
This CSV-formatted dataset provides two years of hourly whole-building chilled-water meter readings for non-residential structures. The Building Data Genome Project 2 provides a comprehensive collection of energy consumption metrics derived from real-world non-residential assets. This specific dataset focuses on chilled water meter readings, offering time-series data captured at hourly intervals over a continuous two-year duration. These readings are essential for researchers and engineers seeking to understand HVAC demand profiles and cooling load fluctuations across diverse building types. The structured CSV format ensures compatibility with standard data analysis software, allowing for immediate integration into energy prediction models, benchmarking studies, and long-term performance monitoring workflows within professional AEC environments.
The download is a single file containing .CSV files.
Professionals integrate this dataset into energy analysis software, data science environments like Python or R, or spreadsheet applications for rigorous statistical evaluation. Users typically import the CSV files to perform exploratory data analysis, train machine learning algorithms for energy forecasting, or develop baseline models for building retrofits. The consistency of the hourly timestamps allows for seamless synchronization with external weather data or occupancy schedules, facilitating high-fidelity simulations for HVAC optimization and sustainability reporting in complex non-residential building projects.
Building Data Genome Project 2 Chilled Water Meter Data
Free
Engineers and sustainability consultants utilise this data when developing energy conservation measures or validating building performance against industry benchmarks. It serves as a foundational resource for projects involving large-scale commercial offices, educational facilities, and healthcare campuses where cooling load management is critical. By leveraging this open-source data, practitioners can test predictive models before applying them to live building management systems. The dataset is particularly valuable during the design and operations phases of high-performance buildings, supporting informed decision-making regarding mechanical system efficiency and operational energy reduction strategies.
File formats
What changed in 1.0.0
Initial release.
CC BY SA 4.0
Attribution: Clayton Miller et al. — Building and Urban Data Science (BUDS) Lab, National University of Singapore — licensed under CC-BY-SA-4.0
Full terms have not been published for this listing. Contact the seller before using it on a commercial project.