Modeling of a solar-powered thermoelectric air-conditioning system using a random vector functional link network integrated with jellyfish search algorithm

In this study, the performance of a solar thermoelectric air-conditioning system (STEACS) is predicted using advanced optimized artificial intelligence models. A STEACS powered by solar PV panels is experimentally tested under different cooling loads varying from 65.0 to 260 W. The obtained experimental data are used to train and test the model. The model consists of a random vector functional link (RVFL) network optimized by one metaheuristic optimizer such as jellyfish search algorithm (JFSA), artificial ecosystem-based optimization (AEO), manta ray foraging optimization (MRFO), and sine cosine algorithm (SCA). The inputs of the model were time, solar irradiance, ambient temperature, wind speed, and humidity. The predicted responses of the investigated system are the input current of PV, the average temperature of the air-conditioned room, the cooling capacity, and the coefficient of performance. The accuracy of the four models is evaluated using eight statistical measures. RVFL-JFSA outperformed the other models in predicting all responses with a correlation coefficient of 0.948-0.999 and, consequently, it is recommended to use it to model STEACS system.