Applied Research On Adaptive Fuzzy Control Of Vapor Compression Chiller In Central Air Conditioning System

Central air conditioning systems are widely used in large buildings to provide comfortable indoor air environment. The system load is changed with many factors, such as the users’ demands, seasons, weather and so on. How to maintain good performance of the control system under variable system load and variable system working point conditions, is vital to the stability, safety and energy efficiency of the central air conditioning system. Therefore, the study and application of advanced and intelligent control algorithms have great significance and value in theory and practice.In order to study the modeling, control and optimization of the central air conditioning system, and to combine theoretical research with practical application, design and build of the hardware platform and software platform are completed in this study. The hardware platform achieves the working of the central air conditioning system, variable frequency operating of the motors and the acquisition of the field operating parameters. The industrial computer software platform achieves the real-time online monitoring and the implementation of intelligent control algorithms.The chiller is the core part of the central air conditioning cooling system. Under variable load conditions, the vapor compression chiller exhibits strong nonlinear characteristics. To conquer the system nonlinearity, an adaptive fuzzy logic controller based on an on-line tuning mechanism of the scaling factors is used in this study. The controllers adjust the compressor frequency and the electronic expansion valve opening pulses, satisfying the control objectives of the cooling capacity and the superheat degree of the refrigerant evaporation.To test and verify the control performance of the controllers, field real experiments are carried out. By doing set-point tracking experiment, disturbance rejection experiment and variable working load experiment, the controllers performance indexes are tested and analyzed to compare the adaptive fuzzy logic controller with PID controller. Experimental results show that the adaptive fuzzy logic controller performs much better and more satisfactory than the PID controller under the condition of variable working load.