The Water Temperature Prediction of a Double Exposure Solar Cooker

Abstract

This article presents an experimental and theoretical examination of solar box type cooker. In the experimental work, conventional and double-exposure cookers have been compared. For experimentation, a new design was developed for solar cooker. Additional reflectors were installed on the lower part of the developed cooking room and experiments were conducted in Turkey (36-42 N and 26-45 E) in July for 0.5, 1, and 1.5 L of water. In theoretical examination, artificial neural networks (ANNs) and multiple linear regression (MLR) were used for modeling the temperature of water in the cooker and the adaptation abilities of models were compared. Rootsquare of mean square error, mean absolute error and correlation coefficient were used for comparison. When these models were compared according to the mentioned criteria, it has been observed that ANN model provided better adaptation to prediction of water temperatures used in this study compared to linear multiple regression model. Therefore, it has been concluded that ANNs could provide an alternative method to regression analysis. (c) 2013 American Institute of Chemical Engineers Environ Prog, 33: 629-635, 2014