33rd Annual CREWES Sponsors Meeting

Online [Zoom], December 2-4, 2021

Note: All dates and times on this page are Mountain Standard Time (MST), which is UTC -7 hours

* Zoom invitations and passcodes for all meeting events will be sent to registrants on Monday Nov 29

Day 1 (December 2rd, 2021)

Session 1 (D. Lawton). Acquisition, field applications and DAS: 8:00 am
 Welcome and overview: 8:00 am
 Talk 1. CO2 monitoring at the CaMI.FRS - updates
 Talk 2. Fibre trace registration by cross-correlations - can we successfully predict helically wound fibre pitch angle from recorded data?
 Talk 3. An encoder-decoder CNN for DAS-to-geophone transformation
 Talk 4. Rock physics analysis of well-log data
 Talk 5. Source wavefield estimation in VSP full waveform inversion
 Talk 6. Pre-processing of the 2018 CaMI VSP and parameterization for DAS FWI
 Talk 7. The 2018 CREWES VSP survey: FWI of the accelereomter and straight DAS fiber datasets

Session 2 (D. Trad). Machine Learning and Data Science: 10:40 am
 Talk 8. Seismic inversion with gradient boosting
 Talk 9. GPU applications for modelling, migration, FWI, and machine learning
 Talk 10. Convolutional neural network-based reverse time migration with multiple energy
 Talk 11. Quantum machine learning
 Talk 12. Learning the elastic wave equation with Fourier Neural Operators
 Talk 13. Facies classification with parametric and nonparametric statistics

Wrap up: 12:40 pm

Day 2 (December 3th, 2021)

Session 3 (K. Innanen). Full Waveform Inversion and Amplitude Inversion: 8:00 am
 Talk 14. Implicit neural representation FWI
 Talk 15. Stepsize sharing in time-lapse full-waveform inversion
 Talk 16. Statistical-mechanical models as tools for seismic data analysis and inversion
 Talk 17. Local optimization approaches for simultaneous AVO inversion based on re-parameterized Zoeppritz equations
 Talk 18. Reducing the influence of remnant noises on FWI with misfit modification
 Talk 19. Amplitude-encoding FWI using different bases
 Talk 20. Inversion of azimuthal PP-wave amplitudes without NMO correction for fracture weaknesses

Session 4 (D. Trad). CO2, EOR and Geothermal Applications: 10:20 am
 Talk 21. Time-lapse analysis of CaMI.FRS CO2 VSP data
 Talk 22. Time-lapse EFWI of CO2 injection at CaMI.FRS using VSP: a feasibility study
 Talk 23. Passive source location by diffraction scanning
 Talk 24. Time-lapse rock physics CO2 monitoring with FWI
 Talk 25. Shiny web applications for optimization of DFIT event detection with unsupervised learning
 Talk 26. Geothermal: geoscientific and engineering challenges
 Talk 27. Full waveform source and medium determination from seismic-while-drilling data
 Talk 28. Getting up and running with the short course codes

Wrap up: 12:40 pm

Day 3 (December 4th, 2021)

Short Course: Creating Machine Learning Apps in R with Shiny: 9:00 am

Setup (please complete this before the course begins if you want to follow along on your own computer):

  • Download and install R, select the mirror location and OS version: R-Project
  • Download and install RStudio Desktop - free version - for your OS: RStudio
  • Download course material, unzip, and open all the coding files in RStudio so necessary libraries can be automatically installed: Course materials
  • Create an account at Nasdaq: For the WTI Oil Price Forecast app, a QuandL key is needed. For that, go to your "Account Settings" to access your QuandL key: Nasdaq

Agenda:

  • Introduction to R (30 minutes)
  • Introduction to HTML (15 minutes)
  • Introduction to CSS (15 minutes)
  • Shiny App for Facies Classification (60 minutes)
  • Break (10 or 15 min)
  • Flexdashboard App for Oil Price Forcasting (60 minutes)