NGUYEN Frédéric

Professeur ordinaire

NGUYEN Frédéric

Géophysique appliquée
Faculté des Sciences appliquées
Vice-doyen à la recherche (Sciences appliquées)
Urban and Environmental Engineering

ULiège address
Bât. B52/3 Géophysique appliquée
Quartier Polytech 1
allée de la Découverte 9
4000 Liège 1
Belgique
Local
-1/523
ULiège phone number
+32 4 3663797
ULiège Fax
+32 4 3669520
Email
Captcha : Put the image in the right direction
captchaImage

Conseil sectoriel à la recherche et à la valorisation
Sciences et Techniques
Personal website (s)
Applied Geophysics Research Unit (to be updated)

University degrees
2005 : Docteur en sciences appliquées (Université de Liège)

Biography

The candidate develops research in applied geophysics both from the point of view of methods (technological aspect) and the field of application (contribution of new knowledge). Geophysics consists of studying the subsoil by sending for example seismic waves and analyzing the response received by sensors usually positioned on the surface. Unlike medical imaging, geophysical imaging is limited in resolution and is subject to uncertainty. This uncertainty is inherent either to the physical principle or to a mathematical point of view. It is therefore essential to (1) improve the imaging algorithms and to be aware of the uncertainties; (2) be diagnostic of the phenomena studied; and (3) integrate multiple data sets into a unified context. To improve the imaging, the candidate and his team developed algorithms that integrated independent data, e.g., measured in drilling or observed along rock faces. This resulted in significantly more realistic images. They also proposed image quality indicators and innovative data processing. Secondly, the candidate and his team demonstrated that the temporal dimension of geophysical observations contributes to obtain diagnostic and -relatively- unambiguous information in the study of dynamic processes. For example, the candidate and his team have shown the existence of a seasonal geophysical signature related to the degradation of diesel in the subsurface, which could improve the efficiency of in-situ pollution control techniques. Finally, subsurface systems are by nature difficult to predict either in terms of geometry or evolution over time. It may be more useful to quantify the uncertainties in the predictions rather than provide a single wrong answer. To do this, the candidate and his team have developed a methodology where a large number of models capable of simulating a desired response and observational data are generated in order to learn a direct relationship between the observables and the response of interest. Once this simulation work is done, this framework makes uncertainty assessment easier and faster. The results of this research will allow significant advances in a large number of fields.

Recent publications: 

Integrated methodology to link geochemical and geophysical-lab data in a geophysical investigation of a slag heap for resource quantification

Iterative Prior Resampling and rejection sampling to improve 1D geophysical imaging based on Bayesian Evidential Learning (BEL1D)

Geophysical Inversion Using a Variational Autoencoder to Model an Assembled Spatial Prior Uncertainty

 

Duties or mandates

  • Vice-Dean for Research, Engineering School
  • Head of Research, Urban and Environmental Engineering Department (UEE), 2018-2021

ULiège Course

Prospection géophysique, 26h Th, 20h Pr, 5j T. t., 20h Proj., NGUYEN Frédéric

Site investigation, 5h Th, 40j Proj., 40h Labo., 5j T. t., BROUYÈRE Serge, NGUYEN Frédéric

Project in inverse modelling : from field to algorithms, 5h Th, 40h Pr, 30h Proj., 4j T. t., NGUYEN Frédéric

Integrated project on hydrogeophysics, 4h Th, 38h Pr, NGUYEN Frédéric

Introduction aux méthodes numériques et projet, 10h Th, 2h Labo., 28h Proj., BRULS Olivier, LOUVEAUX Quentin, NGUYEN Frédéric

ULiège Outside Course

()