Research On Control Strategy Of Heating End System

With the improvement of living standards,people have higher and higher demands on heating comfort during heating.At present,the "full-open" and "full-close" control methods of the central heating end of urban buildings both reduce indoor thermal comfort and cause unnecessary heat loss.At present,how to achieve energy saving in heating and how to improve heating comfort is an urgent problem to be solved in central heating of buildings.The sub-topic of the "National Key R & D Program of the 13 th Five-Year Plan" is "Development and Application Demonstration of Real-time Optimization and Control Tools for Electromechanical Equipment Systems"(No.2017YFC0704207-03),under which funded,the thesis takes the central heating terminal as the research object.We carry out the research on the central heating end control strategy,the main work completed is as follows:(1)Based on heat transfer theory,we analyzed the thermal dynamic process of the heating space from various aspects such as maintenance structure,windows,disturbing heat sources and solar radiation,and established a mathematical model of the heat balance of the heating space to study and control the control strategy of the heating end.The design of the device provides a theoretical basis.(2)Based on the analysis and comparison of traditional PID control parameter tuning methods and the introduction of reinforcement learning theory,we propose a Q-learning online optimization PID parameter heating end flow control algorithm.The algorithm aims at the minimum energy consumption during the room temperature adjustment process,and uses the temperature difference change as the agent’s reward and punishment strategy.Through Q learning,the optimization of the Q table and the online adjustment of the PID control parameters are realized.(3)In order to verify the effect of the control algorithm and control strategy,we carried out a comparative simulation experiment of the control effect of the heating end under different control methods.The experimental results show that,compared with the traditional PID control algorithm,the heating end flow control algorithm based on Qlearning online optimization of PID parameters has a gentler temperature change in the heating space and changes in the opening of the regulating valve,and can save 33% of heating Energy saving effect.