Distributed Prediction And Optimization Strategy For Dynamic System Of Central Heating Network

Central heating saves energy and protects environment. As the main form of heating at present, it is one of the infrastructures in urban construction development. Central heating system will directly have an impact on the living quality of people. However, heating system is a large, complex and multivariable system, which undoubtedly has increased the difficulty of operational control. With the improvement of people’s awareness of environmental protection and the increase of heating needs, heating operation optimization strategy was put on the agenda to solve the problem of heat maladjustment and improve the heating quality. At the same time, in order to enhance enterprise efficiency and social and economic benefits, it is imperative to realize comprehensive monitoring, management and optimization of heating operation by the modern technical means.In this paper, the physical models of the each part of the heating operation are derived based on conservation law of energy and mechanism of heating equipment. The model parameters are identified by the least square method combined with the practical response curve. Finally, the quantity and quality regulation process model of heating is established and the dynamic model of the heating system is built.As a hot issue of central heating, operation optimization has received much concern in the control field, because just reasonable and feasible optimization strategies can make the most use of energy saving of central heating. This paper explored the optimization of heating process in MATLAB environment. The simulation results show that distributed predictive control is more effective. However, the optimal solution of distributed predictive control based on Nash does not necessarily converge to the Pareto solution with the different coupling degree. Therefore, an improved algorithm for distributed predictive control is put forward based on cooperative game theory, so that the optimal solution of the improved algorithm can converge to Pareto solution whether the coupling is strong or not. The MATLAB simulation results show that the improved algorithm has good control performance on iterative tracking. Combining these strategies, this paper puts forward a regulation scheme of heating operation, distributed control on subsystem and decentralized control on whole system, which can reduce the computational complexity of on-line implementation and meet the need of users’ independent adjustment easily.In order to improve the automation level of the heating system, enhance the maneuverability of optimization strategies and save energy in maximum, this paper has finally developed the monitoring optimization software of heating. The system realized the function of real-time monitoring, history search, system optimization, etc. from three aspects of interface design, system function design and database design. It lays the foundation for the integrated operation and intelligent management of heating operation.