| With the rapid development of intelligent buildings, building automation systemshave gradually strengthened the scientific, economic and rationalized control andmanagement of construction equipment. This not only enhances the building’sfunctionality and efficiency, but also achieves the result of energy savings. As the coreof the Building Automation System, HVAC system’s intelligence is the key of asignificant reduction in building energy consumption and the effect of environmentalcontrol.The variable air volume air conditioning system has a good comfort, flexibilityand energy efficiency, so it has gradually become the mainstream trend.VAV BOX is the key component of VAV air conditioning system. Its operatingconditions directly affect the operating results of the air conditioning system. This paperdiscusses the precise control and immunity issue of temperature in the room, and usesfuzzy self-tuning PID control algorithm and neural network predictive control algorithmto achieve precise control of VAVBOX. At first the VAVBOX is identified. The airdamper is analyzed using an open-loop identification method, the room model analyzedwith a closed-loop identification method. In theory, this paper uses fuzzy control andneural network control, combined with the traditional PID control and predictive control,to constitute the fuzzy PID controller and neural network model predictive controller tocontrol the VAV mathematical model, and uses SIMULINK toolbox of MATLABsoftware simulate the control system. Simulation results show that the fuzzy PID controland neural network model predictive control are significantly superior thanconventional PID in the control accuracy and immunity, and laid the foundation forfuture applications in engineering practice. |
