ai-driven rock physic to leverage the seismic characterization in exploration context
Objectives
Leading consulting firm in geoscience, guiding client decisions across all energy sectors, supported by research and development conducted by IFPEN Group, Beicip-Franlab is exploring the potential of AI throughout its entire value chain. Historically data-driven and utilizing machine learning techniques, Beicip-Franlab's seismic characterization group is now investigating an innovative approach to reintroduce rock physics into their industrial workflows, leveraging recent breakthroughs in AI.
Main tasks undertaken during the internship
- Catalog building of rock physics templates, mineral and fluid constants.
- Testing AI (MCDA, PINNS…) to automatically optimize parameters and select the best model(s). Discussion about uncertainties.
- Application to build missing logs and detect anomalous values in row data.
- Synthetic AVO, synthetic gathers and angle-stacks. Comparison with real dataset.
- Application to validate inversion results and petro-elastic models. Application to populate training samples for facies prediction. Application to real project data.
- Development of an interactive view (plots vs seismic) into InterWell.
The internship will be supervised by a senior geophysicist and software engineer.
Software used
IDE depending on the language (Anacoda, Visual studio, Netbeans…) as well as specialist software InterWell.Final year student enrolled in a master's degree program with geosciences skills. Fast learner, rigorous and eagerness to understand and master technical software. Knowledge about rock physics & programming (Python and/or C++ and/or Java).