Determination of thermal performance calculation of two different types solar air collectors with the use of artificial neural networks

In this study, two different surface shaped solar air collectors are constructed and examined experimentally; corrugated and trapeze shaped. Experiments are carried out between 09.00 and 17.00 in October under the prevailing weather conditions of Elazığ, Turkey. Thermal performances belonging to experimental systems are calculated by using data obtained from experiments. A feed-forward neural network based on back propagation algorithm was developed to predict thermal performances of solar air collectors. The measured data and calculated performance values are used at the design of Levenberg–Marquardt (LM). Calculated values of thermal performances are compared to predicted values. It is concluded that, ANN can be used for prediction of thermal performances of solar air collectors as an accurate method in this system.