城市集中供热网参数预测研究

Forecasting City Central Heating Network Supply Parameters

城市集中供热系统已经成为我国北方冬季供暖的一种主要形式,而预测集中供热系统未来一段时间内的参数变化趋势,对热网的运行、调节均起着重要作用。通过对热网中的不确定信息的分析,采用基于模型的模糊神经网络结构,用误差反传算法来训练网络的连接权系数,以呼市热力公司某供热站的供热网参数为研究对象予以建模和预测,其结果表明该算法具有较好的泛化、学习、映射能力。

The city central heating network has become a dominating form of supplying central heating in the north of China in winter. Ghanging tendency of the parameters forecasting will play a key role in controlling and running the city central heating nefivork, This Paper presents a method for thermal power supply parameters modeling and forecasting based on a fuzzy neural network model by an- alyzing the indefinite information on central heating network , in which the Back Propagation algorithms is used to train the connection weights of the fuzzy neural network. The modeling and forecasting results about the central heating network parameters of a supply site in Huhhot Thermal Power Company has shown that the method has good properties of generalization, learning and reflecting caoabilities.