Electricity Demand and Weather Data (2021–2023)

This dataset contains hourly electricity demand data and corresponding weather indicators collected from 2021 to 2023. The electricity data was sourced from the U.S. Energy Information Administration (EIA), covering both winter and summer periods across three years. Weather features—including temperature, wind speed, and humidity—were collected to capture the external conditions affecting demand. All files are stored in CSV format and aligned by timestamp. This dataset supports research in time series forecasting, demand prediction, and energy systems modeling. It has been used in our study comparing classical machine learning and deep learning models (e.g., Random Forest, LightGBM, LSTM, Transformer, and TFT) for electricity load forecasting. The dataset is suitable for testing short-term and long-term predictive models and supports calendar feature engineering. 

本数据集涵盖2021年至2023年采集的每小时电力需求数据及配套气象指标。其中电力数据源自美国能源信息署（U.S. Energy Information Administration, EIA），覆盖三年中的冬、夏两个时段。气象特征包含气温、风速与湿度，用于捕捉影响电力需求的外部环境因素。所有文件均以CSV格式存储，并按时间戳对齐。本数据集可支撑时间序列预测、需求预测及能源系统建模相关研究。此前已在我们的研究中用于对比经典机器学习与深度学习模型（如随机森林（Random Forest）、轻量梯度提升机（LightGBM）、长短期记忆网络（LSTM）、Transformer以及时序融合Transformer（TFT））在电力负荷预测任务中的表现。该数据集适用于短期及长期预测模型的测试，同时支持日历特征工程。