Description

One of the most important outstanding questions in fluid dynamics is the study of how and why uniform flow transitions to chaotic motion, or turbulence. Turbulent flow is a vital aspect in the design of airplanes, cars, and boats, but also in climate science and chemical engineering. Unsupervised machine learning will be used to find coherent structures in simulated flow in a pipe. Results will be used to develop a statistical theory of the transition to turbulence. For more details, see the recent article: Allawala, A., Tobias, S. M., & Marston, J. B. (2020). Dimensional reduction of direct statistical simulation. Journal of Fluid Mechanics, 898, 533–18. http://doi.org/10.1017/jfm.2020.382.

Task ideas

• Apply dimensionality reduction techniques to a dataset of fluid simulations, such as proper orthogonal decomposition (POD), principal component analysis (PCA), or autoencoder neural networks. Determine the most important structures in predicting the onset of turbulence.

Expected results

• Comparison between different dimensionality reduction techniques, with investigation of the numerical and physical validity of each.

Requirements

Programming experience with either Python, Julia, or Swift.

Mentors

• Brad Marston (Brown University)
• Steven Tobias (University of Leeds)
• Chenyu Zhang (Brown University)

Please DO NOT contact mentors directly by email, and instead please send project inquiries to ml4-sci@cern.ch with Project Title in the subject and relevant mentors will get in touch with you.

Corresponding Project

• TURBULENT

Participating Organizations

• Brown
• Leeds

---