Two-phase flows can be found in many industrial applications, including nuclear power plants, cooling of electronics and heat transfer devices in spacecraft. In order to provide better design guidelines aiming at higher efficient and safer operation, the fundamental understanding of two-phase flows is required. At STMF, we investigate the physics of two-phase flows by performing experiments with and without phase change (bubble growth in nucleate boiling, Taylor bubble motion in a capillary, bubbly flow) employing high-resolution and high-speed optical diagnostics such as shadowgraphy, particle image velocimetry, interferometry, laser-induced fluorescence and infrared thermography. The shape of these bubbles can be complex due to coalescence phenomenon and/or even partially obstructed due to an overlapping of bubbles on the image, for instance, which makes the classical post-processing techniques quite time-consuming and limited. This internship aims at using the artificial intelligence as a tool to aid the post-processing analysis in two-phase flows. Within this framework, the student will perform the following tasks:

·         Perform a preliminary literature review on artificial intelligence methodologies applied in two-phase flows.

·         Establish a case testing scenario.

·         Test different methodologies using synthetic images.

·         Interpret the results.

·         Present and discuss the work during technical meetings and write a final report.

The candidate is a master student who has a good knowledge of applied mathematics with some experience on neural networks. Knowledge on fluid-dynamics is a plus. Knowledge on Python and or Matlab is desired. A good knowledge of English is required. French would be a plus.