자료유형  

 학위논문

Control Number  

0017161492

International Standard Book Number  

9798382230696

Dewey Decimal Classification Number  

550

Main Entry-Personal Name  

Josue Sa da Fonseca.

Publication, Distribution, etc. (Imprint  

[S.l.] : Stanford University., 2024

Publication, Distribution, etc. (Imprint  

Ann Arbor : ProQuest Dissertations & Theses, 2024

Physical Description  

295 p.

General Note  

Source: Dissertations Abstracts International, Volume: 85-11, Section: B.

General Note  

Advisor: Tapan Mukerji.

Dissertation Note  

Thesis (Ph.D.)--Stanford University, 2024.

Summary, Etc.  

요약Effective decision-making in exploring and developing subsurface resources hinges on accurately estimating geologically plausible and realistic properties. Nevertheless, conventional inverse modeling formulations for property estimation encounter computational challenges arising from high-dimensional problem settings and the substantial geological uncertainty inherent in subsurface estimation problems. Especially for pore pressure forecast, this challenge yields to disregarding the geological processes occurring in the subsurface in the predictions by standardizing the use of empirical models calibrated only on offset wells. This dissertation introduces and investigates methodologies rooted in the Bayesian Geophysical Basin Modeling (BGBM) workflow to assess pore pressure. The aim is to incorporate geological processes into estimations and utilize probabilistic models, facilitating a comprehensive evaluation of pore pressure predictions across exploration and exploitation stages in subsurface resource activities.BGBM methodology is an interdisciplinary workflow that incorporates data, geological expertise, and physical processes through Bayesian inference in sedimentary basin models. Its application culminates in subsurface models and properties that integrate the geo-history of a basin, rock physics relations, well log and drilling data, and seismic information. Monte Carlo basin modeling realizations are performed by sampling from prior probability distributions on facies parameters and basin boundary conditions. After data assimilation, the accepted set of posterior subsurface models yields uncertainty quantification of subsurface properties. This procedure is especially suitable for pore pressure prediction in a predrill stage. However, the high computational cost of seismic data assimilation decreases the practicality of the workflow. Therefore, we introduce and investigate seismic travel times criteria as computationally faster proxies for analyzing the seismic data likelihood when employing BGBM. A real field case from the Gulf of Mexico using a 2D section for pore pressure prediction considering different kinematics criteria is evaluated. We validate and compare the outcomes for predicted pore pressure with mud weight data from a blind well. We demonstrated that the proposed fast proxies make the BGBM methodology efficient and practical.Nevertheless, the computational time for running 3D Monte Carlo forward basin modeling simulations to quantify pore pressure uncertainty remains a bottleneck. In order to tackle the latter challenge, we propose integrating the directness of empirical relations with process-based simulations of Earth models for further 3D extrapolation. The quantified uncertainty of pore pressure estimation from the posterior 2D results obtained after the BGBM workflow application is encoded into standard empirical equations. Eaton's model was selected to evaluate this procedure. Then, the BGBM 2D section of the real case provides the empirical model calibration. Validation at the blind well in the 2D section was remarkably similar to the BGBM result. The calibrated empirical parameters were then carried out to be extrapolated in a 3D test area near the BGBM study. Mud weight measurements of a test well in the 3D area confirm promising pore pressure forecasts.Ideally, the BGBM workflow should be rerun after new data is observed. BGBM robustly incorporates geological processes in pore pressure prediction. However, this workflow requires reasonable computational resources to simulate Earth models conditioned with data in a Monte Carlo fashion. Thus, real-time revised pore pressure information of deep targets with new data of shallow layers acquired during drilling becomes unfeasible for BGBM. Therefore, we suggest combining probabilistic graphical models with BGBM posterior results to create computationally efficient pore pressure uncertainty workflows to rapidly update beliefs when data arrives while drilling. Bayesian Networks (BNs) are proposed to represent subsurface pore pressure per geological compartment (e.g., stratigraphic layers) connected to facies parameters of layers possibly observed at the vertical location of the well perforation. The BNs are trained with the dataset derived from the BGBM posterior results. Finally, computational evaluations analyzing the inferred results of discrete and continuous BNs are demonstrated and discussed, concluding that BNs offer a viable probabilistic model for rapid pore pressure updates during drilling, provided that the BN has been calibrated with the posterior outputs from BGBM results.

Subject Added Entry-Topical Term  

Geophysics.

Subject Added Entry-Topical Term  

Engineering.

Subject Added Entry-Topical Term  

Environmental science.

Index Term-Uncontrolled  

Bayesian Networks

Index Term-Uncontrolled  

Computational challenges

Index Term-Uncontrolled  

Seismic data likelihood

Index Term-Uncontrolled  

Empirical relations

Added Entry-Corporate Name  

Stanford University.

Host Item Entry  

Dissertations Abstracts International. 85-11B.

Electronic Location and Access  

로그인을 한후 보실 수 있는 자료입니다.

Control Number  

joongbu:655460