Research On Multi-variable Adaptive Control Of Central Air-conditioning System

With the improvement of people’s living standard, the central air conditioning system has a rapid development in construction industry, air conditioning system plays a more important role than ever before, especially in the aspect of building energy consumption. So how to use intelligent control technology to control it has become a hot issue in control field nowadays. However variable air volume system has most significantly energy saving effect in central air conditioning system, it will be a core object of air conditioning research field, but variable air volume is a multi-variable, large delay, nonlinear and uncertain system, the traditional PID control has been difficult to meet the requirements of its multi-variable dynamic control, In order to improve the control performance of air conditioning system to achieve better energy-saving effect, based on actual project, this paper proposed three improved controller design schemes, and use different intelligent control algorithms to optimize them and to achieve effective control of temperature. And then though MATLAB simulation experiment, we analyzed the results to get the best control strategy.The main work in this paper is as follows:1. Take actual project as background, combined with the design of a mall’s C part, we illustrated the building monitoring system, composition and working principle of air conditioning system, and the control principle and characteristics of all parts in air conditioning system, such as cooling and heating system, fresh air handling unit and air conditioning unit.2. Aiming at the characteristics which central air conditioning system is a multi-variable, nonlinear,large delay and uncertain system, this paper put forward three different controller design schemes, respectively recurrent wavelet neural network predictive control, fuzzy neural network predictive control and T-S fuzzy recurrent neural network predictive control. And then we use intelligent control algorithm to carry on control research.3. We established the mathematical model of VAV terminal control room, and determined model parameters. And then we used the process map to illustrate the achievement of control method. On the basis of those, we analyzed the results of MATLAB simulation experiment, and compared them with traditional PID control to get best control scheme.