ASHRAE
The American Society of Heating, Refrigerating and Air-Conditioning Engineers, a professional organization that develops industry standards such as building energy consumption prediction validation; its Guideline 14 specifies calibration error thresholds for energy models.
Timeline (last 90 days)
Compared with existing methods, DySENSE improves prediction performance by 8.7% (accuracy) and reduces prediction error by 9%–18% (CVRMSE, ASHRAE standard).
CVRMSE (coefficient of variation of the root mean square error) is the error metric recommended by ASHRAE for validating building energy consumption predictions, calculated as RMSE divided by the mean of observed values, and is a dimensionless relative error.
The ASHRAE Great Energy Predictor III dataset was released by ASHRAE in 2019 and covers hourly energy consumption records from more than 1,400 buildings across 16 global sites, including four energy types—electricity, chilled water, steam, and hot water—with a time span of two years.
All Facts (3)
Compared with existing methods, DySENSE improves prediction performance by 8.7% (accuracy) and reduces prediction error by 9%–18% (CVRMSE, ASHRAE standard).
50%UnverifiedCVRMSE (coefficient of variation of the root mean square error) is the error metric recommended by ASHRAE for validating building energy consumption predictions, calculated as RMSE divided by the mean of observed values, and is a dimensionless relative error.
50%UnverifiedThe ASHRAE Great Energy Predictor III dataset was released by ASHRAE in 2019 and covers hourly energy consumption records from more than 1,400 buildings across 16 global sites, including four energy types—electricity, chilled water, steam, and hot water—with a time span of two years.
50%