针对供热系统调峰热水锅炉集群的参数优化问题，应用基于数据驱动的神经网络算法，接入DCS系统锅炉数据，辨识建立锅炉的运行性能模型，以锅炉运行参数为决策变量、锅炉集群运行经济性最优为目标函数，应用粒子群算法优化获得最优的锅炉集群运行方案。以三河新源供热调峰锅炉集群的运行数据为实例，对3台和4台锅炉运行参数进行了优化，结果表明：通过优化获得的方案，与实际运行方案对比后，可产生约7%的节能量。

This paper aims at optimizing the parameters of hot water boiler cluster to achieve peak shaving in heating system through three steps.The first step is to connect the boiler data from DCS and use neural network algorithm to identify and establish boiler performance model.The second step is to take the boiler operation parameters as decisive variables,and choose the operation economic of boiler cluster as the objective function.The third step is to optimize the operation economy by using particle swarm optimization.In this study,we use the operation data of boiler cluster with three boilers and four boilers of Sanhe XinYuan Heating Company,to show the effectiveness of such approach.In addition,we determine the optimal operation scheme for the boiler cluster.The comparison between the optimized scheme and the actual operation scheme indicates the optimized scheme has the potential to save nearly 7% energy consumption.

TK39

国家科技支撑计划课题（2014BAA06B01）