基于DHP算法的热力站一次网热量分配控制  

Heat Distribution of Primary Network of Heating Substation Based on DHP Algorithm

作　　者：李琦[1] 于明伟 赵峰 LI Qi;YU Mingwei;ZHAO Feng(College of Information Engineering,Inner Mongolia University of Science and Technology,Baotou 014010,China)

机构地区：[1]内蒙古科技大学信息工程学院,内蒙古包头014010

出　　处：《信息与控制》2018年第6期737-744,共8页Information and Control

基　　金：国家自然科学基金资助项目(61463040)

摘　　要：针对集中供热系统非线性、大时滞、强耦合的特点,热源到各热力站热量难以实现按需分配、均匀分配.本文提出了自适应动态规划(ADP)解决一次网热量优化分配控制的方法.首先进行热负荷预测,得到各热力站的热负荷值,然后分析集中供热系统一次网运行机理,建立热量优化分配问题的数学描述,最终通过双启发式动态规划(DHP)对该问题进行求解,得到各热力站一次侧的供水流量、供水温度的优化设定值.仿真结果表明该方法优化热量分配的有效性,避免供热不均现象的出现,提高热量的利用率,实现了按需供热的目标.The characteristics of the central heating system-nonlinearity,large time delay,and strong couplingmakes it difficult to distribute demand and uniformly distribute heat from the heat source to each heat station.We propose the use of adaptive dynamic programming (ADP) to solve the optimal allocation of heat in the primary network. First,we forecast the heat load and obtain the heat load values of the heating stations.Then,we analyze the central heating system’ s primary network operation mechanism and establish a mathematical description of heat allocation problem. Using dual heuristic dynamic programming (DHP) to solve the problem,we then optimize the set value of the water flow and temperature on one side of each heating station.The simulation results show that the availability of DHP in optimizing heat distribution can prevent uneven heating and improve heat utilization,as well as achieve on-demand heating.

关 键 词：自适应动态规划 双启发式动态规划 集中供热系统 一次网 热量分配 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]