 2 , a n d 6.3666 × 10 - 4 ) when compared with other competitor algorithms. Further, a lower RMSE ( 9.79352834 × 10 - 4 ) was obtained in the case of the double-diode model by adapting some parameters ranges. Also, the high closeness between the simulated current–voltage ( I – V ) curve is achieved by AISA compared with the experimental data.