Coyote Optimization Algorithm for Parameters Estimation of Various Models of Solar Cells and PV Modules

Abstract

Recently, building an accurate mathematical model with the help of the experimentally measured data of solar cells and Photovoltaic (PV) modules, as a tool for simulation and performance evaluation of the PV systems, has attracted the attention of many researchers. In this work, Coyote Optimization Algorithm (COA) has been applied for extracting the unknown parameters involved in various models for the solar cell and PV modules, namely single diode model, double diode model, and three diode model. The choice of COA algorithm for such an application is made because of its good tracking characteristics and the balance creation between the exploration and exploitation phases. Additionally, it has only two control parameters and such a feature makes it very simple in application. The Root Mean Square Error (RMSE) value between the data based on the optimized parameters for each model and those based on the measured data of the solar cell and PV modules is adopted as the objective function. Parameters' estimation for various types of PV modules (mono-crystalline, thin-film, and multi-crystalline) under different operating scenarios such as a change in intensity of solar radiation and cell temperature is studied. Furthermore, a comprehensive statistical study has been performed to validate the accurateness and stability of the applied COA as a competitor to other optimization algorithms in the optimal design of PV module parameters. Simulation results, as well as the statistical measurement, validate the superiority and the reliability of the COA algorithm not only for parameter extraction of different PV modules but also under different operating scenarios. With the COA, precise PV models have been established with acceptable RMSE of 7.7547x10-4 , 7.64801x10-4 , and 7.59756 x10-4 for SDM, DDM, and TDM respectively considering R.T.C. France solar cell.

Optimal Energy Scheduling for Data Center With Energy Nets Including CCHP and Demand Response

“…PV generation is mainly influenced by solar irradiance and ambient temperature [42]. According to [43], the modelling of PV generation is given as: (9) (10) (11)…”

A Forensic-Based Investigation Algorithm for Parameter Extraction of Solar Cell Models

“…A triple-phase teaching-learning-based optimization (TPTLBO) [24], Coyote Optimization Algorithm (COA) [25], an interval branch and bound algorithm [26] Tree Growth Algorithm (TGA) [27], are applied to extract the parameters of different PV models of the three models. shuffled complex evolution (SCE) [28] technique was developed for only extracting the intrinsic parameters of the PVTDM.…”

Identification of Solar Photovoltaic Model Parameters Using an Improved Gradient-Based Optimization Algorithm With Chaotic Drifts

“…These structures are influenced by natural events, such as swarming activities, mechanisms focused on nature, and physics. Genetic algorithm (GA) [18], [19], particle swarm optimization [20]- [22], enhanced leader particle swarm optimization algorithm (PSO) [23], niche particle swarm optimization in parallel computing algorithm [24], several versions of differential evolution (DE) [25]- [28], penalty-based DE algorithm [29], sunflower optimizer [30], grey wolf optimizer (GWO) [31], whale optimizer algorithm (WOA) [32], harris-hawk optimizer (HHO) [33], improved salp swarm algorithm (ISSA) [34], several version of JAYA algorithm [35], multiple learning backtracking search algorithm [36], coyote optimization algorithm [37], teaching-learning-based optimization and its various versions [38]- [44], political optimizer (PO) [4], evolutionary shuffled frog leaping algorithm [45], slime-mould optimizer (SMO) [46], [47], marine predator algorithm (MPA) [48], equilibrium optimizer (EO) [49], ions motion optimization (IMO) [50], improved PSO (IPSO) [51], Forensic-based investigation algorithm [52], and improved learning-search algorithm [53] are among good heuristic-based structures. Some studies have endeavored to hybridize a few of these strategies to boost their performance, such as hybrid grey wolf optimizer with cuckoo search algorithm [54], hybrid firefly with pattern search algorithms [55], hybrid grey wolf optimizer with particle swarm algorithm [56], hybrid WO with DE algorithm [26], hybrid GA with simulated annealing algorithm [18], etc.…”