Yun-Hang Cho

Research Project: An integrated numerical and physical modelling approach to enable remote measurement of turbulent free-surface flows.

To develop the science and engineering behind a new ability to infer shallow flow conditions based on remote assessment of free-surface patterns, and to deepen our understanding of turbulent flows. This will be achieved by combining the experimental expertise at Sheffield with the numerical modelling expertise at Institute of High Performance Computing A*Star Singapore.

Objectives:

1. Collect novel dataset of free-surface flows by enhancing an existing laboratory flume with equipment and expertise available in the department.
2. Develop deterministic numerical model, based on existing models in IHPC, for predicting free-surface dynamics from given flow and boundary conditions.
3. Invert the model to enable inference of flow and boundary conditions from measured free-surface data.
4. Blind-test’ the inverted model via new laboratory and field experiments (river; sewage treatment channel), supported by industry partners.

Turbulent flow surface patterns have been shown to contain information regarding the underlying flow conditions [2], but existing relationships are empirical in nature.

This PhD study will test the hypothesis that flow conditions can be directly determined from free-surface measurements. This ability would be a powerful tool for flow measurement and management, with numerous applications in river and sewer monitoring, process plant control and coastal management.

However, it will require a suitable inverse numerical model to relate free-surface behaviour to the turbulent flow field. Such a model does not currently exist, due to a historical lack of (i) understanding of free-surface dynamics, and (ii) ability to collect reliable free-surface data for calibrating and validating models. Recent advances at IHPC and Sheffield mean it is now possible.

An integrated approach, with close collaboration between Sheffield and IHPC will align the numerical and experimental work to enable the bespoke data set necessary to develop and validate an appropriate model.

A detailed planning stage involving both research groups will identify the format, resolution and volume of data required, and a laboratory flume at Sheffield will be enhanced with existing flow measurement equipment in the department to provide it.

Flow field and free-surface data will thereby be recorded for a range of flows. Existing deterministic numerical models in IHPC will be further developed and validated using the experimental data. The numerical model will simulate the flows established in the laboratory, and include the dynamics of the free-surface.

Once a numerical model to recreate the free-surface flows is developed and validated, an inverted model will be derived to enable the inference of turbulent flow fields based on measured free-surface data.

This will be achieved via backward invert modelling or forward modelling with parameter optimisation to minimise the error between the generated free-surface and the measurements. The model will enable the inference of turbulence properties and boundary conditions based on free-surface data, and will be ‘blind tested’ against new conditions in the laboratory.

It will also be tested at two field locations (river; treatment channel) to demonstrate the potential to remotely assess flow conditions in real applications.