应用Matlab神经网络工具箱函数,建立了可用于热网供暖短期预测的前向反馈BP网络模型,并根据乌鲁木齐市华源热力公司的实际供热网络数据,对所建立的网络进行了训练和检验.结果表明:该模型具有较好的预测特性,为热网监控与调节提供了依据,从而可实现按需供热,节约能源. Using the Matlab neural network toolbox function, a forward feedback BP network model that can be used for short-term forecasting of heating network was established, and the network was trained and tested according to the actual heating network data of Huayuan Heating Company in Urumqi The results show that this model has good predictive properties and provides the basis for heat network monitoring and regulation, which can realize on - demand heating and energy saving.