EWELD: A Large-Scale Industrial and Commercial Load Dataset in Extreme Weather Events

Area of Science:

• Power Systems Engineering
• Data Science
• Climate Science

Background:

• Accurate load forecasting is vital for power system stability and economic operation.
• Extreme weather events significantly disrupt typical electricity consumption patterns.
• A lack of public datasets hinders research on load forecasting under extreme weather conditions.

Purpose of the Study:

• To address the data gap in extreme weather load forecasting.
• To introduce a comprehensive dataset of electricity consumption during extreme weather events.
• To facilitate research and development in resilient power grid management.

Main Methods:

• Collected 15-minute interval electricity consumption data from smart meters.
• Gathered data over six years from 386 industrial and commercial users in southern China.
• Annotated data with 5,741 specific records of extreme weather events (typhoons, extreme heat).

Main Results:

• A novel, large-scale dataset comprising over 50 million electricity consumption records.
• Detailed data covering 17 diverse industries.
• Marked records specifically identifying periods of extreme weather impact.

Conclusions:

• The released dataset provides a valuable resource for advancing load forecasting models.
• Enables improved understanding and prediction of power demand during extreme weather.
• Supports the development of more robust and reliable power systems.