The Control Study On Terminal Appliance Of VAV Air-Conditioning System Based On Fuzzy Neural Networks

Since energy-saving has become the global issue for development of all countries to consider, as the biggest proportion of building energy consumption in the air conditioning has become the focus of attention. In recent years, variable air volume (VAV) has drawn more and more attention of researchers at home and abroad for its high efficiency and energy-saving. For VAV air-conditioning systems, its control system plays a decisive role on effects of the energy-saving. The control of VAV air-conditioning system is the control of its core--VAV Terminal Appliance.In this paper, the characteristics of PID (Proportion Integration Differentiation) control method and fuzzy control method are analyzed on the basis of the characteristics of the VAV air-conditioning system terminal appliance. Compared with PID control, the fuzzy control although has made great progress in technology, it is hard to say whether expert experience will directly influence the effect of air conditioning control, because membership functions and the control rules rely on the experience of experts.In order to solve the above problems, this paper explores the use of combining fuzzy control and neural network to control the VAV terminal appliance. According to the actual operation characteristics of the air condition, the temperature deviation and the change ratio of temperature deviation that is used as the input in the control system are determined. On this basis, the influences of the linguistic input and output variables and the corresponding mapping function are studied, and by combining actual operating experience, the fuzzy control rules of the control system are concluded. Using the above findings, fuzzy neural network model that is used to control the terminal appliance of VAV air-conditioning system is studied and designed, and the corresponding learning algorithm is determined. Through the study of input and output variables, membership functions and control rules are self-adjusted to improve intelligence of VAV air-conditioning system.Finally, in the simulation environment of Matlab/SimuLink, the above models are simulated. Simulation results show that both overshoots and transition time, the control method is superior to PID control and fuzzy control. As a result the feasibility of using fuzzy neural network to control the terminal appliance of VAV air-conditioning system is proved theoretically, which provides theoretical basis for further study of theory and experimental research.