Research Of A Sampling Feedforward Control Strategy On Mixing Water Heating Network

Along with the development of urbanization in our country, the central heating engineering has made considerable progress in recent years. With the advancements of computer technology and heating network supervisory control technology, installing the mix water pumps of heating system has increasingly become a hot area of research. The mixing water heating network has many features, such as low heat loss, saving energy, less initial investment and low maintenance cost, etc. But control and regulation of mixing water heating network are complex tasks. Design valid control strategies of mixing water heating network have been an urgent task for heating and control specialists to solve.The article focuses on mixing water heating network. The dynamic model of mixing water heating network is established. Through studies with the hydraulic condition, thermal condition and control technology of mixing water heating network, a sampling feed forward control strategy is proposed on the basis of genetic algorithm(GA). The control strategy is experimentally simulated and validated.In accordance with the basic equipments used in mixing water heating network, the heating system is broken into five separate device module, including pump, valve, boiler, heating user, node. According to the law of conservation of mass and energy, the device modules’ dynamic equations of hydraulic condition and thermal condition are deduced, and the mathematical model of device modules are built. The features of heating network, such as tight coupling and large time delay, are fully considered to uplift the model’s artificiality. Based on the mathematical models of device modules, program modules are written using programming languages. Every program module can exchange data with every other one to build contacts. Many different structural mixing water heating network simulation softwares can be established by using the program modules.Based on mixing water heating network simulation software, the features of mixing water heating network, such as tight coupling and large time delay, are discussed. Influencing factors of large time delay are qualitative studied. For the characteristics of tight coupling and large time delay, a control strategy that combined feed forward control and sample adjustment is adopted in this paper. A sampling feed forward controller on the basis of genetic algorithm (GA) is designed and written using programming language. The static simulation and simulant results demonstrate this controller can finish controlling task accurately and effectively.Based on a engineering project of mixing water heating network, the controller designed in this paper is used to dynamic simulations in24time intervals in a day. The results have shown that this control strategy is able to achieve the predictive control by using genetic algorithm (GA). Revolving speed of variable frequency pumps and opening of control valves arc globally optimized though feed forward control. The method can reduce the impact of network tight coupling and degrade response time to achieve the goal of stability control for mixing water heating network.According to the requirement of metered heating engineering, self-control condition of heating users is simulated. It shows that the controller designed in this paper is basically able to eliminate the impact of heating network’s tight coupling. The room temperature regulating is able to implement independently. It allows heating users control their room temperature by themselves,which provides a technical reference for achieving metered heating engineering.