基于数据驱动的供热系统优化调控研究

Research on data-driven optimization and regulation of heating systems

针对传统供热系统运行过程中调节方式简单造成能源浪费等情况,基于系统运行数据构建了以输配能耗最小为目的的机器学习调控模型,提出一种基于数据驱动的供热系统优化调控方法。通过机器学习建立了单元阀门调控模型、短期热负荷预测模型和换热站流量调控模型,对各模型进行研究分析,得到最优模型参数组合。对优化前后的运行数据进行实验对比,结果表明:在供热量不变的情况下,优化后平均循环流量和泵耗明显降低,泵耗的节能率达到38%,具有良好的节能效果。

In view of the situation that the simple regulation method in the operation of traditional heating system causes energy waste,a machine learning regulation model aiming at minimizing the energy consumption of transmission and distribution is constructed based on the system operation data,and a data driven optimization regulation method for heating system is proposed.A unit valve control model,short-term heat load prediction model,and heat exchange station flow control model were established through machine learning.The optimal model parameter combination was obtained through research and analysis of each model.Through experimental comparison of operational data before and after optimization,the results show that:under the condition of constant heat supply,the average circulating flow rate and pump consumption after optimization are significantly reduced,and the energy-saving rate of pump consumption reaches 38%,which has a good energy-saving effect.