H51C-1374
Improved Modeling of Naturally Fractured Reservoirs by Quantitatively Handling Flow Convergence into the Wellbore

Friday, 18 December 2015
Poster Hall (Moscone South)
Dustin Crandall, Matthew Stadelman, W Neal Sams and Grant S Bromhal, National Energy Technology Laboratory Morgantown, Morgantown, WV, United States
Abstract:
Complex fractured networks in the subsurface control the flow of fluids in many applications, and accurately modeling their interaction with wells is critical to understanding their behavior. For tight sand and shale formations, fluid flow is primarily restricted to fractures within each rock layer. NFFLOW was designed by the Department of Energy to model gas well production from naturally fractured reservoirs. NFFLOW is a discrete fracture simulator, with every fracture and rock matrix in the domain handled individually. One-dimensional models are used calculate the flow through connected fractures and flow from the surrounding rocks into fractures. Flow into wellbores are determined from the combined flux from connecting fractures and adjacent rock matrices. One-dimensional fluid flow equations are used because they are extremely fast to solve and represent a reasonable approximation of the physical behavior of fluids in most of the reservoir. However, near the wellbore those models become inaccurate due to gas flow convergence, which is a multidimensional situation. We present a method to correct the one-dimensional models, using data from two-dimensional fluid flow models, while maintaining the original simulator speed. By applying corrections from the two-dimensional model, the one-dimensional models can better account for gas flow convergence into the wellbore as well as the location of the wellbore within the rock strata. Corrections were successful in scaling the one-dimensional flow rates to match the two dimensional values over a wide range of parameters for both fracture flow and porous media flow into the wellbore. This is shown to increase the accuracy of history matching to production data for a wide range of wells, allowing for better modeling and prediction of future productivity. With an accurate history match established, NFFLOW can then be used to investigate issues such as the ability of the formation to sequester carbon dioxide or the effects of additional hydraulic fracturing on fluid transport.