Deadline:
As soon as possible
Location(s)
France
Location : The role is based in Massy, a town in the greater Parisian region of France.
Overview
Details
What will you learn?
The world beneath our feet is incredibly complex. It is also the source of natural resources fundamental to the functioning of our society. In Viridien, we use knowledge of signal processing, wave propagation, numerical optimization and AI to develop and test high-end algorithms, sequences, or workflows with predefined geophysical or engineering content for our clients.
During your internship, you will be working in the subsurface imaging R&D team, in close collaboration with our software and production departments, to meet their demands and industry trends.
You will have the opportunity to engage in cutting-edge research on reservoir-oriented full-waveform inversion (FWI). In Viridien, FWI has been widely used for velocity model building and seismic imaging in complex geological contexts around the world to accurately construct highly detailed, data-driven physical medium parameters of subsurface. Until recently, most FWI industrial applications have relied on acoustic-approximation-based modeling to reduce computational costs. A step forward from acoustic modeling to elastic modeling allows FWI to interpret elastic phenomena in the seismic data.
This approach not only enhances P-wave–based velocity model building and seismic imaging beneath complex overburdens, but also provides an additional S-wave velocity model, which is highly valuable for distinguishing fluid types (water, oil, and gas) and monitoring fluid movements during hydrocarbon reservoir characterization.
During this internship, you will investigate the potential of using neural networks to incorporate petrophysical information for guiding full-waveform inversion (FWI) from both methodology and development perspectives. The main objectives are as follows:
- Develop a neural network model to map petrophysical parameters (porosity, lithology, saturation) to elastic parameters (Vp, Vs, density, …)
- Integrate the neural network mapping into the FWI framework, allowing the inversion to be guided by petrophysical information
Opportunity is About
Eligibility
Candidates should be from:
Description of Ideal Candidate
Qualifications
- Last year of Masters degree in any of the following disciplines: Mathematics, Physics, Computer science, Artificial intelligence, Data science, Natural Sciences with a Mathematics/Physics specialization.
Key Skills And Experiences
- Solid understanding of machine learning algorithms and neural network architecture
- Excellent coding skills with one or more programming languages, such as Python, C/C++, CUDA, FORTRAN
- Exceptional analytical and problem-solving skills
- Knowledge of basic geophysical and wave-propagation modeling concepts is preferred but not mandatory
- Highly motivated and eager to learn
- Fluent English (read, written, spoken)
Dates
Deadline: As soon as possible
Cost/funding for participants
Un-save Opportunity
Internships, scholarships, student conferences and competitions.
