Design for outdoor thermal comfort : on-site monitoring and modeling of airflow in building surroundings

Abstract

Residents wish to have outdoor spaces to enjoy walking, cycling, and other recreational activities, which are often hindered by unfavorable thermal comfort conditions, particularly in the summer. High building density lowers the average wind speed and this intensifies the urban heat island effect at the urban scale. The conscientious use of building morphology to create a local thermal comfort zone at selected spots in a large precinct is becoming a pressing issue for sustainable urbanization. Previous studies regarding outdoor thermal comfort were limited to either field measurement or wind environment prediction around buildings in the urban mesoscale. Therefore, this thesis provides a microscale measurement in a precinct at the scale of approximately 200 m at first. Then the wind flow prediction around a single building (or the building with elevated design) surrounding is considered because it is normally employed in practice than the complicated features of building clusters in urban areas. Specifically, on-site monitoring and accurate prediction of the isolated building without an elevated design are the prerequisites of the successful prediction of wind environment and thermal comfort. Three sub-works, namely, (1) assessment of local thermal comfort differences in a precinct, (2) quality assessments and improvements of Computational Fluid Dynamics (CFD) predictions of air flow around an isolated building with different turbulence models, and (3) turbulent flow predictions, wind comfort and thermal comfort assessments around the building with and without the elevated design, were conducted to achieve the research objective. A proof of concept study was conducted in advance via continuous monitoring of the pedestrian level winds and thermal parameters at two sample days in summer, which include instantaneous air temperature, globe temperature, wind speed, and relative humidity. Three outdoor locations at a university campus (at the scale of approximately 200 m) are selected, and daytime thermal perceptions at the three sites were evaluated using Physiological Equivalent Temperature (PET). Meanwhile, a new PET-based index is defined, which is called the normalized environmental parameter difference for air temperature, mean radiant temperature, and wind speed. From the analysis of the simultaneous differences of the three parameters among the monitored spots, wind speed and mean radiant temperature differences that result in significant differences in thermal comfort are shown. This pilot study clearly indicates that wind amplification combined with shading effects can generate thermally comfortable conditions in the open ground floor beneath an elevated building, even on a sunny, hot summer day in a subtropical city. Then, the variations of thermal perceptions are assessed for hot summer, temperate autumn, and cool winter in two outdoor sites on a campus via on-site monitoring at pedestrian level winds and thermal parameters at two sample days (sunny and cloudy). The daytime wind directions are also recorded from a nearby urban weather station and used for the analysis of the influences on the differences of wind and thermal comfort between the two survey sites in different seasons. Several predictive methods for the outdoor wind environment were evaluated and compared to identify that CFD is the most appropriate predictive method. The choice of appropriate turbulence models in CFD is still a challenging issue for accurately predicting outdoor microclimate and thermal comfort conditions in urban planning. The performances of the Steady-state Reynolds Averaged Navier-Stokes (SRANS), Large Eddy Simulation (LES), and Detached Eddy Simulation (DES) modeling approaches in simulating the wind flow around an isolated building were compared with that of a benchmark wind tunnel experiment. The effects of the computational parameters were analyzed, including the grid resolution for all cases and the discretization time step ({439}t) and non-dimensional sampling time (t*) for the LES and DES cases. The results of the LES and DES cases are consistent with the experimental results for the leeward and lateral regions in the vertical and horizontal planes. The Delayed Detached Eddy Simulation (DDES) and LES models predict similar results in the building wake region, but the DDES has a lower overall mesh requirement. The DDES model encouragingly provides not only the mean flow field but also the instantaneous wind characteristics, which can be useful for more accurate analysis of outdoor wind and thermal comfort. The outdoor thermal comfort in the urban design stage can potentially enhance the livability of a subtropical city as Hong Kong. This is done by first comparing the CFD simulation results of wind velocities around a single building with and without an elevated design with those obtained from a wind tunnel experiment, and three turbulence modeling approaches, the DDES and the steady-state and unsteady-state RNG k-{464} models were assessed. The effects of grid resolution and inflow fluctuating algorithm are also discussed. The mean velocity field obtained using the DDES model is consistent with the wind tunnel measurements, particularly in the wake region and at the open space beneath the elevated building. The building elevation modified the mean flow pattern around the building. Then the transient wake flows and turbulent flow characteristics around the buildings with and without elevation are provided by instantaneous wind velocity, lift coefficient, and turbulence intensity. The periodical wind flow pattern and the gust wind field around the two building cases are presented as well. The potential impact on pedestrian thermal comfort is predicted using a simplified method by combining the simulated wind velocity and the on-site monitored radiant and air temperatures and air humidity on two summer days. It is revealed that the elevated design improves thermal comfort, but only in the limited neighboring area. However, the open space beneath the elevated building provides better thermal comfort in the summer condition. This work demonstrates that CFD simulation of wind conditions can be used to assess outdoor thermal comfort in the planning stage without being coupled with thermal simulation. These findings contribute to the enhanced understanding of outdoor thermal comfort in a precinct and the improved predictive strategies of the thermal perceptions of pedestrians in the planning stage and further help to alert city planners of additional options available in precinct planning to encourage outdoor activities.