Study On Modeling And Optimizing Of Water-system In The Heating Ventilating And Air-conditioning System

In the heating ventilating and air-conditioning system (HVAC), the energy consumption of the water-loop sub-system, which consists of cooling towers and chiller plants, is about 55 percent of that of the whole system, in which the energy consumption of the cooling towers and chiller plants is about 15 percent and 40 percent, respectively. Thus, it is important to reduce the energy consumption of the water-loop system in the HVAC system.In general, the design of the HVAC system is based on the assumption that the system is working on the largest loads, but in fact the HVAC system is usually run on the part loads. If the optimum work point of the HVAC system could be found by certain methods and the energy matching of different devices in the system could be adjusted by control systems, the energy consumption of the system would be greatly reduced.According to the operating process and the characteristics of energy consumption of various devices in the HVAC system, this paper proposes to optimize the work point of the water-loop system on-line, to make the HVAC system run on the optimum process during the all kinds of condition and reduce the energy consumption. Furthermore, I also study the modeling method and optimizing method according to the characteristics of crucial device, and solve the real-time optimization control problem of water system in HVAC. For the water-loop system, the configuration of the different devices is so complicated that the traditional mechanism analysis method and system identification method is not suitable to establish the model of the HVAC system, which is of highly non-linear and high coupling. In this paper, the neutral network method is developed to identify the static model of the HVAC system and the feedforward network with BP algorithm taking into account the experimental data is presented to obtain the neutral network model, which is nearly consistent with the experiment data. The optimization of the HVAC system is a kind of multi-dimension and non-linear optimization problem with constraint and in this paper the algorithm method (GA), which can find the globaloptimum point, is used to solve the optimization problem. Based on the traditional GA method, this paper presents an improved GA method combined with the compound method to optimize the work point of the static model of the HVAC system on line. By using the combined optimization method proposed in this paper, some key problems can be solved, such as the local optimum point problem of the HVAC system, which is resulted from the complexity of the system, the large mount of the variables and the difficulty of determination of the initial data, and the real time problem. The laboratory simulation is used to verify the validity of the methods proposed in this paper and program compiled by the authors.