A Decision Analytic Approach to Geosteering Operations

Kullawan, K.; Bratvold, Reidar Brumer; Bickel, J. Eric

doi:10.2118/167433-pa

Cited by 29 publications

(14 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Ref. [54] a decision-analytic framework to support highquality geosteering decisions is implemented based on Bayesian inference. In Refs.…”

Section: Inversion Of Ultra-deep Directional Resistivity Measurementsmentioning

confidence: 99%

Generalized Transitional Markov Chain Monte Carlo Sampling Technique for Bayesian Inversion

Han¹,

Catanach²,

Chen³

et al. 2021

Preprint

View full text Add to dashboard Cite

In the context of Bayesian inversion for scientific and engineering modeling, Markov chain Monte Carlo sampling strategies are the benchmark due to their flexibility and robustness in dealing with arbitrary posterior probability density functions (PDFs). However, these algorithms been shown to be inefficient when sampling from posterior distributions that are high-dimensional or exhibit multi-modality and/or strong parameter correlations. In such contexts, the sequential Monte Carlo technique of transitional Markov chain Monte Carlo (TMCMC) provides a more efficient alternative. Despite the recent applicability for Bayesian updating and model selection across a variety of disciplines, TMCMC may require a prohibitive number of tempering stages when the prior PDF is significantly different from the target posterior. Furthermore, the need to start with an initial set of samples from the prior distribution may present a challenge when dealing with implicit priors, e.g. based on feasible regions. Finally, TMCMC can not be used for inverse problems with improper prior PDFs that represent lack of prior knowledge on all or a subset of parameters. In this investigation, a generalization of TM-CMC that alleviates such challenges and limitations is proposed, resulting in a tempering sampling strategy of enhanced robustness and computational efficiency. Convergence analysis of the proposed sequential Monte Carlo algorithm is presented, proving that the distance between the intermediate distributions and the target posterior distribution monotonically decreases as the algorithm proceeds. The enhanced efficiency associated with the pro-

show abstract

“…In Ref. [54] a decision-analytic framework to support highquality geosteering decisions is implemented based on Bayesian inference. In Refs.…”

Section: Inversion Of Ultra-deep Directional Resistivity Measurementsmentioning

confidence: 99%

Generalized Transitional Markov Chain Monte Carlo Sampling Technique for Bayesian Inversion

Han¹,

Catanach²,

Chen³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…A decision analytic framework for geosteering is discussed in [23] [24] [25]. It aims to provide unbiased and consistent decision support under uncertainty by optimization with respect to multiple weighed geosteering and drilling objectives.…”

Section: A2 a Potential Future Geosteering Workflowmentioning

confidence: 99%

A Novel Method for Locally Updating an Earth Model While Geosteering

Suter¹,

Cayeux²,

Friis³

et al. 2017

IJG

View full text Add to dashboard Cite

Earth models are important tools for support of decision making processes for optimal exploitation of subsurface resources. For geosteering and other realtime processes where time is a major constraint, effective model management is decisive for optimal decision support. During drilling, subsurface information is received which should optimally be used to modify the 3D earth model. Today this model is typically not altered during the operation. We discuss the principles of a novel method that enables a populated earth model grid to be locally modified when the topology (connectivity) of the geological structure is locally altered. The method also allows local updates of the grid resolution. The modelled volume is split into closed regions by the structural model. Each region is individually discretized and obtains its own subgrid. Properties are stored in separate functions, e.g. for each layer, and transferred into each subgrid via a mapping. A local update of the geological structure implies that only subgrids in regions that are directly affected by the updated structure must be discarded and rebuilt, and the rest of the populated earth model grid is retained. Our focus is on decision support for optimal well placement while geosteering. The proposed method aims to manage multiple model realizations that are never fixed and always locally updated with the most recent measurements and interpretations in real-time, and where each realization is always kept at an optimal resolution.

show abstract

“…To reach high-quality geosteering decisions, it is convenient to use statistical (Bayesian) interpretation, which provides uncertainties (Kullawan et al, 2014). Bayesian workflows for updating geomodels while drilling provide promising results in synthetic testing (Chen et al, 2015;Luo et al, 2015;Alyaev et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

Modeling extra-deep electromagnetic logs using a deep neural network

et al. 2021

View full text Add to dashboard Cite

Modern geosteering is heavily dependent on real-time interpretation of deep electromagnetic (EM) measurements. We present a methodology to construct a deep neural network (DNN) model trained to reproduce a full set of extra-deep EM logs consisting of 22 measurements per logging position. The model is trained in a 1D layered environment consisting of up to seven layers with different resistivity values.A commercial simulator provided by a tool vendor is utilized to generate a training dataset.The dataset size is limited because the simulator provided by the vendor is optimized for sequential execution.Therefore,we design a training dataset that embracesthe geological rules and geosteering specifics supported by the forward model. We use this dataset to produce an EM simulator based on a DNN without access to the proprietary information about the EM tool configuration or the original simulator source code.Despite employing a relatively small training set size, the resulting DNN forward model is quite accurate for the considered examples: a multi-layer synthetic case and a section of a published historical operation from the Goliat Field.The observed average evaluation time of 0.15 milliseconds per logging position makes it also suitable for future use as part of evaluation-hungry statistical and/or Monte-Carlo inversion algorithms within geosteering workflows.

show abstract

A Decision Analytic Approach to Geosteering Operations

Cited by 29 publications

References 23 publications

Generalized Transitional Markov Chain Monte Carlo Sampling Technique for Bayesian Inversion

Generalized Transitional Markov Chain Monte Carlo Sampling Technique for Bayesian Inversion

A Novel Method for Locally Updating an Earth Model While Geosteering

Modeling extra-deep electromagnetic logs using a deep neural network

Contact Info

Product

Resources

About