Developing a High-Efficiency Method for Field-Scale Simulation of a Tight and Naturally Fractured Reservoir in the Williston Basin

Natural fractures are important for oil and gas production in tight reservoirs that are usually characterized by ultralow matrix porosity and permeability. Natural fractures can be characterized by core sample analysis, image logs, production data, and other data sets; however, some of these data may not always be available. Addressing the uncertainty of natural fractures in the modeling and simulation processes is challenging, especially for a field-scale simulation that requires intense computational efforts. The objective of this study was to develop a high-efficiency simulation workflow that quantifies drained reservoir volumes (DRVs) and reproduces the production and water injection history for a tight and naturally fractured reservoir in the Williston Basin. The production history of all the wells and logging data were collected to characterize reservoir heterogeneity. Core samples from the target reservoir were tested to estimate petrophysical properties and identify natural fractures. The reservoir and fracture information was used to build a field-scale reservoir model. History matching was performed using a compositional reservoir simulator with the embedded discrete fracture modeling (EDFM) method. A fracture and DRV optimization (FDO) workflow was developed to improve history-matching results. The workflow enables dynamic updating of fracture and DRV parameters during history matching to reflect the change of key fracture properties in the production process. Core sample analysis showed that the permeability of most samples ranged from 0.001 to 0.1 mD, with a few samples at a permeability higher than 1 mD. Cumulative oil production per well ranged from 17 to 200 thousand barrels (Mbbl). The results of permeability and cumulative oil production distribution indicated strong heterogeneity within the reservoir. Open fractures were detected from the tested core samples. Experimental and production data were integrated within the field-scale reservoir model. Assisted by the FDO workflow, the production and water injection histories for all 17 wells were successfully matched. The simulation results showed that water injection led to increasing water cut in two offset wells and restoration of reservoir pressure. A novel FDO workflow was proposed in this study to better reproduce the production and water injection history of a tight reservoir with natural fractures in the Williston Basin. Compared to most of the published work in this area, the FDO workflow can significantly improve the simulation efficiency for field-scale simulation models considering the presence of a complex natural fracture network. The workflow can be further applied to reservoir fracture network characterization and history matching of other unconventional reservoirs with complex natural fracture systems.