Short-term power load forecasting based on TCN-LSTM model

In order to improve the accuracy of short-term power load forecasting and provide a stronger guarantee for the stable operation of the power system, this paper proposes a short-term power load forecasting method TCN-LSTM, which combines time convolution network ( TCN ) and long short-term memory ( LSTM ). Firstly, the time series features in the original load data are extracted by using the characteristics of TCN. Then, the extracted temporal features and non-temporal features are input into the LSTM model to train the model ; finally, the trained model is used to predict the short-term power load data. Through the simulation experiment of the load data of a certain area in the south, the TCN-LSTM combined model is compared with the single model. It is found that the prediction accuracy of the TCN-LSTM model is higher, indicating that the model has better short-term power load forecasting ability.The experimental simulation comparison shows that the prediction accuracy of the combined model proposed in this paper is 0.66 % higher than that of the basic model.