Unsteady aerodynamics of wings in complex environments

The atmospheric boundary layer is never stationary. This is because the wind often changes direction and magnitude, and all these interact with the complex urban landscape. Changing meteorological conditions which could trigger non-stationary phenomena and change urban flow, and turbulence. These have significant impact on aerodynamics of buildings, small vehicle (bikes, UAVs) and dispersion (e.g. point source release of pollutants). In this project, we have identified the understanding of non-stationary phenomena and the
consequent uncertainties as a key to improve the current predictive capability. One case study is efficient prediction for emergency response. Such predictions typically have high uncertainties because details of the source are limited and because that the meteorological conditions are determined from a single nearby observation station or forecast grid point.
This project will exploit recent advances in computational methods and facilities to enable statistical analyses to be conducted which address two of the most challenging issues in CB hazard prediction:
1. How is uncertainty in flow and dispersion prediction affected by changes in the meteorological conditions and simplified building geometries?
2. How should meteorological data be processed (such as by using machine learning) to define the inputs for flow and turbulences, and to improve the prediction of near-source dispersion?
Answers to these are required so that an assessment can be made as to whether current emergency response tools are 'fit-for-purpose', and to define the information and data preparation requirements necessary for accurate real-time high-fidelity dispersion simulations.