Electricity Price Forecasting Based on Transfer Learning and CNN-LSTM

The electricity price forecasting can accurately access the electricity price changing situation of the whole area and provide precise consultation for the operation decision-making of the electric power price system. Because the electricity price is affected by multidimensional factors, in order to fully obtain the temporal characteristics of the power price and promote the precision of the power price forecasting, this paper proposes a power price forecasting method based on transfer learning and CNN-LSTM. By analyzing the data set from different places and areas’ electricity prices and change tendency through CNN-LSTM, the predicted electricity price model will be built to predict the future disposition of the price and provide reasonable suggestions for the area. This model can finally promote the accuracy of the prediction of the electricity price. Through the method of the paper, it provides a more accurate approach of predicting electricity prices for whom it may concerned.