Study on Short-Term Electricity Load Forecasting Based on SSA-LSTM-AdaBoost Modeling

Abstract

Power load forecasting is of great significance and plays a vital role in the safe operation of the power system and the stability of power supply. Aiming at the problem of low prediction accuracy of single model, this paper proposes a prediction model based on the combination of Sparrow Search Algorithm (SSA) optimized Long Short-Term Memory Network (LSTM) and integrated algorithm. Multiple weak learners are first integrated through the AdaBoost algorithm to capture patterns and features in the data from multiple perspectives. Secondly, the collective intelligence and group collaboration ability of the SSA algorithm is utilized to ensure the global convergence of the algorithm, thus improving the prediction accuracy and robustness of the LSTM model. Finally, the model is analyzed and compared by examples to verify that the prediction accuracy of the model has been further improved.