A Novel Data-Driven Load Forecasting Method with Decomposition Mechanism

Facing with the problem of load prediction in short-term as well as medium-term, a new hybrid model which contains ICEEMDAN, LS-SVM, as well as LSTM, has been built. Firstly, for decomposing the common dataset, the ICEEMDAN algorithm has been introduced. Secondly, the new algorithm combines LSTM and LS-SVM into a new structure of forecasting model. Then, the common dataset is introduced to train the hybrid model as well as obtain the results of STLF as well as MTLF. Finally, simulation results demonstrate the superiority of our proposed method.