This dataset describes the timeseries energy consumption of the U.S. residential building stock at the end-use level. For details on how it was created and validated, please see the project's final report. For instructions on how to access pre-created aggregate data files, please visit the End Use Load Profiles for the U.S. Building Stock project website. For commercial building stock data, please see the ComStock data viewer. Two versions of this dataset have been created: one with actual weather data from 2018, and another with typical (TMY3) weather data. The TMY3 15-min energy data should not be used for larger geographies because weather events are not regionally aligned. Multiple geographic views of the dataset have been created, one by state, and one for each Census region by PUMA.
|Publication date||Release #||Building stock represented||Upgrades applied||Weather year||Data Viewer Links - Annual and Timeseries Energy||Data Viewer Links - Building Characteristics||Data Table with Characteristics and Annual Energy Use (metadata.tsv)||Supplemental Data and Dictionaries||Input building characteristics|
|October 2021||1||U.S. residential sector circa 2018||None||amy2018||by_state, by_puma_northeast, by_puma_midwest, by_puma_south, by_puma_west||by_state, by_puma_northeast, by_puma_midwest, by_puma_south, by_puma_west||metadata||suppl_data_dict||github|
|October 2021||1||U.S. residential sector circa 2018||None||tmy3||by_state, by_puma_northeast, by_puma_midwest, by_puma_south, by_puma_west||by_state, by_puma_northeast, by_puma_midwest, by_puma_south, by_puma_west||metadata||suppl_data_dict||github|
The Supplemental Data and Dictionaries link will take you to folders on OEDI that contain pre-aggregated timeseries .csv files for different geographies, weather data in .csv format, geographic information, OpenStudio building energy models, and a data dictionary and enumerations dictionary defining all data fields and parameter options.
First release of a national scale model that has been calibrated (at the timeseries resolution) to a large amount of measured electricity data, including hourly AMI data from 8 utilities around the country, as well as end-use level data from a variety of sources. See the project's final report for details about model assumptions, data sources, and results of the calibration process.
All datasets, including those that are privately shared can be accessed here: