Comparative analysis of an evaporative condenser using artificial neural network and adaptive neuro-fuzzy inference system


ERTUNÇ H. M., HOŞÖZ M.

INTERNATIONAL JOURNAL OF REFRIGERATION-REVUE INTERNATIONALE DU FROID, cilt.31, sa.8, ss.1426-1436, 2008 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 31 Sayı: 8
  • Basım Tarihi: 2008
  • Doi Numarası: 10.1016/j.ijrefrig.2008.03.007
  • Dergi Adı: INTERNATIONAL JOURNAL OF REFRIGERATION-REVUE INTERNATIONALE DU FROID
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.1426-1436
  • Anahtar Kelimeler: Refrigeration system, Evaporative condenser, Experiment, Performance, Modelling, Fuzzy logic, Neural network, HEAT-PUMP SYSTEM, REFRIGERATION SYSTEM, PERFORMANCE, MODEL
  • Kocaeli Üniversitesi Adresli: Evet

Özet

This study deals with predicting the performance of an evaporative condenser using both artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) techniques. For this aim, an experimental evaporative condenser consisting of a copper tube condensing coil along with air and water circuit elements was developed and equipped with instruments used for temperature, pressure and flow rate measurements. After the condenser was connected to an R134a vapour-compression refrigeration circuit, it was operated at steady state conditions, while varying both dry and wet bulb temperatures of the air stream entering the condenser, air and water flow rates as well as pressure, Temperature and flow rate of the entering refrigerant. Using some of the experimental data for training, ANN and ANFIS models for the evaporative condenser were developed. These models were used for predicting the condenser heat rejection rate, refrigerant temperature leaving the condenser along with dry and wet bulb temperatures of the leaving air stream. Although it was observed that both ANN and ANFIS models yielded a good statistical prediction performance in terms of correlation coefficient, mean relative error, root mean square error and absolute fraction of variance, the accuracies of ANFIS predictions were usually slightly better than those of ANN predictions. This study reveals that, having an extended prediction capability compared to ANN, the ANFIS technique can also be used for predicting the performance of evaporative condensers. (C) 2008 Elsevier Ltd and IIR. All rights reserved.