Evaluation of Coefficient of Performance of Water– Carrol Vapor Absorption Refrigeration System Using Artificial Neural Network approach

Abstract

The aim of this work is to apply expert system methods, like artificial neural network (ANN), for evaluating the coefficient of performance (COP) of water-carrol (lithium bromide-ethylene glycol)-based vapor absorption refrigeration system. Theoretical performance analysis of vapor absorption refrigeration system is too complex since very tedious second-order mathematical differential equations are used. In order to simplify this complex process, artificial neural networks (ANNs) are used. Instead of complex differential equations and limited experimental data, faster and simpler solutions were obtained by using equations derived from the ANN model. This work had achieved highest value of absolute fraction of variance (0.9999996683 and 0.99999977) and lowest value of mean absolute error (0.001542 and 0.000351) and root mean square error (0.000708 and 0.00029621). The consistency between traditional theoretical approach and ANN’s approach results was achieved by absolute relative error in the range of 0.0001 to 0.0015, which is very satisfactory.

Keywords: Vapor absorption refrigeration system, coefficient of performance, water-carrol, artificial neural networks