电力负荷预测专识数据集

本数据集聚焦电力系统负荷预测建模与验证，覆盖全网统调负荷、地区负荷、母线负荷及行业负荷等类型，时间粒度包含15分钟、小时、日、月等尺度。数据内容包括时间空间标识、负荷有功/无功值、气象协变量（温度、湿度、风速、降雨等）、日历特征（节假日、工作日）、历史统计特征。适用于超短期/短期/中期/长期负荷预测模型训练、母线负荷预测、电价预测辅助、异常场景分析及电力AI算法研发等场景。This dataset focuses on modeling and validation for power system load forecasting, covering various load types including national grid unified dispatch load, regional load, bus load and industrial load, with time granularities spanning 15-minute, hourly, daily and monthly scales. The dataset contains spatiotemporal identifiers, active/reactive power values of loads, meteorological covariates (e.g., temperature, humidity, wind speed, rainfall), calendar features (e.g., holidays, workdays) and historical statistical features. It is applicable to multiple scenarios such as training of ultra-short-term, short-term, medium-term and long-term load forecasting models, bus load forecasting, auxiliary electricity price forecasting, abnormal scenario analysis and research and development of power AI algorithms.

搜集汇总

背景与挑战

背景概述

该数据集专为电力系统负荷预测建模与验证设计，涵盖全网统调、地区、母线及行业等多种负荷类型，时间粒度从15分钟到月不等，并包含负荷值、气象协变量和日历特征等关键数据。它适用于超短期至长期的负荷预测模型训练、电价预测辅助、异常分析以及电力AI算法研发等应用场景。