Title: Identification of fuzzy model of refrigerant condenser via Adaptive-Neuro-Fuzzy Inference System in vapour compression air conditioning system

Authors: Jagdev Singh, Nirmal Singh, J.K. Sharma

Addresses: Department of Mechanical Engineering, Beant College of Engineering and Technology, Gurdaspur (Pb) 143521, India. ' Department of Mechanical Engineering, Beant College of Engineering and Technology, Gurdaspur (Pb) 143521, India. ' Swami Paramanand College of Engineering and Technology, Lalru District, Mohali 140507, Punjab, India

Abstract: In this article, fuzzy model has been identified to study the effect of refrigerant flow and condenser temperature on condenser superheat. Fuzzy model of two input variables, mass flow rate and condenser temperature, one output variable condenser superheat was developed to describe its significance in vapour compression air conditioning system. Adaptive Neuro-Fuzzy Inference System (ANFIS) was used to study the trend of condenser superheat depending upon condensing temperature and refrigerant mass flow rate. To develop an effective Multi Input Single Output (MISO) model, fuzzy rule base was designed for vapour compression air conditioning system. The effect of input parameters on condenser superheat through the developed fuzzy model was studied. Simulation of fuzzy and mathematical model was carried which showed that the MISO control can significantly improve the superheat behaviour at the condenser pressure and hence energy efficiency of vapour compression refrigeration cycle.

Keywords: adaptive neuro-fuzzy inference system; ANFIS; fuzzy modelling; refrigerant condenser; R-134a; vapour compression cycle; air conditioning; fuzzy logic; refrigerant flow; condenser temperature; condenser superheat; input variables; mass flow rate; mathematical modelling; energy efficiency; multi input single output; MISO control.

DOI: 10.1504/IJAAC.2010.030809

International Journal of Automation and Control, 2010 Vol.4 No.2, pp.158 - 176

Published online: 06 Jan 2010 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article