Short-term Load Forecasting of a Technology Park Based on a LightGBM-LSTM Fusion Algorithm

As the marketization level of the electricity retailing keeps expanding, the power load forecasting of the technology park is significant to the power market and energy dispatch of the park itself. To enhance the prediction accuracy and reliability of load forecasting in technology parks, a multi-feature fusion algorithm of LightGBM and LSTM with K-fold cross validation is proposed in this paper. Firstly, the LightGBM algorithm with K-fold cross validation is designed to train the prediction model of the first layer features. Then, the prediction results are used as the additional eigenvalues of the next layer LSTM model training. Finally, the load value predicted by the two-layer model is weighted and averaged into the final load prediction value according to the prediction error. Combined with the data example, the LightGBM algorithm using K-fold cross validation improves the result's accuracy, and the fusion model has more advantages than the single model prediction, and the prediction error can be effectively reduced.