Identification of Fractures and Preferential Flow Paths Using Streamlines and Dynamic Data in Dual Porosity Dual Permeability Reservoir Models
Conference Paper
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
Identification of the location and orientation of fractures and preferential flow paths using the dynamic reservoir response can provide insightful guidance for naturally fractured reservoir development and management. However, data assimilation for high-resolution fractured reservoir models using well responses is time consuming. The streamline-based technology has proven to be effective and efficient for subsurface flow modeling and inverse problems. We propose a novel and robust streamline tracing method in dual porosity dual permeability (DPDK) systems and an efficient workflow to identify fracture location in fractured reservoir models using dynamic data.In DPDK models, the matrix-fracture interactions are typically described by non-neighbor connections (NNCs) which make streamline tracing non-trivial. We use a boundary layer method to reconstruct the fluxes and avoid flux discontinuities at NNCs. The flux reconstruction allows us to trace streamlines and compute the time of flight (TOF) in DPDK models. The fracture identification workflow relies on the streamline-based analytic sensitivity computation and rapid model calibration using dynamic data. Multiple realizations of fracture distribution are initialized and calibrated using observed well responses, for example water cut and BHP. The ensemble of calibrated fracture models is then used for uncertainty analysis, and a probability map of the spatial distribution of fractures is finally generated.The power and utility of our approach are demonstrated using several applications. We first validate the streamline tracing algorithm for DPDK model by comparing its TOF with the result generated by the finite volume formulation of TOF generalized for compressible flow. Next, we validate the fracture identification workflow using a synthetic case by comparing the fracture probability map with the "true" fracture distribution. Finally, we apply our method to a DPDK model of a fractured reservoir and generate fracture probability map by calibrating multiple model realizations to well production and pressure data. The fracture probability maps serve as valuable tools to guide reservoir management and field development strategy.The novelty of this work is the newly developed streamline tracing algorithm for DPDK models and the efficient workflow to generate fracture probability maps based on dynamic data. The proposed approach is easy to implement and can be coupled with commercial simulators for field scale applications.
