Two PhD positions in engineering at Université Libre de Bruxelles (ATM-BURN) and UC Louvain (IMMC) (# of pos: 2)
Universit libre de Bruxelles ULB
State University, Browse similar opportunities
en
State University, Browse similar opportunities
“METRIC – A Multi-fidEliTy framework for the development of Robust dIgital twins of Combustion systems”
The production of renewable electro- or biofuels from excess wind and solar power is a very attractive opportunity to create energy-dense carriers with neutral or even negative carbon balance. These carriers will be converted in advanced combustion technologies able to prevent harmful emissions while assuring high energy efficiency, regardless of the fuel(s) used. Thus, the role of combustion technologies will become even more crucial in the future.
Combustion devices are characterized by vast ranges of time and length scales, phase transitions, bifurcations, and multiple interacting physical phenomena that are extremely complex to predict. High-fidelity simulations are still prohibitive for realistic systems and limited to simple canonical problems: this is mainly due to the size of chemical mechanisms, which consist of hundreds of species and thousands of reactions, even for simple fuels.
Reduced-Order Models (ROMs) are required in a number of applications, including combustion, where fast responses are needed to predict the state of a complex system. Physics-based ROM are gaining interest, to embed critical aspects of detailed simulations and experiments into simplified relationships between the inputs and outputs that can be used in real time. The development of such virtual models, also referred to as digital twins, opens up a number of opportunities, such as the use of data to anticipate the response of a system and brainstorm malfunctioning, and the use of simulations to develop new technologies, i.e., virtual prototyping. However, ROMs are neither robust to parameter changes, nor are they cheap to build. Extending their applicability for a wide range of parameter variations requires a large number of computationally expensive models, while making them robust requires appropriate techniques for feature extraction, classification and regression.
Within the METRIC project, we propose a multi-fidelity framework combining numerical simulations of different accuracy, adaptive chemistry approaches, modal analysis and non-linear regression for the development of robust digital twins of realistic combustion systems. The focus is put on innovative combustion technologies, able to deliver high-combustion efficiencies with very low pollutant emissions, such as flameless and MILD combustion.
We seek two candidates (PhD level) with a background in one of the following areas:
1. Turbulence chemistry interactions & combustion modelling. Expertise in Reynolds-Averaged Navier Stokes simulations, Large Eddy Simulation, turbulence modelling, turbulence-chemistry interaction modelling.
2. Model reduction and feature extraction. Expertise in machine learning techniques, Principal Component Analysis/Proper Orthogonal Decomposition, Vector quantization methods, Neural networks, t-distributed Stochastic Neighbor Embedding.
3. Optimization and Uncertainty Quantification (with application to turbulent flows). Expertise in surrogate model generation (e.g. using polynomial chaos expansion and Gaussian Process/Kriging), parameter estimation using deterministic and Bayesian approaches, data assimilation, …
The positions are readily available. The choice will be based on the candidates’ profiles. The duration of the PhD is 4 years.
Description of the team and the environment
Two PhD positions are available, one at ULB and one at UCL.
Aero-Thermo-Mechanics Department of the Faculty of Applied Sciences, Brussels School of Engineering, Université Libre de Bruxelles (ULB).
The department is composed of four professors and approximately 40 researchers. The department is active in many Belgian and European research projects with strong national and international collaborations. The promoter of the project is Professor Alessandro Parente. Prof. Parente’s research activity includes turbulent/chemistry interaction in turbulent combustion and reduced-order models; non-conventional fuels (hydrogen ammonia and other solar fuels) and pollutant formation; novel combustion technologies, e.g. MILD combustion; numerical simulation of atmospheric boundary layer flows; and verification, validation and uncertainty quantification in computational fluid dynamics.
Professor Parente has authored more than 80 journal papers and 2 patentw. In January 2015, Prof. Parente founded the BURN group (http://burn-research.be ). The group involves 7 full time professors and around 40 researchers between ULB and VUB and aims at developing a world-class research group in combustion simulations and experimental investigations. This project is aligned with the research goals of the BURN group and will complement work currently be undertaken by ULB and partners.
More on ULB: https://www.ulb.be/en/ulb-homepage
Institute of Mechanics, Materials, and Civil Engineering (iMMC) of the Université catholique de Louvain (UCLouvain)
Leading international center for research in engineering. Its core strengths span several of the major engineering disciplines encompassing energy, thermodynamics, chemical and environmental engineering, materials and processes, structural engineering, geomechanics, manufacturing, fluid mechanics, mechatronics, robotics, biomechanics, numerical and computational sciences. Prof. Francesco Contino focuses his research effort on CFD simulations of internal combustion engines, pollutant emissions of vehicles, and the use of non-conventional fuels. He has developed chemistry reduction methods that enable the use of more detailed mechanisms in CFD simulations, hence providing a better description of the combustion. His expertise also extends to UQ and robust optimization.
More on UCLouvain: https://uclouvain.be/en/index.html
Choose the country you wish to travel to study for free, work or volunteer
