Usage
  • 194 views
  • 306 downloads

Data-Driven Approaches to Estimate the Impact and Presence of Shale Barriers in SAGD Reservoirs

  • Author / Creator
    Zheng, Jingwen
  • Steam-assisted gravity drainage (SAGD) is one of the most successful in-situ techniques and has been widely adopted for heavy oil and bitumen recovery. Steam chamber development and SAGD production performance are highly sensitive to the reservoir heterogeneity, thus characterizing shale heterogeneities and quantifying their influence is essential for analysis and optimization of SAGD operations. This thesis presents a novel workflow with data-driven techniques to quantify the uncertain influences of heterogeneous shale barrier configurations and infer potential shale barrier configurations from the SAGD production histories. The workflow is trained and tested on a set of 2D and 3D synthetic models, and it is subsequently applied to a field case study. The data employed for establishing the workflow is derived from a set of synthetic cases based on petrophysical properties and operational constraints representative of the Athabasca oil sands reservoirs. Reservoir heterogeneities are simulated by superimposing sets of idealized shale barrier configurations on the homogeneous model; each shale configuration is parameterized by a unique set of indices that represent the location and geometry of shale barriers. The workflow consists of three major steps: (1) multidimensional scaling and cluster analysis are performed to classify the cases in the training dataset into multiple groups sharing similar production characteristics; (2) for each cluster, an AI-based forward model is constructed to correlate the reservoir simulation predictions (outputs) with the associated shale barrier configurations (inputs); (3) to infer the unknown shale barrier configuration corresponding to a new SAGD production history, its production profile is first analyzed and assigned to one of the clusters identified in step 2, and a hybrid inverse modeling scheme, which integrates the genetic algorithm and previously-trained forward model, is adopted.The workflow is initially tested using several cases with arbitrary shale barrier configurations. Good agreement between the predicted and target shale barrier configurations is observed. The workflow is later applied to examine the actual field production histories extracted from Firebag. The inferred shale barrier features are consistent with those interpreted from the petrophysical log. This work introduces a novel systematic workflow for applying machine learning techniques to analyze the impact and distribution of heterogeneous shale barriers from SAGD field data. It presents an innovative 3D shale barrier parameterization scheme and a cluster-based approach for visualizing and inferring internal structures among many realizations of shale barrier configurations. It offers a potentially efficient way for identifying a reasonable ensemble of initial configurations that can be subjected to further (more detailed) history matching and facilitate the optimization of operational strategies for mitigating the impact of shale barriers.

  • Subjects / Keywords
  • Graduation date
    Spring 2019
  • Type of Item
    Thesis
  • Degree
    Doctor of Philosophy
  • DOI
    https://doi.org/10.7939/r3-k2n7-3470
  • License
    Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.