DeepFlow: History Matching in the Space of Deep Generative Models
The calibration of a reservoir model with observed transient data of fluid pressures and rates is a key task in obtaining a predictive model of the flow and transport behaviour of the earth's subsurface. The model calibration task, commonly referred to as "history matching", can be formalised as an ill-posed inverse problem where we aim to find the underlying spatial distribution of petrophysical properties that explain the observed dynamic data. We use a generative adversarial network pretrained on geostatistical object-based models to represent the distribution of rock properties for a synthetic model of a hydrocarbon reservoir. The dynamic behaviour of the reservoir fluids is modelled using a transient two-phase incompressible Darcy formulation. We invert for the underlying reservoir properties by first modeling property distributions using the pre-trained generative model then using the adjoint equations of the forward problem to perform gradient descent on the latent variables that control the output of the generative model. In addition to the dynamic observation data, we include well rock-type constraints by introducing an additional objective function. Our contribution shows that for a synthetic test case, we are able to obtain solutions to the inverse problem by optimising in the latent variable space of a deep generative model, given a set of transient observations of a non-linear forward problem.
NurtureToken New!

Token crowdsale for this paper ends in

Buy Nurture Tokens

Authors

Are you an author of this paper? Check the Twitter handle we have for you is correct.

Lukas Mosser (edit)
Olivier Dubrule (add twitter)
Martin J. Blunt (add twitter)
Ask The Authors

Ask the authors of this paper a question or leave a comment.

Read it. Rate it.
#1. Which part of the paper did you read?

#2. The paper contains new data or analyses that is openly accessible?
#3. The conclusion is supported by the data and analyses?
#4. The conclusion is of scientific interest?
#5. The result is likely to lead to future research?

Github
Stargazers:
6
Forks:
2
Open Issues:
0
Network:
2
Subscribers:
1
Language:
MATLAB
Pytorch implementation of "DeepFlow: History Matching in the Space of Deep Generative Models"
Youtube
Link:
None (add)
Views:
0
Likes:
0
Dislikes:
0
Favorites:
0
Comments:
0
Other
Sample Sizes (N=):
Inserted:
Words Total:
Words Unique:
Source:
Abstract:
None
05/14/19 06:01PM
9,845
2,753
Tweets
arxiv_cscv: DeepFlow: History Matching in the Space of Deep Generative Models https://t.co/BMZsLfK2ah
arxiv_cscv: DeepFlow: History Matching in the Space of Deep Generative Models https://t.co/BMZsLfK2ah
jesserobertson: Really interesting article using observations of time dependent flow to invert for subsurface properties (e.g. for petroleum or water): https://t.co/wvBAL94179 https://t.co/eXeGHIPrAg
maskot1977: DeepFlow: History Matching in the Space of Deep Generative Models https://t.co/MnIY6JsvkZ
arxiv_cscv: DeepFlow: History Matching in the Space of Deep Generative Models https://t.co/BMZsLfK2ah
udmrzn: RT @StatsPapers: DeepFlow: History Matching in the Space of Deep Generative Models. https://t.co/HEOyprqXhS
udmrzn: RT @arxiv_cscv: DeepFlow: History Matching in the Space of Deep Generative Models https://t.co/BMZsLfK2ah
porestar: Very excited to share our most recent work "DeepFlow: History Matching in the Space of Generative Networks" We couple a GAN with a differentiable simulator to perform history matching in latent space. 🎇 1/n ➡️ Paper: https://t.co/xXCg0UR743 ➡️ Code: https://t.co/oQe93d40uK https://t.co/vViZOMPIhW
StatsPapers: DeepFlow: History Matching in the Space of Deep Generative Models. https://t.co/HEOyprqXhS
arxiv_cscv: DeepFlow: History Matching in the Space of Deep Generative Models https://t.co/BMZsLfK2ah
arxiv_cs_LG: DeepFlow: History Matching in the Space of Deep Generative Models. Lukas Mosser, Olivier Dubrule, and Martin J. Blunt https://t.co/tkzigNMVq3
Images
Related