Artificial neural network analysis of a refrigeration system with an evaporative condenser


Ertunc H. M., Hosoz M.

APPLIED THERMAL ENGINEERING, vol.26, pp.627-635, 2006 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 26
  • Publication Date: 2006
  • Doi Number: 10.1016/j.applthermaleng.2005.06.002
  • Journal Name: APPLIED THERMAL ENGINEERING
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.627-635
  • Keywords: artificial neural network, refrigeration, evaporative condenser, COMPRESSION LIQUID CHILLERS, COOLED CONDENSER, HEAT-EXCHANGER, PERFORMANCE, PREDICTION, PUMPS
  • Kocaeli University Affiliated: Yes

Abstract

This paper describes an application of artificial neural networks (ANNs) to predict the performance of a refrigeration system with an evaporative condenser. In order to gather data for training and testing the proposed ANN, an experimental refrigeration system with an evaporative condenser was set up. Then, steady-state test runs were conducted varying the evaporator load, air and water flow rates passing through the condenser and both dry and wet bulb temperatures of the air stream entering the condenser. Utilizing some of the experimental data, an ANN model for the system based on standard backpropagation algorithm was developed. The ANN was used for predicting various performance parameters of the system, namely the condenser heat rejection rate, refrigerant mass flow rate, compressor power, electric power input to the compressor motor and the coefficient of performance. The ANN predictions usually agree well with the experimental values with correlation coefficients in the range of 0.933-1.000, mean relative errors in the range of 1.90-4.18% and very low root mean square errors. Results show that refrigeration systems, even complex ones involving concurrent heat and mass transfer such as systems with an evaporative condenser, can alternatively be modelled using ANNs within a high degree of accuracy. (c) 2005 Elsevier Ltd. All rights reserved.