Numerical Evaluation and Artificial Neural Network (ANN) Model of the Photovoltaic Thermal (PVT) System with Different Nanofluids

Abstract

The present study investigates the performance of photovoltaic thermal (PVT) systems that employ silver, aluminum oxide, copper, and titanium dioxide nanoparticles with distilled water as a solvent. The volume portions of the nanoparticles considered are 2% and 5% by weight. The study employs an energy balance equation to encompass circular geometries for fluid flow channels and a flow velocity ranging from 1×10−4 to 3×10−4 m/s. A numerical model has been established to investigate the performance of the photovoltaic thermal system and obtained the highest performance in Cu/water nanofluid for a uniform mass flow rate of 0.0670 kg/s and volume portion of 5% compared to other nanofluids, and the average electrical, thermal, and overall performance achieved is 15.8%, 30.2%, and 45.3%, respectively. Moreover, an artificial neural network (ANN) was developed to predict the electrical and thermal efficiency of the PVT system, and the mean absolute percentage error (MAPE) between array error of the thermal and electrical efficiency of the system is 4.98% and 2.61%, respectively. This value shows the strong validation of the numerical and ANN simulation values.