LSTM Recurrent Neural Network Based Interactive Behaviour Learning of Microgrid Under Incomplete Information

Abstract

The interactive operation of Microgrid (MG) in energy market is facing more and more incomplete information decision-making requirements due to the enhanced privacy of users, which brings great challenges to the participation flexibility of MG in demand response program. To address this problem, this paper proposes a novel LSTM recurrent neural network (RNN) based method for MG interactive operation, the proposed method uses LSTM-RNN to map the relationship between tie-line power and the available resources data outside of MG, and improves the prediction accuracy through iteratively learning of the accumulated operating data. On this basis, the interactive behaviour learning and interactive operation of MGs under incomplete information is effectively realized. The example analysis and test are carried out on a typical MG system, the validity of the proposed method is verified by comparing with traditional model-based method.