基于MPC集中供热二次网回水温度控制研究  

Temperature Control of Return Water in the Secondary Network of Centralized Heating Based on MPC

作　　者：董家祥 迟耀丹[1] 赵阳[1] 苏天博 DONG Jiaxiang;CHI Yaodan;ZHAO Yang;SU Tianbo(Key Laboratory of Building Comprehensive Energy Conservation in Cold RegionJilin Jianzhu University,Changchun 130118,China)

机构地区：[1]吉林建筑大学寒地建筑综合节能教育部重点实验室,长春130118

出　　处：《建筑节能（中英文）》2024年第7期102-108,共7页Building Energy Efficiency

基　　金：吉林省科技发展计划资助项目:“大型厂区智慧安全管理系统研究与应用示范”(20230203139SF)。

摘　　要：在换热站质调节系统控制中,因被控对象具有时变性、不确定性及非线性等特点,传统的PID算法难以满足对换热站二次网回水温度控制精度的要求。提出一种基于阶跃响应的模型预测控制策略(MPC),以实现对供热二次网回水温度的精确控制。基于现场采集供热数据,通过采用递推阻尼最小二乘法算法(RDLS)对供热二次网模型的过程参数进行辨识,引用动态矩阵控制算法(DMC)进行控制量参数调节;将目标优化函数用于动态矩阵控制算法中进行控制量寻优。通过MATLAB仿真验证表明,该控制策略与传统的PID控制相比,该方法能够较好地控制换热站二次网回水温度,具有超调量小、调节时间短的优点,最后通过现场可编程门阵列(FPGA)对该方法进行程序设计并实现。In the control of quality regulation system for the heat exchange station,due to the time-varying,uncertain,and nonlinear characteristics of the controlled object,traditional PID algorithms are difficult to meet the accuracy requirements for the return water temperature control of the secondary network of the heat exchange station.A Model Predictive Control(MPC)strategy based on step response is proposed to achieve accurate control of the return water temperature of the heating secondary network.Based on on-site collected heating data,the process parameters of the heating secondary network model are identified using the Recursive Damping Least Squares(RDLS)algorithm,and the Dynamic Matrix Control(DMC)algorithm is used to adjust the control parameters;the objective optimization function is applied to the dynamic matrix control algorithm for control quantity optimization.Through MATLAB simulation verification,it is shown that compared with traditional PID control,this control strategy can better control the return water temperature of the secondary network of the heat exchange station,with the advantages of small overshoot and short adjustment time.Finally,the method is programmed and implemented using a Field Programmable Gate Array(FPGA).

关 键 词：换热站 模型预测控制 现场可编程门阵列 动态矩阵 

分 类 号：TU83[建筑科学—供热、供燃气、通风及空调工程] TK37[动力工程及工程热物理—热能工程]