PhD Studentship: Chaos in motion: reconstructing turbulent flows with physics-informed neural networks

Supervisory Team: Sean Symon & Bharathram Ganapathisubramani

PhD Supervisor: Sean Symon

Project description:

We are seeking a highly motivated PhD candidate to join our research team focused on using advanced machine learning techniques, specifically physics-informed neural networks (PINNs), to reconstruct the time history of turbulent flows from non-time-resolved experimental data. This project will develop and apply PINNs to fill gaps in temporal resolution, transforming sparse experimental measurements into high-fidelity time-resolved flow reconstructions critical for turbulence research.

Physics-informed neural networks (PINNs) are an emerging class of machine learning models that integrate fundamental physical laws directly into the neural network architecture. By embedding governing equations like the incompressible Navier-Stokes equations into the learning process, PINNs enable the reconstruction of complex flow fields with greater accuracy than purely data-driven models, especially when working with sparse or incomplete data. This approach is particularly promising for high-Reynolds number flows, where capturing fine-scale turbulence and temporal dynamics is challenging using traditional experimental techniques. In this project, PINNs will be adapted to overcome these limitations, enabling time-resolved reconstructions from experimental datasets that lack temporal resolution. The PINNs also provide information in regions of the flow where it is difficult to obtain reliable measurements.

Some of the key responsibilities will be to:

• Develop and Implement PINNs: Design and implement PINN models tailored for reconstructing turbulent flows, incorporating physical constraints such as the Navier-Stokes equations and continuity.
• Time History Reconstruction: Generate and evaluate time-resolved flow fields from sparse measurements, validating PINN-based predictions against benchmark simulations.
• Validation and Model Performance Analysis: Assess the accuracy and robustness of reconstructed flow fields across various turbulent regimes.
• Experimental Data Integration: Apply the PINN to experimental data with limited temporal resolution, using the network to infer the missing time evolution and quantities that are not directly measured.

You will be working in the AFM research group which comprises of experts in theoretical, computational and experimental fluid mechanics. We strive to provide an environment in which these different approaches can be combined and focussed on topics of practical importance. You will join a vibrant team of other post-graduate students working in different areas of fluid mechanics ranging from urban flows to canonical turbulent boundary layers. Please visit https://sites.google.com/view/seansymon/home for more information.

Entry Requirements

A very good undergraduate degree (at least a UK 2:1 honours degree, or its international equivalent).

Closing date: 31 January 2025. 

Applications will be considered in the order that they are received, the position will be considered filled when a suitable candidate has been identified.

Funding: We offer a range of funding opportunities for both UK and international students, including Bursaries and Scholarships.

For more information please visit PhD Scholarships | Doctoral College | University of Southampton

Funding will be awarded on a rolling basis, so apply early for the best opportunity to be considered.

How To Apply

Apply online, by clicking the 'Apply' button, above. 

Select programme type (Research), 2025/26, Faculty of Engineering and Physical Sciences, next page select “PhD Engineering & Environment (Full time)”.

In Section 2 of the application form you should insert the name of the supervisor: Sean Symon

Applications should include:

• Research Proposal
• Curriculum Vitae
• Two reference letters
• Degree Transcripts/Certificates to date

For further information please contact: feps-pgr-apply@soton.ac.uk

Advert information