Simulation study of district heating control based on load forecasting

At present, the development of district heating system (DHS) in China is mainly reflected in the scale and structure, there are still great disadvantages in system management and control. The problem of imbalance between the heat supply and the user’s demand is serious. In this paper, a DHS in Kaifeng of China was taken as the research object, and the control model of secondary return temperature of a typical thermal power station which based on the step response is established. Based on the high-precision heating load prediction model of thermal power station, the primary side flow as the control variable, secondary return temperature as the controlled variable, and the generalized predictive control (GPC) algorithm as the control method, the secondary return temperature of the target system is accurately controlled; at the same time, particle swarm optimization (PSO) is used to determine parameters adaptively for parameter tuning in GPC; and the control strategy is simulated. Compared with the traditional Proportion integral differential (PID) control algorithm, The root-mean-square error and mean absolute percentage error of the simulation results of the control strategy and the set value are reduced by 22.24% and 22.33%, respectively, has the advantages of smaller overshoot and faster response, which can achieve the effective control of secondary return temperature and on-demand heating better.当前，我国区域供热系统（District Heating System, DHS）的发展主要体现在规模与结构层面，但其系统管控环节仍存在诸多不足，供热供给与用户需求间的失衡问题尤为突出。本文以中国开封的某区域供热系统为研究对象，构建了基于阶跃响应的典型火电厂二次回水温度控制模型。基于火电厂高精度热负荷预测模型，以一次侧流量为控制变量、二次回水温度为被控变量，采用广义预测控制（Generalized Predictive Control, GPC）算法作为控制手段，实现了目标系统二次回水温度的精准调控；同时，引入粒子群优化（Particle Swarm Optimization, PSO）算法对GPC的参数进行自适应整定，并对所提控制策略开展仿真验证。相较于传统比例-积分-微分（Proportion Integral Differential, PID）控制算法，该控制策略的仿真结果与设定值的均方根误差和平均绝对百分比误差分别降低了22.24%与22.33%，且具备超调量更小、响应速度更快的优势，可更好地实现二次回水温度的有效调控与按需供热。