Cristin-resultat-ID: 1913855
Sist endret: 21. februar 2022, 15:51
NVI-rapporteringsår: 2021
Resultat
Vitenskapelig Kapittel/Artikkel/Konferanseartikkel
2021

Real-Time Ensemble-Based Well-Log Interpretation for Geosteering

Bidragsytere:
  • Nazanin Jahani
  • Joaquin Ambia
  • Kristian Fossum
  • Sergey Alyaev
  • Erich Christian Suter og
  • Carlos Torres-Verdin

Bok

SPWLA 62nd Annual Logging Symposium
ISBN:
  • 000-0-00-000000-0

Utgiver

Society of Petroleum Engineers
NVI-nivå 1

Om resultatet

Vitenskapelig Kapittel/Artikkel/Konferanseartikkel
Publiseringsår: 2021
ISBN:
  • 000-0-00-000000-0

Klassifisering

Fagfelt (NPI)

Fagfelt: Geovitenskap
- Fagområde: Realfag og teknologi

Beskrivelse Beskrivelse

Tittel

Real-Time Ensemble-Based Well-Log Interpretation for Geosteering

Sammendrag

The costs for drilling wells on the Norwegian continental shelf are high, and hydrocarbons are often located in complex reservoir formations. To effectively produce from such reservoirs and reduce costs, optimized well placement in real time (geosteering) is crucial. Geosteering is usually assisted by an updated formation evaluation obtained by well-log interpretation while drilling. A reliable, computationally efficient, and robust workflow that can interpret well logs and capture uncertainties in real time is necessary for this application. An iterative version of the ensemble-based method, namely the approximate Levenberg Marquardt form of the Ensemble Randomized Maximum Likelihood (LM-EnRML), which is a Monte Carlo implementation of Bayesian inference, is integrated into our formation evaluation workflow. We estimate model parameters and related uncertainties for joint resistivity and density in addition to boundary locations by reducing statistical misfit between the observed measurements (well logs) and the numerically simulated logs. The modeled uncertainty is reduced as more data becomes available. The results of analyzing several synthetic cases verified that the applied iterative ensemble-based method could estimate model parameters (e.g., resistivity and bulk density) by employing as few as 40 ensemble members and just a small number of iterations. The results from the iterative ensemble-based method are compared with the classical Monte Carlo method. By comparing the CPU time, we concluded the proposed method has at least 10 times higher computational efficiency and similar accuracy. In addition, the ensemble-based method can run in parallel. The reliability of well-log interpretation is normally sensitive to several parameters, such as formation layer thickness, model parameter’s contrast, and the distance between boundary and measuring tool. Our sensitivity analysis on synthetic examples indicates that uncertainty is higher in cases with thinner layers and higher property contrasts. Furthermore, as expected, the higher uncertainty of properties is preserved in layers further away from the tool. Thus, the iterative version of the ensemble-based method has been proven robust, reliable, and computationally efficient to estimate petrophysical formation properties and their uncertainties, indicating that it is suitable for geosteering.

Bidragsytere

Nazanin Jahani

  • Tilknyttet:
    Forfatter
    ved NORCE Energi og teknologi ved NORCE Norwegian Research Centre AS

Joaquin Ambia

  • Tilknyttet:
    Forfatter
    ved University of Texas at Austin

Kristian Fossum

  • Tilknyttet:
    Forfatter
    ved NORCE Energi og teknologi ved NORCE Norwegian Research Centre AS
Aktiv cristin-person

Sergey Alyaev

  • Tilknyttet:
    Forfatter
    ved NORCE Energi og teknologi ved NORCE Norwegian Research Centre AS

Erich Christian Suter

  • Tilknyttet:
    Forfatter
    ved NORCE Energi og teknologi ved NORCE Norwegian Research Centre AS
1 - 5 av 6 | Neste | Siste »

Resultatet er en del av Resultatet er en del av

SPWLA 62nd Annual Logging Symposium.

Perkins, Tegwyn. 2021, Society of Petroleum Engineers. Vitenskapelig antologi/Konferanseserie
1 - 1 av 1