79th EAGE Conference & Exhibition 2017
Energy, Technology, Sustainability - Time to open a new Chapter
Workshop 1: Data Science for Geosciences
Monday 12 June*
|Title:||Data Science for Geosciences|
C. Agut (Total)
Today, geoscientists have to handle a huge variety of data on a wide range of scales. These data can be derived from measurements in the lab, in wells, at the surface, from satellite… or they can result from different geoscience modelling or simulation approaches. Challenges arise in the integration of the multiphysical features of these data, which is required to provide a consistent understanding of the medium under study. Standard approaches relying on the determination of a model aiming to synthetize all these multiphysical multiscale phenomena are quite complex, time consuming and often hampered by difficulties encountered when coupling the different (thermo-, hydro-, geomecha-, geochemical…) mechanisms as well as their limited capacity to cover scales ranging from core to basin scale. Considering the rapid increase of data in volume and variety, development of sustainable infrastructure, as well as clever data management and analysis approaches seem a promising way, enabling to deduce from data themselves patterns yielding a consistent understanding of the medium. The pending question is the potential of statistical or machine learning methods potential, to name a few, for practical geosciences applications.
Tackling this challenge requires to address following aspects:
– Which efficient infrastructure to manage huge amount of data?
– Raw versus representative database? Which level of data curation, quality control and preprocessing provides a significant added value?
– What kind of data analytics? Which “big data” technologies for specific geosciences applications?
– How to interpret these tool outputs? How to calibrate them? Which confidence can we have on the results?
Examples of what can be expected from application on real or synthetic data would be much appreciated, in particularly to advance on following issues:
– Can we/should we compare big data technologies results to standard approach results?
– Can we estimate the added value of a large range of data?
– Can data science approaches reveal hidden patterns that are not explained by standard modelling approaches?
– How to extrapolate outputs derived from a specific database?
|09:15||Keynote: Data Science & Application to Geosciences - An introduction
M. Lutz* (Total)
|10:00||Results from Experiments with the Use of Hadoop and High Speed Computing for Data Management
A. Smith* (Lunchelan), M. Wildig (Ovation Data)
|10:25||IoT-based Wireless Networking for Geoscience Applications
H. Jamali-Rad* (Shell), X. Campman (Shell)
|11:05||Performing Successful Data Science in the Geoscience Domain
D. Irving* (Teradata UK), J. McConnell (Teradata UK)
|14:00||Keynote speaker: Fostering High-impact Machine Learning Ecosystem in Subsurface Science and Engineering
M. Hall* (Agile Geoscience)
|14:45||Deep Learning on Hyperspectral Data for Land Use and Vegetation Mapping
N. Audebert* (ONERA), B. Le Saux (ONERA), S. Lefèvre, C. Taillandier, D. Dubucq
|15:50||Machine Learning can extract the Information Needed for Modelling and Data analysing from Unstructured Documents
H. Blondelle* (Agile Data Decisions), A. Juneja (Agile Data Decisions), J. Micaelli (Agile Data Decisions), P. Neri (Agile Data Decisions)
|16:15||Unsupervised Identification of Electrofacies Employing Machine Learning
I. Emelyanova* (CSIRO Energy), M. Pervukhina (CSIRO Energy), M. Clennell (CSIRO Energy), C. Dyt (CSIRO Energy)
|16:40||Machine Learning based Workflows In Exploration and Production
J. Limbeck* (Shell), M. Araya (Shell), G. Joosten (Shell), A. Eales (Shell), P. Gelderblom (Shell), D. Hohl (Shell)
|11:30-12:30||Automated Facies Prediction in Drillholes using Machine Learning
M. Blouin* (INRS-ETE), A. Caté (INRS-ETE), L. Perozzi (INRS-ETE), E. Gloaguen (INRS-ETE)
|Optimising Storage for High-speed Data access to Large Volumes of Data - Recent advances and future direction
A. Smith* (Lunchelan), M. Wildig (Ovation Data)
|ForM@Ter: Adata and services centre for Solid Earth
E. Ostanciaux* (IPGP), M. Mandea (CNES), M. Diament (IPGP), O. Jamet (IGN)
|Technical Descriptions in Long-term 115⁰C Borehole Digital Micro-seismic Monitoring at the PTRC Aquistore CO2 Sequestration Project
C. Nixon (University of Alberta), D. Schmitt* (University of Alberta), R. Kofman (University of Alberta), D. White (Geological Survey of Alberta),et al.
|Sensitivity Analysis of Synthetic Seismograms in Sedimentary Basin with Respect to Uncertain Seismological Parameters
F. de Martin* (BRGM French Geological Survey), P. Thierry (Intel), D. Keyes (KAUST), E. Chaljub (ISTerre) et al.
|Automatic Similarity Measures to Manage Geoscience Databases
|Channel Characterization Using Support Vector Machine
A. Mardan* (Amirkabir Univeristy of Technology), A. Javaherian (Amirkabir Univeristy of Technology), M. Mirzakhanian (NIOC)
|Carbonate Reservoir Cementation Factor Modeling Using Wireline Logs and Artificial Intelligence Methodology
F. Anifowose* (Saudi Aramco), C. Ayadiuno (Saudi Aramco), F. Rashedian (Saudi Aramco)
|Adjusting Well Plan Trajectory through 3-D Seismic Litho-fluid Classification - A case study
M. Delnava* (Pars Petro Zagros Geophysics), K. Kazemi (Pars Petro Zagros Geophysics)
|Subsurface Integrity Management of UGS
F. Favret (EDF), R. Del Potro* (Geostock), M. Diez (University of Bristol)
|Accelerated Data Handling using Real-time Data Compression and Decompression for Data Sciences and Geoscience
D. Buch* (Headwaves), M. Vinther (Headwaves)
The Unstructured Data Challenge in E&P
|A Benchmark of Lossy Compression Algorithms in the Context of Geophysics
M. Cuif Sjöstrand* (Total), M. Zmudz (Total), N. Six (Total), H. Puntous (Total)