Performance Prediction of a Solar Water Heater Using Artificial Neural Networks

Abstract

This chapter discusses the Artificial Neural Network (ANN) for predicting the performance of a thermosyphon type solar water heater with minimum input data. This is measured in terms of the useful energy extracted from the system and stored water temperature rise. A four-layer feed forward neural network has been trained based on 29 known performance measurements. These were obtained from tests performed under varying weather conditions. In this way, the network was trained to accept and handle a number of unusual cases. Unknown data were subsequently used to investigate the accuracy of prediction. Predictions with maximum deviations of 1.8 MJ and 2.8°C were obtained respectively. These results indicate that the proposed method can successfully be used for the estimation of the performance of solar water heater operating under any weather conditions.