IP Library › Granted Patent US 12,664,336
Granted Patent B2
US 12,664,336 · App. 17/693,261 · Granted Jun 23, 2026

Machine learning inversion using bayesian inference and sampling

Inventors: Daniele Colombo (Dhahran, SA); Anton Egorov (Moscow, RU); Ersan Turkoglu (Dhahran, SA)
Assignee: SAUDI ARABIAN OIL COMPANY
G06F30/27G01V1/282
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,664,336
App. No.
17/693,261
Filed
Mar 11, 2022
Granted
Jun 23, 2026
Kind
B2
Art Unit
2188
USPC
703/10
Abstract

Methods may include obtaining a preprocessed observed geophysical dataset based on an observed geophysical dataset of the subterranean region of interest and forming a training dataset composed of a geophysical training models and corresponding simulated geophysical training datasets. The methods may further include iteratively determining a simulated geophysical dataset from a current geophysical model, determining a data loss function between the preprocessed observed geophysical dataset and the simulated geophysical dataset, and training a machine learning (ML) network, using the training dataset, to predict a predicted geophysical model. The methods may still further include determining a model loss function between the current and predicted geophysical models and updating the current geophysical model based on an inversion using the data loss and model loss functions.

Claims (60)

1 . A method comprising:

iteratively:

obtaining an observed geophysical dataset from a subterranean region of interest;

determining, using a computer processor, a preprocessed observed geophysical dataset based on the observed geophysical dataset;

obtaining a current geophysical model of the subterranean region of interest;

forming a training dataset comprising a plurality of geophysical training models and a corresponding simulated geophysical training dataset for each of the plurality of geophysical training models;

performing inversion comprising iteratively or recursively, using the computer processor, until a convergence criterion is satisfied:

determining a simulated geophysical dataset by applying forward modeling to the current geophysical model,

determining a data loss function between the preprocessed observed geophysical dataset and the simulated geophysical dataset,

training a machine learning (ML) network, using the training dataset, to predict a predicted geophysical model from the preprocessed observed geophysical dataset,

inputting the preprocessed observed geophysical dataset into a trained machine learning network,

wherein the trained machine learning network comprises Bayesian inference;

producing a predicted geophysical model from the trained machine learning network using the preprocessed observed geophysical dataset,

wherein the predicted geophysical model comprises a plurality of model parameters, and

wherein each model parameter among the plurality of model parameters comprises a probability distribution of each model parameter,

determining a model loss function between the current geophysical model and the predicted geophysical model,

forming an objective function based on the data loss function and the model loss function,

determining a value of the objective function, and

updating the current geophysical model based on the value; and

locating a hydrocarbon reservoir based on the updated current geophysical model;

planning a wellbore path to the hydrocarbon reservoir;

drilling, using a geosteering system, a wellbore guided by the wellbore path comprising updating a position of a drill bit of the geosteering system based on the updated current geophysical model;

determining, using the geosteering system and during drilling, acoustic measurements; and

updating the observed geophysical dataset using the acoustic measurements.

2 . The method of claim 1 , wherein determining the preprocessed observed geophysical dataset comprises sampling the observed geophysical dataset.

3 . The method of claim 1 , wherein training the machine learning network comprises sampling a training dataset.

4 . The method of claim 3 , wherein sampling comprises active learning.

5 . The method of claim 1 , wherein the Bayesian inference comprises a Markov chain Monte Carlo method.

6 . The method of claim 1 , wherein the machine learning network comprises Gaussian process regression.

7 . The method of claim 1 , wherein the observed geophysical dataset comprises a seismic dataset.

8 . The method of claim 1 , wherein the trained machine learning network comprises a supervised machine learning network.

9 . The method of claim 1 , wherein training the trained machine learning network comprises:

augmenting a training dataset with the updated current geophysical model and the simulated geophysical dataset; and

retraining the trained machine learning network based on the augmented training dataset.

10 . A system comprising:

a seismic acquisition system configured to obtain an observed geophysical dataset from a subterranean region of interest;

a computer system configured to, iteratively:

receive the observed geophysical dataset,

determine a preprocessed observed geophysical dataset based on the observed geophysical dataset,

receive a current geophysical model of the subterranean region of interest,

form a training dataset comprising a plurality of geophysical training models and a corresponding simulated geophysical training dataset for each of the plurality of geophysical training models;

perform inversion comprising iteratively or recursively, until a convergence criterion is satisfied:

determine a simulated geophysical dataset by applying forward modeling to the current geophysical model;

determine a data loss function between the preprocessed observed geophysical dataset and the simulated geophysical dataset;

train a machine learning (ML) network, using the training dataset, to predict a predicted geophysical model from the preprocessed observed geophysical dataset,

input the preprocessed observed geophysical dataset into a trained machine learning network,

wherein the trained machine learning network comprises Bayesian inference;

produce a predicted geophysical model from the trained machine learning network using the preprocessed observed geophysical dataset,

wherein the predicted geophysical model comprises a plurality of model parameters, and

wherein each model parameter among the plurality of model parameters comprises a probability distribution of each model parameter;

determine a model loss function between the current geophysical model and the predicted geophysical model;

form an objective function based on the data loss function and the model loss function;

determine a value of the objective function;

update the current geophysical model based on the value, and

locate a hydrocarbon reservoir based on the updated current geophysical model;

a wellbore planning system configured to plan a wellbore path to the hydrocarbon reservoir; and

a geosteering system configured to:

drill a wellbore guided by the wellbore path comprising update a position of a drill bit of the geosteering system based on the updated current geophysical model, and

determine, during drilling, acoustic measurements;

wherein the computer system is further configured to update the observed geophysical dataset using the acoustic measurements.

Assignments (3)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 16, 2023
From: ARAMCO INNOVATIONS LLC
To: SAUDI ARABIAN OIL COMPANY
Reel/Frame 065240/0208 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 25, 2023
From: EGOROV, ANTON
To: ARAMCO INNOVATIONS LLC
Reel/Frame 063425/0766 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 14, 2022
From: COLOMBO, DANIELE; TURKOGLU, ERSAN; EGOROV, ANTON
To: SAUDI ARABIAN OIL COMPANY
Reel/Frame 059255/0588 →
Continuity (1)
Related Publication 20230289499A1 · Sep 14, 2023
References Cited (49)
US 8245795B2 · Dennis · 2012 [cited by examiner]
US 8363509B2 · Colombo et al. · 2013 [cited by applicant]
US 8861308B2 · Virgilio et al. · 2014 [cited by applicant]
US 8923094B2 · Jing et al. · 2014 [cited by applicant]
US 10261215B2 · Miotti et al. · 2019 [cited by applicant]
US 10920585B2 · Colombo et al. · 2021 [cited by applicant]
US 10996372B2 · Denli · 2021 [cited by examiner]
US 20160086079A1 · De Stefano · 2016 [cited by applicant]
US 20190195067A1 · Colombo et al. · 2019 [cited by applicant]
US 20190383965A1 · Salman et al. · 2019 [cited by applicant]
US 20200088896A1 · Schmedes et al. · 2020 [cited by applicant]
US 20200183031A1 · Denli et al. · 2020 [cited by applicant]
US 20200183035A1 · Liu · 2020 [cited by examiner]
US 20200183041A1 · Denli et al. · 2020 [cited by applicant]
US 20200184374A1 · Liu et al. · 2020 [cited by applicant]
US 20200257016A1 · Abdallah et al. · 2020 [cited by applicant]
US 20210041596A1 · Kushwaha et al. · 2021 [cited by applicant]
US 20210239872A1 · Wheelock et al. · 2021 [cited by applicant]
US 20210264262A1 · Colombo et al. · 2021 [cited by applicant]
CN 109313724B · 2021 [cited by applicant]
EP 2020609A1 · 2009 [cited by applicant]
WO 2020257263A1 · 2020 [cited by applicant]
WO 2021130512A1 · 2021 [cited by applicant]
WO 2021168230A1 · 2021 [cited by applicant]
Of Figueiredo et a Gaussian Mixture McMC method for linear seismic inversion, Geophysics, 2019 (Year: 2019). [cited by examiner]
Figueiredo et al Gaussian Mixture McMC method for linear seismic inversion, Geophysics, 2019 (Year: 2019). [cited by examiner]
Gaussian Process Regression within an active learning scheme, IGARSS 2011 (Year: 2011). [cited by examiner]
CN 113553773, Wu et al, Original and translation (Year: 2021). [cited by examiner]
Of Pasolli, et al Gaussian Process Regression within an active learning scheme, IGARSS 2011 (Year: 2011). [cited by examiner]
Colombo D. et al., “High-resolution velocity modeling by seismic-airborne TEM joint inversion: a new perspective for hear surface characterization”, The Leading Edge, vol. 35, Nov. 2016, pp. 977-985 (9 pages). [cited by applicant]
Colombo, D. and M. De Stefano, “Geophysical modeling via simultaneous joint inversion of seismic, gravity, and electromagnetic data: Application to prestack depth imaging”, The Leading Edge, vol. 26, Mar. 2007, pp. 326-… [cited by applicant]
Ronneberger, O. et al., “U-net: Convolutional Networks for Biomedical Image Segmentation”, May 2015, pp. 1-8, URL: <https://arxiv.org/abs/1505.04597v1> (8 pages). [cited by applicant]
Rovetta, Diego and Daniele Colombo, “Analysis of inter-domain coupling constraints for multi-physics joint inversion”, Inverse Problems, IOP Publishing Ltd., vol. 34, Nov. 2018, pp. 1-31 (31 pages). [cited by applicant]
Naeini, Ehsan Zabihi, “A machine learning approach to quantitative interpretation”, SEG International Exposition and 89th Annual Meeting, SEG, 2019, pp. 3176-3180 (5 pages). [cited by applicant]
Biswas, Reetam et al., “Prestack and poststack inversion using a physics-guided convolutional neural network”, Interpretation, Society of Exploration Geophysicists and American Association of Petroleum Geologists, vol. … [cited by applicant]
Colombo, Daniele et al., “Deep-learning electromagnetic monitoring coupled to fluid flow simulators”, Geophysics, Society of Exploration Geophysicists, vol. 85, No. 4, Jul.-Aug. 2020, pp. WA1-WA12 (12 pages). [cited by applicant]
Dogru, Ali H. et al., “A Next-Generation Parallel Reservoir Simulator for Giant Reservoirs”, SPE 119272, Society of Petroleum Engineers, Feb. 2009, pp. 1-29 (29 pages). [cited by applicant]
Archie, G.E., “The Electrical Resistivity Log as an Aid in Determining Some Reservoir Characteristics”, T.P. 1422, Petroleum Technology, Jan. 1942, pp. 54-62 (9 pages). [cited by applicant]
International Search Report and Written Opinion of the International Searching Authority issued in corresponding International Application No. PCT/US2021/018753, mailed on Aug. 11, 2021 (20 pages). [cited by applicant]
Colombo, Daniele et al., “Physics-driven deep learning joint inversion”, SEG Technical Program Expanded Abstracts 2020, pp. 1775-1779, Sep. 30, 2020 (8 pages). [cited by applicant]
Moorkamp, Max et al., “Joint Inversion in Hydrocarbon Exploration: Theory and Applications”, In: “Integrated Imaging of the Earth: Theory and Applications”, John Wiley & Sons, Inc, vol. 218, pp. 167-189, Apr. 22, 2016 (… [cited by applicant]
Colombo et al.; “Coupled physics-deep learning inversion”, Computers & Geosciences; vol. 157; Aug. 25, 2021; pp. 1-13 (13 pages). [cited by applicant]
He et al.; “Physics-informed neural networks for multiphysics data assimilation with application to subsurface transport”, Advances in Water Resources; vol. 141; May 6, 2020; pp. 1-15 (15 pages). [cited by applicant]
Sun et al.; “Physics-guided deep learning for seismic inversion with hybrid training and uncertainty analysis”, Geophysics; vol. 86; Issue 3; May 2021; pp. R303-R317 (15 pages). [cited by applicant]
Ray, A. and Myer, D.; “Bayesian geophysical inversion with trans-dimensional Gaussian process machine learning”, Geophysical Journal International; vol. 217; Feb. 28, 2019; pp. 1706-1726 (21 pages). [cited by applicant]
Berger et al.; “A Survey of Active Learning for Quantifying Vegetation Traits from Terrestial Earth Obersvation Data”, Remote Sensing; vol. 13; No. 2; 287; pp. 1-23 (23 pages). [cited by applicant]
Moss, Adam; “Accelerated Bayesian inference using deep learning”, Monthly Notices of the Royal Astronomical Society; vol. 496; May 2020; pp. 1-12 (12 pages). [cited by applicant]
Jannouch, E. H. and Holmstedt, O.; “Deep-learning-accelerated Bayesian inference for state-space models”, Master's thesis in Mathematics; Chalmers University of Technology; 2020 (91 pages). [cited by applicant]
Hooten et al.; “Making Recursive Bayesian Inference Accessible”, The American Statistician; vol. 75; No. 2; Sep. 2019 (10 pages). [cited by applicant]