A statistical method for selection of sequences of coincident weather parameters for design cooling load calculations

Current design weather data recommended by ASHRAE and CIBSE may result in overestimated peak cooling loads. This is because solar radiation, and outdoor dry-bulb and wet-bulb temperatures selected for design conditions do not occur coincidently. Hence, the data cannot reflect the joint statistical distribution of these three weather parameters. Moreover, the peak cooling load largely depends on the characteristics of both weather and buildings. A statistical method has been developed for the rational selection of sequences of coincident design weather parameters in order to properly determine peak cooling loads. Overall periodic transfer factors responding to different periodic weather heat sources are first derived based on the radiant time series (RTS) method. This allows us to utilize the available thermal and optical properties of a building without the need for tedious regenerating these data. The periodic transfer factors are then equivalently transformed to z-transfer coefficients. The model has been applied to hourly weather records of 25 years in Hong Kong to generate the hourly cooling loads of buildings with any thermal lag. Sequences of coincident design solar irradiance, dry-bulb and wet-bulb temperatures have been rationally determined through statistical analysis of the computed cooling loads. Results indicate that horizontal solar irradiance computed with the method recommended by ASHRAE is always higher, 4–20%, than the measured value in different months. The peak cooling load resulted from the traditional design weather data is always much higher, 12–50%, than the results from the new design weather data. An unreasonably oversized air-conditioning system would cause high initial cost and unnecessary significant use in embodied energy. It would also deteriorate the part-load energy efficiency and the system management effectiveness.