PERFORMANCE ANALYSIS OF SIMPLE VCR SYSTEM BY DEVELOPING ANN MODEL WITH SUCTION & DELIVERY, PRESSURE & TEMPERATURE AS INPUT AND COMPRESSOR WORK AS OUTPUT PARAMETER
Abstract

Author(s): Rakesh Kumar Agrawal, G.K. Agrawal

In this study, artificial neural networks (ANNs) model has been developed to analyse effect on compressor work of the other variables like suction pressure, delivery pressure, suction temperature & d elivery temperature. In developed ANN model network 2,network type feed forward back propagation with training function TRAINLM, adaptation learning function LEARNGDM and with other parameter network has been successfully trained to analyse performance analysis of simple vapor compression refrigeration system using refrigerant R134a, which does not damage ozone layer. Experimentation was conducted to investigate effect of suction pressure and other variables like suction temperature to compressor, delivery pressure, outlet temperature to compressor and compressor work per kg of refrigerant. As we know conventional analytical approach involves more complicated formula & assumptions ,whereas experimental studies are tedious ,so in this paper an attempt has been made to train (ANNs) for suction pressure range(156kPa-425kPa), delivery pressure range(1101kPa-1769kPa) suction temperature range(10 0C - 34 0C) ,outlet temperature from compressor range(68 0C - 88 0C)as input to artificial neural networks (ANNs) model network2and it has been successfully trained for output as compressor work. Experimental output and output predicted from network2 resembles close to each other withR2=0.9999858,RMSE = 0.128kJ/kg, COV=0.379%& ANN with Network type -feed- forward back prop, training function- TRAINLM, adaptation learning function –LEARNGDM, with 8 No of neuron can be successfully applied in the field of performance analysis of simple vapor compression refrigeration (VCR) system. Keywords-CVR System, Suction, Delivery, Pressure & Temperature 

johronline
+44 193 554 8010
Google Scholar citation report
Citations : 18

Journal of Harmonized Research in Engineering received 18 citations as per google scholar report

Indexed In