|Location||Houston, United States|
|Date Posted||January 1, 2018|
The main aim of this VIE assignment will be to help develop and implement preconditioning tools and methods for seismic FWI in order to accelerate and improve its convergence and thereby reduce the computational cost. The VIE will work closely with R&T; staff who have developed prototypes of preconditioning software. The project will be structured as follows:
Phase 1: initiation to FWI and preconditioning: R&T; staff will introduce the VIE to the concepts of FWI with emphasis on the computational and numerical challenges and will then guide him/her through a bibliography review to develop an understanding of the theory and practice of FWI and the numerical issues and problems with its use.
Phase 2: Familiarization with FWI codes: the VIE will develop an understanding of the content, structure and performance of our existing FWI codes by examining the source, running basic tests on synthetic data and undertaking some minor development exercises within them. This should reinforce the lessons learnt in phase 1 and add a practical understanding of FWI and how its algorithms are translated into codes that run on real, industrial datasets.
Phase 3: Code implementation and testing: the VIE will work with his mentor to implement the FWI preconditioning algorithms, currently expressed in prototypes, into operational code libraries. (S)he will validate the codes on synthetic data of increasing complexity and then on real data, and will work to improve their performance on the computers available in Houston and Pau. The VIE will then assess the improvement in the inversion results against the computational effort required; (s)he may further consider applying the new codes to other geophysical inverse problems.
The selected candidate will receive a VIE allowance according to the Business France rates and the subsidiary will provide a furnished accommodation (flatshare).
Please check the following website to make sure that your application complies with the criteria to carry out a VIE assignment: www.civiweb.com
Contexte and environment
The theoretical basis underpinning seismic Full Waveform Inversion (FWI) has been known for over 30 years, but it is only in the last 5-7 of these that it has become established as a model-building tool in industrial imaging for oil and gas exploration. This was due to a combination of the need for algorithmic/methodological refinements to deal with the practical problems of FWI, such as the plethora of local minima that afflict most applications, and the computational cost. Indeed the latter consideration still limits industrial applications in terms of both frequency content and physical realism (acoustic instead of visco-elastic simulation, for example), forcing continued methodological compromises.
Desired qualifications : Master or Ph.D. in applied maths, numerical sciences or physics with at least 1 year of profesionnal experience in the field.
Technical skills : good knowledge of numerical methods, inversion and optimisation ; awareness of geophysical imaging and inversion an advantage.
Computing skills : strong programming skills in C, C++ or Fortran ; experience of parallel programming for HPC applications an advantage.
Languages : Fluent English and intermediary French.
Having a Driving license would be preferable.
Geophysics, General R & D Responsibilities
Experience level required
0 - 3 years, 3 - 6 years
Exploration and Production
Paris, La Défense
Proposed start date
About us / company profile
Give your best to better energy and make the commitment with Total. With over 500-plus professions in 130 countries, we offer high safety and environmental standards, strong ethical values, an innovation culture and wide-ranging career development. Be part of the global team whose mission is already shared by 100,000 employees: to make energy better each and every day.