Intelligent prediction of power grid transmission line engineering investment based on XGBoost algorithm

作者：Ping Zhou, Ying Du, Yuhong Zhang, Jingyong Chen, Ying Zhou, Guangxiu Yu · 年份：2024 · DOI：10.1145/3674225.3674343 · 被引用次数：1 · 研究领域：Energy Load and Power Forecasting、Power Systems and Technologies、Smart Grid and Power Systems

Due to the lack of in-depth technical solutions, the accuracy of project investment for transmission lines determination method based on fixed budget estimation is low and the workload is large, so it is urgent to study the investment machine learning-based predictive model. In view of the high dimensional and nonlinear characteristics of transmission line investment, a method of transmission line project investment prediction based on limit gradient boost (XGBoost) algorithm is proposed. The model is trained and tested by using actual transmission line engineering data. The prediction results show that the XGBoost model is superior to the neural network and support vector machine in terms of prediction accuracy and result deviation, and can output the importance ranking of indicators, providing an effective reference for decision makers in investment and control indicators, and the model has high reliability and interpretability.