Data-driven method for flow and scalar dispersion past realistic canopies

Wangsawijaya, Dea Daniella, Jalalabadi, Razieh, Zhdanov, Oleksandr and Zang, Bin (2024) Data-driven method for flow and scalar dispersion past realistic canopies. National Fellowship in Fluid Dynamics (NFFDy) Summer School: Data in Fluids, Department of Engineering, Cambridge University, Cambridge, United Kingdom. 10 Jul - 18 Aug 2023. pp. 121-132 .

Abstract

Canopy flow encompasses a large number of phenomena in nature and engineering applications. Studies involving such canopies are limited by the wide range of lengthscale (from vegetation to tall buildings and wind farms) and ability to represent a realistic canopy model. In this study, we combine experimental and numerical simulation data of a realistic canopy model, utilising and testing various data-driven methods to gain, reconstruct, and predict unknown information regarding the flow.

Present results suggest that the tested data-driven methods successfully predict the low-rank dynamics of the flow to a varying degree, with some limitations regarding the complexity of the flow, the availability of a priori knowledge of the flow, and number of training data.

This record has no associated files available for download.

Contributors

Author: Dea Daniella Wangsawijaya

Download statistics