Conference Agenda

Movement in window views impacts the indoor experience and comfort of building occupants. Some building standards therefore stipulate the presence of dynamic content as a key window view quality evaluation criterion. However, there is a scarcity of tools for evaluation of dynamic window views, owing to the complex interactions between elements in the views. Computing the compositional ratios of view elements vis-à-vis the amount of movement demands an integrated methodological framework. This paper therefore addresses the insufficiency of existing literature on dynamic window view evaluation tools and methods. A framework was developed using the DeepLabV3 semantic segmentation architecture pre-trained on the Cityscapes dataset, for decomposition of dynamic content in fifty recorded urban window views. Movement within the views was calculated using functions contained in the OpenCV library. The pre-trained model yielded accurate predictions of twenty Cityscapes urban object classes, thus facilitating calculation of compositional ratios in the window views. A case study was also discussed to illustrate the practical application of the framework in determination of preferred amount of movement in office window views. This study affirms the suitability of semantic segmentation as a dynamic window view content evaluation tool.

HVAC systems are a key focus of demand response initiatives due to their high energy consumption and control flexibility. However, most studies examined this issue from a grid-operator perspective, while largely overlooking the end-user perspectives on thermal comfort and focusing on demand reduction, optimization strategy, and energy saving. This study addresses this gap by employing Computational Fluid Dynamics (CFD) simulations to assess thermal comfort under global temperature adjustments in a demand response context. Two representative building ventilation strategies—mixing and displacement—and three levels of internal load intensity (low, medium, and high) are examined for a simulated small office environment based on the Department of Energy (DOE) reference building. Thermal comfort is evaluated using Predicted Mean Votes (PMV), draft risk, vertical temperature difference during both response and recovery periods. Results reveal that displacement ventilation enables faster temperature adjustments in the ASHRAE breathing zone across varying levels of internal load, as compared to mixing ventilation. Specifically, the rate of temperature change was 31% to 54% faster during the response period and 32% to 41% faster during the recovery. Furthermore, PMVs were found to decrease more swiftly under displacement ventilation as internal loads increased, indicating higher potential for energy savings. However, the study also found that displacement ventilation is particularly vulnerable to changes in supply air temperature, leading to more fluctuations in draft risk and vertical temperature difference.

This paper explores the benefits of incorporating natural ventilation (NV) simulation into a generative process of designing residential buildings to improve energy efficiency and indoor thermal comfort. Our proposed workflow uses the Wave Function Collapse algorithm to generate a diverse set of floor plans. It also includes post-COVID occupant presence models while incorporating adaptive comfort models. We conduct four sets of experiments using the workflow, and the simulated results suggest that multi-mode cooling strategies combining conventional air conditioning with NV can often significantly reduce energy use, about 18% to 40%, while introducing slight reductions in thermal comfort. This workflow could be beneficial for architects and other stakeholders to generate diverse design options and make decisions based on NV performance.

Personalized cooling and ventilation is a proven effective air distribution strategy to provide individual comfort conditions. Many studies explored the personalized ventilation performance levels with different HVAC systems and design strategies, however, when it comes to India, where there are different climates in tropical region, very few studies have reported personalized conditioning effects from fields. Thus, our present study explores the preliminary level subjective responses to personalized conditioning under three different setpoint temperatures 24°C, 26°C, and 28°C and flowrates between 20 l/s to 34 l/s in combination with ceiling fan airflow. In this paper, we aim to evaluate the occupants’ perception and acceptance to the thermal environment under real field conditions.