Advances and Evaluation of Intelligent Techniques in Short-Term Load Forecasting

In order to maintain the stability of the power grid, various load forecasting methods have emerged in endlessly. However, due to different characteristics such as algorithm generalization capabilities and model complexity, their applicability to load forecasting varies. This article discusses short-term power load forecasting in the past five years. In summary, this paper summarizes the various dimensions such as experimental data sets, data preprocessing, prediction algorithms, optimization models, and evaluation methods of current research status in power load forecasting. In addition to the advantages, disadvantages, and applicability of various prediction algorithms, this paper also sums up and prospects the development trends of the short-term power load forecasting system, so as to provide a reference for the selection of power system load forecasting models in the future.