SEED Research topics

Exploring Realistic Skin Numerical Models in THz Spectroscopy

ID: SEED-PhD-2-skin-models PDF version

This project aims to develop an EM solver for highly realistic skin models. The simulation tool will combine a full- wave skin model together with a physical optics-based model of the spectrometer.

Contributions to a Distributed ORAN-based Digital Twin for Non-Terrestrial Networks (NTNs)

ID: SEED-PhD-3-twin-ntn PDF version

Digital twins (DTs) for non-terrestrial network models can utilize tools from optimization theory, game theory, and artificial intelligence (AI) to aid in instantaneous monitoring, optimization, and control. These capabilities can significantly enhance the design and management of these complex networks, while allowing for proactive identification of technical issues. DTs can also function as robust research and development (R&D) platforms, supporting the creation of innovative network architectures and standards.

Development of a Comprehensive 3D Statistical Model of Fetal Facial Growth Patterns Across Gestational Ages using MRI Data

ID: SEED-PhD-4-fetal-model PDF version

This project addresses key challenges in developing a 3D morphable model of prenatal faces. Accurately representing the evolving facial morphology across different gestational ages using MRI data is a primary challenge. Extending this model to 3D ultrasound introduces difficulties such as lower soft-tissue contrast, higher noise, and motion artifacts.

Study and optimisation of heat transfer in an isothermal liquid piston compressed air energy storage system, including the age-related physical changes in the system

ID: SEED-PhD-5-heat-transfer PDF version

The ongoing development of isothermal liquid piston compressors has led to the optimization of the internal heat exchange with complex and very thin regenerative heat exchangers in the compression chamber. In contact with moving water and atmospheric air, these exchangers are subject to many age-related physical changes (fouling, clogging, corrosion, physical deformation, erosion…). This thesis aims at optimizing the heat exchanger design for the whole life of the system considering these physical modifications.

3D Vision and Cybersickness in Virtual Reality

ID: SEED-PhD-6-vision-vr PDF version

The mismatch between natural vision and visual simulation in VR causes a mismatch between convergence and visual accommodation, which can cause eye fatigue and cybersickness. In this context, eye tracking could be a solution, but current devices do not yet offer sufficient speed and precision. The aim of this thesis is to better understand how attention is projected in 3D, to better predict and mitigate motion sickness and to improve the use of eye tracking in VR.

End-to-End Deep Learning for Ocean Forecasting and Reanalysis Systems

ID: SEED-PhD-7-ocean-forecasting PDF version

This topic aims to combine deep learning, data assimilation and uncertainty quantification methods for forecasting and reanalysis performance as well as the representation and reduction of uncertainties for poorly-observed ocean processes.

Study of the resuspension phenomenon of realistic microparticles under accelerated airflow in HVAC systems: combined numerical – experimental approach

ID: SEED-PhD-8-resuspension-microparticles PDF version

This topic addresses the study of particle resuspension induced by airflow in HVAC systems, focusing on fan start situations, i.e., accelerated airflow, as resuspension mainly occurs during such transient events. The goal is to adapt two existing models to new particle deposits used in experiments and validate the assumptions with more complex morphologies, aiming to achieve more accurate and realistic models of the resuspension phenomenon.