Developed adaptive neuro-fuzzy algorithm to control air conditioning system at different pressures
The paper developed artificial intelligence technique adaptive neuro-fuzzy controller for air conditioning systems at different pressures. The first order Sugeno fuzzy inference system was implemented and utilized for modeling and controller design. In addition, the estimation of the heat transfer rate and water mass flow rate in/out the system was determined by fuzzy IF-THEN rules. The combination between back propagation algorithm and least square method helps to optimize the membership functions and generated the fuzzy rules that describe the relationship between the input/output data of the air conditioning system which changes with pressure values. The fuzzy rules are tuned by adaptive neural fuzzy inference system ANFIS. Statistical indices such as Root Mean Square Error (RMSE) and mean relative error (MRE) are used to evaluate performance of the ANFIS model. The values of the indices show that ANFIS model can accurately and reliably be used to air conditioning system and the simulated model of the system shows a good performance results.
Keywords: fuzzy logic, adaptive fuzzy, air conditioning system, Intelligent control