A Electricity Price Prediction Method Based on Improved Long Short Term Memory for High Proportion New Energy Electricity Markets

The electricity price of electricity market with high proportion of new energy has great volatility, but the prediction effect of traditional deep learning model is not good. This paper proposes a price forecasting method using ATT (attention mechanism) to improve LSTM (long short term memory) and DBO (dung bee optimizer) to obtain the optimal parameters. Firstly, the attention mechanism is used to improve the extraction ability of the input features that play a key role in the electricity price forecasting. Secondly, the DBO algorithm is used to optimize the model to obtain the optimal parameters. Finally, the optimal parameters ATT-LSTM model is used to obtain the optimal forecasting results. This paper uses the real data of the electricity market with a high proportion of new energy to verify the proposed method. Compared with the LSTM algorithm, the prediction accuracy of the proposed method is improved by 65% under the MAE index, and compared with the single GRU and BP algorithm, the prediction accuracy is improved by 74.9% and 83.1%, respectively, which has high prediction performance.