A simple, reliable and adaptive approach to estimate photovoltaic parameters using spotted hyena optimization: A framework intelligent to predict photovoltaic parameters for any meteorological change

Abstract

Photo-voltaic (PV) parameter estimation is inevitable for researchers and industrialists to comprehend accurate PV models. Particularly, modeling and simulation studies are pivotal for optimal design, control, testing, and forecasting the performance of PV systems. However, PV parameter estimation for different irradiation and temperature conditions demands numerous trial runs. Moreover, control parameter tuning associated with optimization techniques increases the computation complexity of the parameter estimation process. In this context, an adaptive 12-parameter model with the inherent capability to predict the two-diode model parameters for any meteorological data without trial runs is proposed in this work. To evolve an adaptive PV model, PV parameters of PV modules such as (i) S36, (ii) SM55, (iii) S75, (iv) ST40, (v) S60, and (vi) SQ150 are conventionally extracted at various temperatures and irradiations using spotted hyena optimization method. These parameters are plotted on the curve-fit toolbar of MATLAB to evolve mathematical equations for series resistance (Rs), parallel resistance (Rp), short circuit current (Isc) and open-circuit voltage (Voc) as a function of temperature and irradiation. It is important to note here that these equations are independent to predict PV parameters (Rs,Rp,Isc&Voc)for any meteorological change. Meanwhile, the unknown 12-parameters of these equations are optimized via the spotted hyena optimization method. Despite these parameters, the remaining two-diode model parameters are computed using analytical equations. From the evolved 12-parameter model, PV parameters of five different PV modules are (Kyocera KC200GT, SP140W, KS20T, Kyocera 20 T, and SP190) extracted and results concerning root mean square error are tabulated. The results attained not only assure high accuracy but also showcase substantial improvement in the reliability of the parameter estimation approach, which has also been validated by five various trial runs exemplifying the consistent performance.