Benchmarking Recovery Factors for Carbonate Reservoirs: Key Challenges and Main Findings from Middle Eastern Fields

摘要

The objective of this paper is to discuss the process of reservoir/field data aggregation, verification and validation process, to extract some key determinant reservoir parameters to be used in recovery factor (RF) benchmarking against global analogs. The subsequent step is to perform the benchmark of the reservoir performance against fields with similar reservoir characteristics and field development plans (FDPs), using digital databases and analytic tools. The use of data mining and analytics is also attempted to estimate RFs for undeveloped reservoirs. RF improvement is the ultimate goal of any field development project being managed by National Oil Companies (NOC) or International Oil Companies (IOC). The extent to which the recovery of hydrocarbon resources is achieved, depends on our knowledge of the reservoir complexity (rock and fluid system), the field development strategy and the economic parameters. Each reservoir is unique in its characteristics and thus the need for the implementation of an appropriate field development scheme to maximize the recovery. One of the main objectives of this work was to identify a process that can assess the RF from a given reservoir while comparing it against a mature analogue across the globe, and then identify development opportunites that can further improve RF. To do this we need to 1)-classify ADNOC's reservoirs in a systematic and consistent manner, 2)-extract key determinant parameters that contribute to RF and 3)-compare them with performance of similar fields extracted from worldwide databases. Once this is done and, with the assistance of digital databases and analytical tools, the benchmark can be performed. This paper addresses all the challenges involved in the database preparation, data verification and validation process and presents a summary of the main findings in establishing the key determinant parameters used for the final reservoir classification (reservoir complexity index), which is used to select similar reservoirs for the recovery factor benchmark process. The paper exhibits also examples of RFs benchmarks for different carbonate reservoir complexities and architectures including but not limited to: 1)- reservoirs in transition zones, 2)-reservoirs affected by karstification and fracturation. 3)-reservoirs formed in complex clinoform (prograding) systems, 4)-reservoirs formed in aggradational layer-cake systems and 5)-reservoirs from sub-tidal sabkha environments with inter-bedded anhydritic layers. The novelty of this work is the judicious integration of all static and dynamic data, the proper parameterization of key determinant reservoir parameters, statistics extracted from 3D geostatistical models, the use of predictive analytic tools including self-clustering and classification and dimensionality reduction to estimate RFs for undeveloped reservoirs with similar reservoir complexity.