Projects
Projects
Embedded Files
Data-driven energy optimization for supporting the design process of developing net-zero energy communities (SoftwareEcom)
[2024. 7~2027.3]: Funded by Korea Agency for Infrastructure Technology Advancement (KAIA)
Data-driven energy optimization for supporting the design process of developing net-zero energy communities (SoftwareEcom)
[2024. 7~2027.3]: Funded by Korea Agency for Infrastructure Technology Advancement (KAIA)
A cross-nation (Korea-United Kindoms) consortium-based research project aiming to develop a software suite to design and operate net-zero energy buildings at a community scale.
Our group's objectives are to: i) develop a computationally efficient simulation method to estimate the solar harvesting potential of photovoltaic panels either attached or integrated into the buildings, at a community or urban scale; ii) experimentally validate the simulation method developed, based on public datasets; and iii) develop a time-resolved design optimization algorithm for BAPV and BIPV design at a community scale, that maximizes the energy autonomy of the communities.
Development of an Ultra-Fast Urban Solar Irradiance Prediction Simulation Method Based on Augmented Urban Digital Twins and Physics-Informed Neural Networks
[2025.9~Present] : Funded by
Development of an Ultra-Fast Urban Solar Irradiance Prediction Simulation Method Based on Augmented Urban Digital Twins and Physics-Informed Neural Networks
[2025.9~Present] : Funded by
The ultimate objective of this project is to develop a model that supports the design and operation of net-zero energy buildings at the community scale, thereby enabling informed decision-making regarding the urban applicability of photovoltaic systems.. The specific objectives are as follows: (1) to preprocess and database public GIS data and construct foundational datasets for urban-scale machine learning and simulation; (2) use generative AI technics to massively augment realistic 3D urban models; and (3) to develop a physics-informed neural network (PINN) surrogate model that embeds radiative transfer constraints into the loss function, enabling fast prediction of urban-scale solar irradiance and photovoltaic potential, and to experimentally validate the proposed approach.
Page updated
Google Sites
Report abuse