Abstract
This study evaluates for the first time the suitability of typical meteorological year (TMY) weather data for simulating the performance of buildings that are entirely conditioned by ambient energy. A home in Durango, CO, was simulated with TMY data, with real data for 1998–2020 and with extreme meteorological year (XMY) data. For this climate, the indoor temperature in the house designed with TMY data drops below the range of comfortable indoor temperature (20–25 °C) for 16 of 23 years, including as low as 13 °C during 2008. With the thermal time constant of the house adjusted for each data set to maintain comfort, the required time constants for the real data ranged from 1.178 to 7.56 days with a mean of 3.14 and a median of 2.38, while the TMY value was 1.862 for a percentile rank of 0.318. XMY data did not produce significantly better results. The correlation of the time constant to weather parameters showed that the maximum interval during which the 24-h average solar load ratio remains below 1 is a promising index for identifying the most challenging year. Until more representative TMY and XMY weightings are developed for ambient-conditioned buildings across other climates, it is advisable that current TMY data are used only for preliminary design and that multi-year simulations are conducted for final design.
