Short-term forecasting for heating load of centralized heat-supply system

In order to coordinate the heat supply of heat source and users´ heat demand and also maintain the whole system work in a high efficiency, the wavelet analysis was utilized to make short-term prediction on heating load. The heating load of heat supply system was chosen as the output variation. After applying wavelet transform to the heat load sequence, the low-frequency signals and high-frequency signals were predicted by neutral network method and time series method. The predictive value of heat load was then obtained through the synthesis of predicted results. It can be seen that the predicted results are reasonable and able to fulfill the project demand.