Ongoing research projects

IMMC

Ongoing research projects in iMMC (June 2022)


This a short description of research projects which are presently under progress in iMMC.
Hereunder, you may select one research direction or choose to apply another filter:

Biomedical engineering

Computational science

Civil and environmental engineering

Dynamical and electromechanical systems

Energy

Fluid mechanics

Processing and characterisation of materials

Chemical engineering

Solid mechanics


Research direction:
Listed keyword:
Other keyword:
Division:
Supervisor:

List of projects related to: machine learning




AI-based control policies towards efficient collective behaviours of flow agents and their application to fish schooling
Researcher: Denis Dumoulin
Supervisor(s): Philippe Chatelain

The principal objective is to shed light on mechanisms allowing anguiliform swimmers to swim very efficiently either on their own or in group.
Simulations rely on an unsteady panel method with vortex shedding and on reinforcement learning.



Reconciling engineering models of ejectors with experiments and CFD using physics-informed machine learning
Researcher: Jan Van den Berghe
Supervisor(s): Yann Bartosiewicz

Although Computational Fluid Dynamics (CFD) have proven sufficiently accurate to analyse the complex flow fields in ejectors, their computational cost remains too high to drive the design and optimization phases at system scale operation.


In that case, Lumped Parametric Models (LPM) are vastly preferable because of their lower computational cost. LPM models are based on integral balances and usually 0D formulations from isentropic gas dynamics. However, the lumping of complex physical phenomena such as turbulent mixing, oblique shock patterns and shock-boundary layer interaction into LPMs requires several closure parameters such as isentropic efficiencies. Moreover, LPMs are designed to predict global quantities such as the entrainment ratio or efficiency but are unaware of the stream-wise evolution of local parameters such as velocity, pressure, or Mach number, which only CFD can access. Finally, none of the LPMs presented in the literature consider the problem of modelling ejectors in transient conditions, which can be of primary importance at system scale operation.


This thesis aims to bridge the gap between classic LPMs and CFD approaches by introducing a new family of self-calibrating 1D models that use physics informed machine learning. In particular, the proposed models will combine 1D and unsteady gas dynamics with closure parameters that depend on local variables and properties. The functional relation linking closure models and flow parameters will be encoded in the form of Artificial Neural Networks (ANNs), and their calibration (i.e., the training of the ANN) will be automatic and online, i.e., while data is progressively collected. The model and the automated calibration procedure will be tested on experimental and numerical data. Hence the machine learning techniques will be used here to help the physical model to be closed, i.e., where our knowledge of the governing equations reaches its limit to derive universal relations for exchange of mass, momentum and energy in complex situations.



The Impact of Energy Policies on the Energy System
Researcher: Panagiotis Varelas
Supervisor(s): Francesco Contino

There are many difficulties in the upcoming years toward a carbon-neutral planet. The main issue that renewable energy sources are facing in the energy transformation is the security of supply. Alternative fuels like hydrogen and ammonia might be a solution to this challenge as possible energy carriers. The main barrier they are currently facing is the increased cost compared to other fuels. This project is aiming to promote alternative fuels and green applications in mobility and other sectors and to highlight the importance of alternative fuels as a solution to tackle climate change and reach our goals.









In this direction, academia should play a vital role by providing the necessary facts and guidelines to the policymakers. Currently, researchers and policymakers are acting independently, and coordinating actions are very often lacking. The main goal of this research is to provide evidence and facts that can help policymakers take decisions. This implies measuring the impact of the energy policies on energy systems. Thus, every time an energy policy will be implemented or modified our model will be able to predict the potential impact both for investors and consumers.









Existing models are not considering the effect of the energy policies on the energy systems. Traditionally these models are scenario-oriented. According to their input parameters such as demand, consumption, and available technologies, they can suggest a couple of different scenarios to the policymakers.









Modern societies are rapidly changing by the decisions of political institutions. Recent major events in Europe and United States proved that energy transition is a dynamic condition affected by a plethora of different parameters that should be considered. Designing energy models needs to be a vice versa task proposing the optimal scenarios to the policy makers, considering at the same time the impact of energy policies.









The second novelty of this research proposal is that it will take into consideration both the financial and the societal impact of energy policies.