Prediction of internal pressure and door operation risk in high-rise buildings under time-varying external conditions

In high-rise buildings, the “stack effect” occurs when temperature differences between the interior and exterior cause outside air to enter through lower-floor entrances and flow to upper floors via atriums, elevator shafts, and stairwells. This undesirable airflow can result in noise issues in fixtures, difficulties in door operation, and other problems. The adoption of revolving doors at entrances could mitigate these effects, but in Japan, their use is challenging due to past accidents.

Other measures, such as adding fixtures along airflow paths, have been considered; however, excessive additions are undesirable due to usability and cost concerns. Therefore, predicting and visualizing the stack effect during the design phase is essential for balancing risk, usability, and cost while serving as a tool for risk communication between designers and building owners.

For predicting the stack effect, ventilation network calculations are commonly used, which require inputs such as outdoor temperature, wind speed, and wind direction. However, the vast combinations of these conditions often lead to assumptions like setting wind speed to 0 m/s, which may reduce the accuracy of predictions under actual operating conditions.

This study aims to develop a method for predicting the stack effect considering time-varying external winds and temperatures, while minimizing computational load. Additionally, it proposes a new indicator for risk assessment called PER (Prediction Evaluation Ratio) to visualize risks. The results demonstrate that considering varying external conditions provides a more accurate estimation of building failure risks due to the stack effect. The newly proposed PER indicator effectively visualizes these risks, offering a practical tool for design and risk management.