ANN modeling of forced convection solar air heater

Abstract

The design and applicability of solar air heating system require a satisfactory prediction of collector outlet air temperature and the useful energy delivered over a wide range of climate conditions. The ANN modeling is extensively used for this purpose. This article presents the results of a study carried out to compare the performance prediction by ANN. In this, an ambient temperature, solar intensity and air velocity were used as input layer, while the outputs are collector outlet temperature and first and second law efficiency of the solar air heater. The back propagation learning algorithm methods were used training and test the data. Comparison between predicted and experimental results indicates that the proposed ANN model can be used for estimating some parameters of SAHs with reasonable accuracy.