Complicated geology and scarce data are the two major problems in simulation of the subject large carbonate reservoir. Sharp contrasts in permeabilities, ranging from tens to more than one million milli-darcy-feet, are observed in a few hundred meters radius of the reservoir. Field data and past experience in reservoir simulation indicate that the field performance is completely dominated by these super high permeability streaks. Recent studies have shown that these high permeability streaks are very localized. The occurrence of these streaks is random and not correlatable. Consequently, they may be only amenable to geostatistical techniques. The Indicator Conditioned Estimation (ICE) technique described in this paper is designed to honor the spatial structure inherent in the source data and to allow for the prediction of super permeability streaks in areas of no well data. The use of pulsed neutron capture logging and production logging has been very successful in monitoring the sweep of the subject reservoir which is under a peripheral waterflood. At the same time, the logging data has also provided a wealth of information for the model construction; however, only the data derived from dry wells has been used in the past. A new procedure is developed to use information from wet as well as dry wells to augment the database for the model construction. This procedure in conjunction with ICE technique has formed an innovative way to generate the initial permeability distributions for a simulation model. It is expected that this initial permeability distribution should serve as a better starting point for the subsequent history matching process.
A new methodology was developed to better characterize the lithology, lithofacies and permeability of a pilot area in the Arab-D reservoir of the Ghawar Field. The study area has one hundred and seven wells. All wells have a conventional suite of openhole logs. There are only nine cored wells with detailed petrographic descriptions of the Arab-D. Four of these cored wells only partially sample the Arab-D reservoir. The methodology targets wells lacking core data and where all subsurface interpretation has to be based on wireline log data. The estimation of lithology from well logs was done using explicit stochastic techniques to calculate a multi-mineral volumetric model. A methodology was developed to incorporate various open hole log suites. Rigorous cross validation assured results consistent with core data. Lithofacies were defined in the cored wells combining depositional textural, and porosity-permeability relationships. Permeability and lithofacies were then estimated from the open hole well logs using multivariate statistical techniques. The cored wells were used to create a multi-dimensional ‘Learn’ histogram/database with core permeability, lithofacies porosity, grain density, and well log derived textural parameters as variables. The validation process involved using the database of each cored well to estimate the permeability and lithofacies in every other cored well. The combined histogram/database from the cored wells was then used to estimate the permeability and lithofacies for the uncored wells. As a result, a more accurate quantification of dolomite is developed, major lithofacies units are identified, and permeabilities are estimated from open hole log data. The methodology developed in this study provides reservoir characterization control at uncored well locations. The accurate estimation of these parameters and their spatial distribution is essential for building a technically sound geological reservoir model for managing this Arab-D reservoir. Introduction One of the most difficult problem faced when building a geological reservoir model is not having enough wells with core data to estimate the distribution of the parameters controlling fluid flow within the reservoir (Figure 1). Detailed evaluation of the core data from the Arab-D reservoir in four wells in a pilot area of the Ghawar Field demonstrated that high values of core plug permeability were often associated with the better flow intervals as indicated by flow meter data. It was also established that the core plug permeability could be related to the lithology and lithofacies from the core descriptions (Figure 2). However, the data from these detailed core descriptions coupled with less detailed information from five other cored wells in the area were deemed insufficient to predict the distribution of permeability throughout the reservoir. Additional information is available from the open hole log suites of 107 wells in the study area. Estimation of the core plug equivalent permeability, lithology, and lithofacies at the uncored wells would provide the control to better distribute these parameters and model the fluid flow in the reservoir. Project Objective The objective of this project was to develop a methodology to estimate the lithology, lithofacies, and core plug permeability in the uncored wells for a pilot area in the Arab-D reservoir of the Ghawar Field (Figure 3). The accurate estimation of these parameters and their spatial distribution is essential for building a technically sound reservoir geological reservoir model for managing this Arab-D reservoir. P. 245^
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.