Diurnal and seasonal predictability of envelope pressures driving natural infiltration in residential buildings

Abstract

This study examines the dynamics of residential building envelope pressures by predicting and comparing time series site-specific weather conditions at minute-level resolution. Utilizing theoretically established relationships of both stack and wind effects, this research examines the predictability and accuracy of envelope pressures under different weather conditions. When high wind effects are removed, the Mean Absolute Error (MAE) in stack pressure predictions are minimized, typically falling below 0.24 Pa. The use of airport weather data, even after correcting for height difference and terrain, was found to be inconducive to prediction, highlighting the preference for site-specific measurements to enhance prediction accuracy. This research utilizes minute-level data for real-time environmental monitoring, aiming to inform pressurization or integrate predictive models for dynamic indoor air quality management. The findings contribute to the field by offering a practical approach to measuring and predicting residential air exchange rates, providing insights that could lead to improved health outcomes and energy efficiency in homes.