WS Monitoramento Geofísico de Barragens

Programa Técnico

Day 1

Session 1: Application to Reservoir Characterization

Start Duration Presentation Title Speaker Affiliation
8:00 AM 0:25:00 Integration of Deep Learning and Physics-based techniques to drive reservoir characterization in Brazil Diego Hernandez
Bernard Laugier
8:25 AM 0:25:00 Propagation of Well-log Data Imputation Uncertainties on the Kriging's Variance of a 3D Petrophysical Map using Epistemic Kernels Jorge Luis Guevara Diaz IBM Research
8:50 AM 0:25:00 Multi-azimuthal pre-stack seismic facies prediction via deep convolutional autoencoders: an application to a brazilian Pre-salt reservoir Marcos de Carvalho Machado Petrobras S. A.
9:15 AM 0:50:00 Keynotes: Digital transformation: challenges and opportunities in Petrobras Exploration and Reservoir Characterization Pedro Henrique Silvany Sales & Ricardo Fernandes Reservoir Geophysicist, & EXP Coordinator, Petrobras
10:05 AM 0:30:00 Break

Session 2: Application to Reservoir Characterization

10:35 AM 0:25:00 Seismic Facies Inversion with Deep Learning and Ensemble Smoother Rodrigo Exterkoetter LTrace/UFSC
11:00 AM 0:25:00 Deep learning applied towards rock core tomography scan stylization João Fernando Sampaio Ralha Petrec
11:25 AM 0:25:00 Stochastic Simulation of Geological Models from Seismic using GANs Erick Costa e Silva Talarico Petrobras
11:50 AM 0:25:00 3D Convolutional Neural Network for aspect ratio estimation from CT-images Rodrigo Exterkoetter LTrace Geophysical Solutions
12:15 PM 2:00:00 Lunch

Session 3: Applications to Seismic Interpretation

2:15 PM 0:25:00 Self Organizing Maps with Well Calibration Improves Stratigraphic Resolution in an Unconventional Reservoir: A Case Study of the Niobrara Formation, Onshore U.S. Carrie Laudon Geophysical Insights
2:40 PM 0:25:00 Improving UNet Fault Prediction using RGT panels Marcelo Oliviera da Silva Center for Petroleum Studies (CEPETRO), University of Campinas (Unicamp)
3:05 PM 0:25:00 Interactive Deep Learning Assisted Seismic Interpretation Technology Applied to Seismic Interpretation Offshore Santos Basin in Brazil Juliana Fernandes Araujo IesBrazil / UFF
3:30 PM 0:50:00 Keynote Speaker: OSDU and the impact on the Energy market Johan Krebbers GM Digital Emerging Technologies/VP IT Innovation, Shell
4:20 PM 0:30:00 Break

Session 4: Application to Seismic Processing

4:50 PM 0:25:00 Sequence-to-sequence learning of artificial bandwidth extension Mehdi Aharchaou ExxonMobil
5:15 PM 0:25:00 Noise Removal with Generative Adversarial Networks Jonlenes Silva de Castro Universidade Estadual de Campinas (UNICAMP)
5:40 PM 0:25:00 Low Bit Rate 2D Seismic Image Compression With Deep Autoencoders João Paulo Navarro UFJF
6:05 PM 0:25:00 Siamese Neural Networks for Edge-Aware Filtering Mehdi Aharchaou ExxonMobil

Day 2

Session 1: Application to Reservoir Characterization

Start Duration Presentation Title Speaker Affiliation
8:00 AM 0:25:00 Feature Extraction from Tomograms Reservoir Rock Samples for Image Retrieval, Using Deep Jonatas Einsiedler Petrobras
8:25 AM 0:25:00 Permeability Prediction Using Support Vector Machine in a Carbonate Reservoir of Campos Basin Mohammad Saad Allahham Petroleum Engineering and Exploration Laboratory, Darcy Ribeiro Northern Rio de Janeiro State University
8:50 AM 0:25:00 Petroleum Well Data Fitting Based on Deep Learning Techniques Leodécio Braz da Silva Segundo University of Campinas
9:15 AM 0:50:00 Keynote Speaker Maurice Nessim President/CEO, WesternGeco and President, Society of Exploration Geophysicists
10:05 AM 0:30:00 Break

Session 2: Applications to Seismic Interpretation

10:35 AM 0:25:00 Machine Learning Techniques to Estimate Reservoir Property Changes directly from a 4D Seismic Attribute and Geographic Coordinates Marcos Vinicius Mussel Cirne Instituto de Computação, Universidade Estadual de Campinas (UNICAMP)
11:00 AM 0:25:00 Unsupervised Machine Learning Models Applied to Seismic Interpretation: Aggregating Lithologic Properties through the Inclusion of Well Log Profiles Alimed Celecia Ramos Pontifiacal Catholic University of Rio de Janeiro; PUC-Rio
11:25 AM 0:25:00 Acoustic Wave Inversion using Radial Basis Neural Networks Miguel Paredes Quinones IBM
11:50 AM 0:25:00 Using Unsupervised Neural Classification to Understand Complex Reservoir Systems Deborah Sacrey Auburn Energy
12:15 PM 1:20:00 Lunch

Session 3: Application to Seismic processing

1:35 PM 0:25:00 Seismic Slopes Estimation using Residual Neural Networks Marcelo Silva HPG / CEPETRO
2:00 PM 0:25:00 Application of Principal Component Analysis and Deep Learning along Directional Image Gathers for High-Definition Classification of Subsurface Features Bruno de Ribet Emerson
2:25 PM 0:25:00 Synthetic Data Generation with Neural Noise Enhancement Marcelo Silva HPG / CEPETRO
2:50 PM 0:25:00 Machine Learning and tools to support a computational workflow for reverse time migration with quantified uncertainty Fernando Alves Rochinha Universidade Federal do Rio de Janeiro
3:15 PM 0:50:00 Keynote Speaker: How AI is Digitally Transforming the Workforce Mark Spieler NVIDIA
4:05 PM 0:30:00 Break

Session 4: Case Studies in Geoscience

4:35 PM 0:25:00 A benchmark case for machine learning with well log data curated from New Zealand's public reports Breno W.S.R de Carvalho IBM/UFF
5:00 PM 0:25:00 Machine Learning Techniques Applied to Carajás Magnetic and Gammaspectrometric Data for Targeting Telma Aisengart Independent Consultant
5:25 PM 0:25:00 Slowness prediction from well log data in Candeias Formation - Recôncavo Basin, Brazil Matheus Radamés Silva Barbosa UFBA - Instituto Nacional de Ciências e Tecnologia (INCT-GP)
5:50 PM 0:25:00 Workshop Wrap up and Discussion

The presentations are scheduled for Rio time zone (GMT-3)