Comprehensive Load Forecasting Archive
收藏资源简介:
本研究提供了一个全面的电力负荷预测数据集,包含11个不同来源的数据集,主要来自UCI机器学习数据库、Kaggle数据竞赛平台和全球能源预测竞赛。数据集涵盖了从建筑级到聚合级的负荷数据,以及温度等外部变量,旨在为电力系统的日前调度提供重要参考。数据集的创建过程中,特别关注了温度和日历变量对负荷的影响,并通过特定的特征工程方法来优化预测模型。该数据集的应用领域主要集中在电力负荷预测,特别是为电力系统的经济调度和运营决策提供支持。
This study presents a comprehensive electricity load forecasting dataset comprising 11 datasets from diverse sources, mainly originating from the UCI Machine Learning Repository, Kaggle Data Competition Platform, and Global Energy Forecasting Competition. The dataset covers load data ranging from building-level to aggregated-level, as well as external variables such as temperature, aiming to provide critical references for day-ahead scheduling of power systems. During the dataset creation process, particular attention was paid to the impacts of temperature and calendar variables on load, and specific feature engineering methods were adopted to optimize forecasting models. The application scenarios of this dataset mainly focus on electricity load forecasting, particularly supporting economic dispatch and operational decision-making of power systems.




