A5055820861- Wind and Air Flow Studies
- Aerodynamics and Fluid Dynamics Research
- Model Reduction and Neural Networks
- Fluid Dynamics and Vibration Analysis
- Fluid Dynamics and Turbulent Flows
- Building Energy and Comfort Optimization
- Aeolian processes and effects
- Noise Effects and Management
- Evacuation and Crowd Dynamics
- Meteorological Phenomena and Simulations
- Infection Control and Ventilation
- Urban Heat Island Mitigation
- Tropical and Extratropical Cyclones Research
- Seismic and Structural Analysis of Tall Buildings
- Aerodynamics and Acoustics in Jet Flows
- Indoor Air Quality and Microbial Exposure
- Fire dynamics and safety research
- Structural Health Monitoring Techniques
- Tree Root and Stability Studies
- Plant Water Relations and Carbon Dynamics
- Atmospheric aerosols and clouds
- Nuclear Engineering Thermal-Hydraulics
- Air Quality and Health Impacts
- Hydraulic flow and structures
- Plant responses to elevated CO2
Hong Kong University of Science and Technology
2013-2023
University of Hong Kong
2013-2023
City University of Hong Kong
2023
Sun Yat-sen University
2021
This study used explainable machine learning (XML), a new branch of Machine Learning (ML), to elucidate how ML models make predictions. Three tree-based regression models, Decision Tree (DT), Random Forest (RF), and Extreme Gradient Boost (XGB), were predict the normalized mean (Cp,mean), fluctuating (Cp,rms), minimum (Cp,min), maximum (Cp,max) external wind pressure coefficients low-rise building with fixed dimensions in urban-like settings for several incidence angles. Two types XML —...
Scientific research and engineering practice often require the modeling decomposition of nonlinear systems. The Dynamic Mode Decomposition (DMD) is a novel Koopman-based technique that effectively dissects high-dimensional systems into periodically distinct constituents on reduced-order subspaces. As mathematical hatchling, DMD bears vast potentials yet an equal degree unknown. This serial effort investigates nuances sampling with engineering-oriented emphasis. Part I aimed at elucidating...
The present work extends the parametric investigation on sampling nuances of dynamic mode decomposition (DMD) under Koopman analysis. Through turbulent wakes, study corroborated generality universal convergence states for all DMD implementations. It discovered implications range and resolution—determinants spectral discretization by discrete bins highest resolved frequency range, respectively. reaffirmed necessity state independence, too. Results also suggested that observables derived from...
This serial work presents a Linear-Time-Invariance (LTI) notion to the Koopman analysis, finding consistent and physically meaningful modes addressing long-standing problem of fluid-structure interactions: deterministically relating fluid structure. Part 1 (Li et al., 2022) developed Koopman-LTI architecture applied it pedagogical prism wake. By systematic procedure, LTI generated sampling-independent linearization that captured all recurring dynamics, six corresponding, orthogonal, in-synch...
This work augments a Linear-Time-Invariance (LTI) notion to the Koopman analysis, finding an invariant subspace on which consistent modes are expanded with fluid mechanics implications. The also develops Koopman-LTI architecture—a systematic procedure associate excitation and structure surface pressure by matching eigen tuples, establishing fluid–structure correspondences that examine interactions (FSIs) at new angles. data-driven, modular architecture exhibits potential evolve advances in...
Wind tunnel tests and computational fluid dynamics (CFD) simulations remain the main modeling techniques in wind engineering despite being expensive, time-consuming, requiring special facilities expert knowledge. There is a clear need for fast, accurate, but, at same time, computationally economical substitute. This study proposes Gaussian Process-based (GP-based) emulator to predict pedestrian-level environment near lift-up building – an isolated, unconventionally configured building. The...