Prediction of energetic performance of a building integrated photovoltaic/thermal system thorough artificial neural network and hybrid particle swarm optimization models

The objective of this work is to evaluate the feasibility of an optimized artificial neural network and particle swarm optimization for estimating the energetic performance of a building integrated photovoltaic/thermal system. A performance evaluation criterion is defined in this study to assess the overall performance of the considered system. Then, the mentioned methods are used to identify a relationship between the input and output parameters of the system. The performance evaluation criterion was taken as the essential output of the system, while the input parameters were the channel length, channel depth, channel width, and the air mass flow rate. The results revealed that the coefficient of determination for the training phase of the artificial neural network and particle swarm optimization-artificial neural network methods is respectively 0.9982 and 0.9997, while it is 0.9980 and 0.9997 for the testing phase. Moreover, the root mean square error for the training phase of the artificial neural network and particle swarm optimization-artificial neural network techniques was respectively 0.0462 and 0.0175, while it was 0.0493 and 0.0178 for the testing phase. Therefore, it was concluded that the particle swarm optimization-artificial neural network model could slightly perform a better performance compared to the conventional artificial neural network.