This paper presents an ensemble-based computer Assisted History Matching (AHM) of a real life carbonate oil field. The field-level reservoir pressures were matched with a fine-scale Dual-Porosity DualPermeability (DPDP) model spanning a long production history under primarily peripheral water injection pressure support. The well-level AHM workflow presented was validated with a DPDP high-resolution sector model of a fracture dominated carbonate reservoir. This sector model was ~17 million active grid cells with no application of simulation grid upscaling. The AHM workflow integrates probabilistic Bayesian inference using Ensemble Smoother with Multiple Data Assimilation (ES-MDA), which simultaneously assimilates the data and generates maximum a-posteriori updates of reservoir model parameters in a variance- minimizing update scheme. A detailed uncertainty matrix was built with ensemble of sensitivity scenarios, based on varying free water level, corresponding matrix porosity and the initial water saturation combined with geostatistical realizations of dynamic permeability derived from dynamic PLT logs and fracture characterization, where the varied parameters were the variogram attributes in terms of correlation length and geometric anisotropy. Five data assimilation iterations with ES-MDA method were required to achieve acceptable convergence and minimization of objective function, defined as a joint misfit of well-level static pressures and watercut for the key producing wells. Practical DPDP model simulation times were achieved through utilization of Massive Parallel Processing technology. This study presents the first ensemble-based approach to integrated reservoir modeling for a mature oil field with the objective to deliver geologically-constrained history matched models with better predictive value for production optimization and forecasting.

Advancements in numerical well testing packages in interpreting pressure transient behavior of complex well geometries and reservoir structures have lead to an improved understanding of the multi-scale heterogeneity encountered in dual-porosity dual-permeability (DPDP) reservoirs. This paper demonstrates the power of numerical well testing models in handling conceptual cases of increasing complexity in dual-porosity dual-permeability (DPDP) reservoirs where a high permeability matrix system interact with super-k intervals, fractures, and faults systems with different levels of complexity. Numerical well test models are built using data from multiple scales and sources (image logs, flowmeter responses (PLTs), petrophysical logs (FALs), and seismic attributes) to match pressure transient responses of wells completed in dual-porosity dual-permeability reservoirs. Six generalized conceptual cases are presented in this paper; a vertical power water injector that initiated induced fractures due to injection above fracture pressure, a vertical well near an area of intersecting faults, a 40-degree deviated well intersecting diffuse fracture network, a deviated well near a conductive fracture corridor, a horizontal well intersecting a finite conductivity fracture, and a horizontal well intersecting an infinite conductivity fracture. An integrated approach was used to match the pressure responses in all cases. Experience shows that the most representative well test solution comes from a thorough integration of well-test data with all available static and dynamic data (e.g. image logs, flowmeter responses (PLTs), geophysical, and petro-physical data). In addition, inclusion of pressure transient responses of offset wells in the area understudy as part of the analysis is crucial in choosing the most representative well test model. In summary, this paper provides guidance and best practice to reservoir engineers in building numerical well test models to analyze fracture and matrix responses in dual-porosity dual-permeability (DPDP) reservoirs. Conceptual cases emphasize integration of multiple sources of static and dynamic data to model different well responses.

This paper demonstrates the effect of pore systems and mineralogy on imbibition capillary pressure (Pci) of carbonate rocks. A systematic workflow is developed and followed to ensure the data quality of Pci, minimize uncertainty in deriving the Pci from centrifuge tests, and analyze the data together with pore-size distribution from mercury injection capillary pressure (MICP) and mineralogy from Quantitative Evaluation of Minerals by Scanning Electron Microscopy (QEMSCAN). The workflow starts with assessing the centrifuge production data for gravity-capillary equilibrium at each speed. Then the quality-checked data is used to produce six different Pci curves using the analytical and numerical models. The analytical and numerical solutions assess the variability in solutions for various rock types, and ultimately, lead to the selection of the most-representative Pci curve. Finally, the representative Pci curves of varying rock types are analyzed together with the MICP and QEMSCAN data to examine the change in Pci curves as a result of changes in the number and character of pore systems, dominant pore throat radii, and mineralogy. Findings from this study present insights into the impact of mineralogy and pore systems on the behavior of the Pci curves. From the mineralogy perspective, the presence of dolomite, microporous calcite, or rutile and anatase (TiO2) within the rock composition has a strong influence on the Pci behavior of carbonate rock. The data reveals that the contrast between the micropore and macropore systems of bi-modal carbonates has the strongest influence on Pci. We find that Pci can be clustered based on mineral content for bi-modal carbonate rocks and the degree of communication between micropore and macropore systems. The novel approach presented in this study links the MICP and QEMSCAN data to the imbibition process making the way toward a better dynamic rock typing.

Summary In this paper, we present an experimental study that explores the potential links between the imbibition capillary pressure Pci and the pore systems and/or mineralogy for carbonate reservoirs undergoing waterflood. A systematic workflow has been formulated to ensure the data quality of Pci, minimize uncertainty in deriving Pci from centrifuge tests, and analyze the data considering the pore-size distribution from mercury injection capillary pressure (MICP) and mineralogy from Quantitative Evaluation of Minerals by Scanning Electron Microscopy (QEMSCAN). The workflow starts with assessing the centrifuge production data for gravity-capillary equilibrium at each speed. Then, the quality-checked data are used to generate six different Pci curves using analytical and numerical models. The resulting curves provide a measure of the variability in solutions for various rock types and assist in the selection of the most-representativePci curve. Finally, the representative Pci curves of all rock samples are analyzed together with the MICP and QEMSCAN data to examine the change in Pci curves as a result of changes in the number and character of rock types, pore systems, dominant pore-throat radii, and mineralogy. Findings from this study shed light on the impact of mineralogy and pore systems on Pci. From the mineralogy perspective, the presence of dolomite, microporous calcite, or rutile and anatase (TiO2) within the rock composition is found to affect the Pci of the carbonate samples used in this study. The rock samples with these minerals should be separated from other bimodal samples before attempting to obtain a correlation between Pci and pore systems. The data analysis further reveals that some bimodal samples of medium permeability yield a better waterflood imbibition efficiency than those of the high-permeability samples. This observation is attributed to a better communication between the micropore and macropore systems, and a closer proximity of the peak radii of the micro- and macropore systems of the medium-permeabilitysamples.

Advancements in pore-scale and core-scale studies have provided an improved understanding of the micro- and macro-porosity nature of carbonate rocks and how the two systems interact. The interaction of the two systems in the presence of a third (fracture) and fourth component (vugs) has not been fully investigated in the industry. This paper demonstrates applicability and some limitations of permeability conditioning practices in dual-porosity dual-permeability (DPDP) systems. In addition, this work demonstrates how the permeability conditioning process can be used as a tool for dynamic classification and calibration of extreme permeability (super-k) intervals in dual-permeability systems. A highly scalable parallel DPDP finite difference simulator is used to: Firstly, demonstrate the permeability conditioning process and how it impacts reservoir dynamics. Secondly, present cases where flowmeter (PLT) responses show a limitation in characterizing super-k intervals and its impact. Thirdly, demonstrate the role of enhancement factor in representing flood front movement for multiple super-k dominated reservoir realizations constrained by flowmeter and pressure transient permeability-thickness controls. The results of this work expands on the representation of super-k intervals in dual-permeability systems in three main areas. Firstly, the decision to explicitly model super-k intervals as a fractured media or to implicitly model these features as a matrix permeability enhancement should be evaluated with use of enhancement factor combined with water breakthrough trends observed in the field. Secondly, the use of PLT responses to characterize super-k intervals should be made after careful evaluation of their responses before and after any well intervention. This step is crucial for proper permeability conditioning and in capturing reservoir dynamics of masked high flow intervals, i.e., new flow dominating features that appear only after the original super-k intervals have been plugged. Thirdly, as part of the integration of pressure transient results into a DPDP finite difference model, special consideration is needed for wells with a non-intersecting conductive fracture signature due to a limitation in the Peaceman formulation for DPDP reservoirs, which only considers cells intersecting the well for productivity index and PLT response calculations. In summary, this paper provides guidance for geologists and reservoir engineers, through use of a permeability conditioning process, to dynamically classify and calibrate fractured/super-k intervals during the process of constructing full-field dual-porosity dual-permeability reservoir simulation models.