Synthesizing hourly meteorological data to improve the accuracy of calibrated simulation models

Conference · Thu Jul 01 04:00:00 EDT 1999

OSTI ID:20030597

Kissock, K; Joseph, H

Dynamic building energy simulation models typically require a year of hourly meteorological data as input. For design purposes, the most common source of these data are the typical meteorological year TMY2 data sets. When calibrating simulation models to measured building energy consumption, the use of typical meteorological data introduces a source of error because the weather during the calibration period is different than the typical weather. Unfortunately, this is still common practice because of the difficulty of obtaining recent hourly meteorological data. In this paper, the error from using typical meteorological data to calibrate simulation models is estimated by simulating annual energy consumption of a residence and large commercial building for 20 years in four US cities. To reduce this error, the authors propose a method to synthesize hourly dry-bulb temperature, global solar radiation on a horizontal surface and specific humidity from readily available average daily temperatures. The method is based on regression results form TMY2 data sets. The synthetic data are shown to have very little bias and, in weather sensitive buildings, significantly reduce the error associated with calibrating models using typical meteorological data.

• Research Organization: Univ. of Dayton, OH (US)
• OSTI ID: 20030597
• Country of Publication: United States
• Language: English

Related Subjects

14 SOLAR ENERGY
32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATION
AMBIENT TEMPERATURE
CALIBRATION
COMMERCIAL BUILDINGS
DATA COVARIANCES
ENERGY CONSUMPTION
HUMIDITY
MATHEMATICAL MODELS
METEOROLOGY
RESIDENTIAL BUILDINGS
SOLAR RADIATION