DeepLight презентация

Июль 31, 2022

Содержание

2. https://www.reddirtreport.com A map of world oil reserves, 2013. World hydrocarbon resources
3. well correlation 3d model frame 3d physical model [Thenin, Larson, 2014] hydrodynamic simulation
4. Machine Learning in Oil & Gas Asset Optimization Reservoir Management Exploration Drilling, Well Logging Production Image
5. Laboratory on Machine Learning in Oil & Gas Industry Partnership Research and Innovation Projects: applied projects
6. Interpretation
7. Fault Detection 0 Image Recognition (CNN) Generative Models GAN, VAE, Bayesian
8. Fault detection in slices Fault surface construction Fault Detection 1 Image Recognition (CNN) Generative Models GAN,
9. Well logging www.saltworkconsultants.com sanuja.com infolupki.pgi.gov.pl
10. Petrophysical model Convert well logs to petrophysical models http://wallace-international.com/ Image Recognition (CNN) Generative Models GAN, VAE,
11. Lithological model [Hall, 2016] Convert well logs to rock types Image Recognition (CNN) Generative Models GAN,
12. Well Correlation Image Recognition (CNN) Generative Models GAN, VAE, Bayesian
13. Production
14. Production 0 Physical Systems (explicit PDE) Production Modelling using Proxy Models. History matching of physical Proxy
16. Examples. Monthly well production.
17. Production 1 Physical Systems (explicit PDE) Production Modelling using Proxy Models. History matching of physical Proxy
18. Production 2 Physical Systems (learn from simulator) Production Modelling using simulator. History matching of simulator models
19. Production 3 Physical Systems (explicit PDE) Optimization of production using simulator with uncertainty. Generative Models (GAN,
21. Скачать презентацию

Слайд 2

https://www.reddirtreport.com
A map of world oil reserves, 2013.
World hydrocarbon resources

Слайд 3

well correlation
3d model
frame
3d physical
model
[Thenin, Larson, 2014]
hydrodynamic
simulation

Слайд 4

Machine Learning in Oil & Gas
Asset Optimization
Reservoir Management
Exploration
Drilling, Well Logging
Production
Image

Recognition

Generative Models
GAN, VAE, Bayesian

Physical Systems

Reinforcement Learning

Maintenance

Interpretation

Field Development

Слайд 5

Laboratory on Machine Learning
in Oil & Gas Industry
Partnership
Research and Innovation

Projects:
applied projects
partnership with oil/gas companies
Student Training:
student thesis projects
publications
student professional activities

Слайд 6

Interpretation

Слайд 7

Fault Detection 0
Image Recognition
(CNN)
Generative Models
GAN, VAE, Bayesian

Слайд 8

Fault detection in slices
Fault surface construction
Fault Detection 1
Image Recognition
(CNN)
Generative Models
GAN,

VAE, Bayesian

Слайд 9

Well logging
www.saltworkconsultants.com
sanuja.com
infolupki.pgi.gov.pl

Слайд 10

Petrophysical model
Convert well logs to petrophysical models
http://wallace-international.com/
Image Recognition
(CNN)
Generative Models
GAN, VAE, Bayesian

Слайд 11

Lithological model
[Hall, 2016]
Convert well logs to rock types
Image Recognition
(CNN)
Generative Models
GAN, VAE,

Bayesian

http://www.reddoggeo.com/

Слайд 12

Well Correlation
Image Recognition
(CNN)
Generative Models
GAN, VAE, Bayesian

Слайд 13

Production

Слайд 14

Production 0
Physical Systems
(explicit PDE)
Production Modelling using Proxy Models. History matching of

physical Proxy Models to production data.

Слайд 15

Слайд 16

Examples. Monthly well production.

Слайд 17

Production 1
Physical Systems
(explicit PDE)
Production Modelling using Proxy Models. History matching of

physical Proxy Models to production data with uncertainty.

Generative Models
(GAN, VAE, Bayesian)

Слайд 18

Production 2
Physical Systems
(learn from simulator)
Production Modelling using simulator. History matching of

simulator models to production data with uncertainty.

Generative Models
(GAN, VAE, Bayesian)

Слайд 19

Production 3
Physical Systems
(explicit PDE)
Optimization of production using simulator with uncertainty.
Generative Models
(GAN,

VAE, Bayesian)

Reinforcement Learning