Load prediction of district heating systems based on the multiscale mining method

Adopts the multiscale mining method to realize the operation optimization of the thermal engineering process. Based on the operation data of a district heating system, with the method of best wavelet packet analysis, calculates the contributions of time series in different scale to the development trend of heating load. Predicts the quantity of heat needed in future using the sampling combination prediction model. Simulation result shows that the multiscale mining method can predict on realtime with high precision and be generalized.