Using Streamline-Derived Injection Efficiencies for Improved Waterflood Management

Abstract

Summary This paper describes a novel approach to predict injection- and production-well rate targets for improved management of waterfloods. The methodology centers on the unique ability of streamlines to define dynamic well allocation factors (WAFs) between injection and production wells. Streamlines allow well allocation factors to be broken down additionally into phase rates at either end of each injector/producer pair. Armed with these unique data, it is possible to define the injection efficiency (IE) for each injector and for injector/producer pairs in a simulation model. The IE quantifies how much oil can be recovered at a producing well for every unit of water injected by an offset injector connected to it. Because WAFs are derived directly from streamlines, the data reflect all the complexities impacting the dynamic behavior of the reservoir model, including the spatial permeability and porosity distributions, fault locations, the underlying computational grid, relative permeability data, pressure/volume/temperature (PVT) properties, and most importantly, historical well rates. The possibility to define IEs through streamline simulation stands in contrast to the ad hoc definition of geometric WAFs and simple surveillance methods used by many practicing reservoir engineers today. Once IEs are known, improved waterflood management can be implemented by reallocating injection water from low-efficiency to high-efficiency injectors. Even in the case in which water cannot be reallocated because of local surface-facility constraints, knowing IEs on an injector/producer pair allows the setting of target rates to maintain oil production while reducing water production. We demonstrate this methodology by first introducing the concept of IEs, then use a small reservoir as an example application. Introduction Local areas of water cycling and poor sweep exist as a flood matures. Current flood management is restricted to surveillance methods or workflows centered on finite-difference (FD) simulation, where areas of bypassed oil are identified and then rate changes, producer/injector conversions, or infill-drilling scenarios are tested. However, identifying and testing improved management scenarios in this way can be laborious, particularly for waterfloods with a large number of wells and/or a relatively high-resolution numerical grid. For mature fields that have potential for improved production without introducing new wells or producer/injector conversions, the main goal is to manage well rates so as to reduce cycling of the injected fluid while maintaining or even increasing oil production. Reservoir engineers have no easy or automated way to identify injection patterns, well-pair connections, or areas of inefficiency beyond simple standard fixed-pattern surveillance techniques (Baker 1997; Baker 1998; Batycky et al. 2005). Such methods are approximate at best owing to the need to define geometric allocation factors and fixed patterns, which suffer from "out-of-pattern" flow. These limitations are removed through streamline-based surveillance models (Batycky et al. 2005). By adding a transport step along streamlines, streamline simulation (3DSL 2006) can additionally identify how much oil production results from an associated injector, quantifying the efficiency down to an individual injector/producer pair. It is this crucial piece of information—the efficiency of an injector/producer pair—that allows an improved estimation of future target rates, leading to improved reservoir flood management.

An approach to waterflood optimization: case study of the reservoir X

“…The streamline-based simulation workflow used for computing well allocation factors (WAFs) and injection efficiencies was proposed by Thiele and Batycky (2006). These efficiencies were used to optimize oil recovery by effectively reallocating water available for injection.…”

A pattern-based approach to waterflood performance prediction using knowledge management tools and classical reservoir engineering forecasting methods

“…This allocation is rather simplistic that ignores deviations due to the heterogeneity in the reservoir. If the heterogeneity and connectivity in the reservoir can be well-defined, a more accurate set of allocation factors of the produced volumes could be determined through methods such as 3D streamline simulation (Thiele and Batycky, 2006). Figure 4 illustrates the allocation process for inverted water-injection patterns:…”

Industrial Compositional Streamline Simulation for Efficient and Accurate Prediction of Gas Injection and WAG Processes

“…This allows decoupling of the transport problem into a sum of 1D problems along streamlines. Previous studies have shown that streamline methods can predict the global sweep of water floods in heterogeneous reservoirs effectively (see, for example, Thiele and Batycky (2006)). A preliminary extension of streamline methods to compositional simulation was discussed in Thiele et al (1997).…”