Automatic Machine Learning Participation in Power Load Forecasting Under the Background of Big Data

Power load forecasting is one of the important tasks in controlling system costs, and accurate and effective load forecasting can reasonably arrange the operating status of power grid generators. Machine learning, as an important prediction method, has gradually become an important choice for complex load forecasting. However, due to the numerous machine learning methods, it is difficult to accurately and effectively select suitable algorithms in different regions when selecting prediction methods, and the parameter settings of algorithms often require debugging. This article is based on the automatic machine learning platform PyCaret, analyzing the convenience and effectiveness of automatic selection algorithms, and comparing them with the prediction accuracy of grey prediction and neural network algorithms. The results indicate that using existing platforms can accurately fit data and achieve portable and effective power load forecasting.