The Centre of Urban Studies and Urban Planning is pleased to invite you to attend TWO RPG Research Seminars on 30 November 2022, (Wednesday) via Zoom as follows:
Please login 15 minutes before the session starts.
*Pre-registration required, zoom link will be provided upon successful registration. [Deadline: 29 November 2022 (Tuesday) 12:00 p.m. (HKT)]
A market-based approach to providing volumetric urban design around metro stations in China
Mr Dongsheng HE
PhD Student, Department of Urban Planning and Design, The University of Hong Kong
Metro systems have been widely advocated in high-density Asian cities. In the process of developing metro systems, transit-oriented development (TOD) is a commonly used planning concept that focuses on integrating transit systems with nearby land use patterns. However, the project implementation may not always follow the advocated design principles due to different barriers. “Market approach” in this thesis is defined that developers provides public goods (e.g., metro access and broader urban design elements) and private goods (e.g., housing property) together and sells them as a bundle, and the charge level is determined by supply and demand of different market actors. Three cities in China (Hong Kong, Shenzhen, and Guangzhou) were selected as the case study area due to the rich TOD practice and comparability of different cities. This thesis will provide institutional design for creating volumetric urban design around station areas in a market manner in Chinese cities.
A hybrid deep learning framework for urban building energy modeling and estimation
Mr Zheng LI
MPhil Student, Department of Urban Planning and Design, The University of Hong Kong
Buildings account for about 36% of global total energy demand, which makes them have great energy-saving potential. To help formulate building energy conservation strategies, many studies have been conducted to support the design of building energy-efficiency. The tasks of urban building energy estimation and post-analysis are critical for quantifying and assessing energy-saving measures. However, current city-scale studies are ineffective to capture the interdependency of inter-building and building-urban environment due to their high-dynamic and non-linear characteristics. It may influence the accuracy of building energy estimation. To bridge the main research gap, the study proposes a hybrid deep learning framework combining physical and data-driven approaches. The framework can 1) capture the interdependency by graph attention networks, 2) use hybrid models with micro and macro variables to estimate building energy consumption. Then we can reveal the hidden relationships between influential factors and energy usage and conduct energy-saving potential analysis, which can provide strategic guidance for decision-makers.
~~ ALL INTERESTED ARE WELCOME ~~
Enquiries: 3917 2721
CENTRE OF URBAN STUDIES AND URBAN PLANNING
THE UNIVERSITY OF HONG KONG