The Research On Multi-mode Control Of Variable Air Volume Air Conditioning Systems

As the foundation to support the process of the human civilization and the modern social development, energy has become one of the most concerned problems nowadays. China is not only a big energy production country, but also a great energy consumption country, in which, the building energy consumption accounted for the largest proportion of total energy consumption, and the air conditioning system energy consumption is the most important part of building energy consumption. Because of its ability to effectively reduce energy consumption and flexibly control variable air volume air conditioning system has become the major research object in air conditioning field. Therefore, it has important theoretical significance and practical application value for variable air volume air conditioning system to research on how to take into account the indoor thermal environment and energy saving effectThis paper is based on the idea of saving energy and adjusting indoor environment comfort. Firstly, the paper briefly introduces the variable air volume air conditioning system, and gives a detailed analysis and selection of the system control method combined with engineering example. Then, the paper put forward multi-mode control strategy based on temperature difference for the large disturbance and time variation of the air conditioning system, and research on the large temperature difference control strategy. Finally, the paper research on the small temperature difference control strategy.The main work in this paper is as follows:(1) Combined with engineering examples, this paper designs the overall air conditioning system, terminal device control solution, airflow control solution, supply air temperature control solution and fresh air control solution, and the paper also analyses the advantages and disadvantages of these control methods in detail.(2) This paper establishes compensation amount of supply air temperature Elman neural network which optimizes supply air temperature operating point aims to make the PID closed loop work in the optimal state. Fruit fly optimization algorithm(FOA) tunes of PID controller to improve the control performance. Simulation results show that Elman neural network can establish a more accurate supply air temperature compensation model; FOA- PID control strategy based on Elman neural network has higher control accuracy, faster response speed and smaller overshoots. It improves the system dynamic performance and steady-state accuracy.(3) This paper establishes the temperature model of terminal device based on the least squares support vector machine(LSSVM) and enables the system to adapt to the dynamic change of the nonlinear process. The optimal control values of air conditioning system are obtained by the rolling optimization of FOA. Simulation results show that the least squares support vector machine can rapidly and accurately establish the temperature model and has the better generalization ability. LSSVM predictive control based on FOA strategy can effectively overcome uncertainty factors of the controlled object such as nonlinearity, time variation and so on, reduce the system overshoot and oscillations, accelerate the system response speed and have strong robust and good tracking ability.