New Realistic Hydraulic and Technogenic Fracture Modeling Approach in Full-Scale Dynamic Models
- K. Yu Bogachev (Rock Flow Dynamics) | V. Shelkov (Rock Flow Dynamics)
- Document ID
- Society of Petroleum Engineers
- SPE Russian Oil and Gas Conference and Exhibition, 26-28 October, Moscow, Russia
- Publication Date
- Document Type
- Conference Paper
- 2010. Society of Petroleum Engineers
- 5.5.8 History Matching, 4.1.2 Separation and Treating, 2.2.2 Perforating, 4.3.4 Scale, 3.2.3 Hydraulic Fracturing Design, Implementation and Optimisation
- 2 in the last 30 days
- 408 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 8.50|
|SPE Non-Member Price:||USD 25.00|
Realistic simulations of full-field models including hydraulic fractures made at producing and injecting wells, and fractures created automatically, when the bottom hole pressure at injectors exceeds the overburden pressure remains problematic in the most commercial hydrodynamic simulators. The existing software can either provide a very detailed fracture dynamics description for one or few wells with high grid resolution and precise fracture geometries, or provide full-field modeling with many wells and skin-factors used to account for hydraulic fractures. With model grids getting finer resolution it becomes more common to have the fractures half length to exceed the grid block sizes.
In the existing hydrodynamic simulators it is the realistic description of large fractures that remains the most problematic issue. In the absence of adequate modeling solutions the reservoir engineers are forced to come up with a set of unphysical approximations causing significant distortions of model properties and deterioration of its forecasting power. The new technology proposed in this paper is based on building networks of new ("virtual??) well perforations in the model grid blocks intersected by the fracture. With the help of this new approach, reservoir engineers can realistically model fracture dynamics in full-field models with hundreds and even thousands of fractured wells.
Let's consider a case when the half length of hydraulic fracture or auto fracture at injector is larger than the model grid block size it is located in. In this situation the introduction of negative skin-factors is not going to be enough for reliable description of the model hydrodynamics. To overcome this problem, the existing practices for handling large fractures encourage engineers to build a set of high permeability "channels?? along the assumed fractures directions by combining manually grid blocks of the original model and significantly increasing the permeabilities in them, as shown in Fig. 1. Such approach can partially reproduce the effect of fast fluid or gas breaking through the fracture, but slows calculations significantly. More important, since the channels are formed from regular (or locally refined) model grid blocks, the passage of liquid or gas inevitably gets modeled by filtration mechanism.
The physics of filtration dictates that liquid or gas can penetrate through a fracture to the well only by pushing fully or partially (according to phases relative permeabilities and capillary effects) all the movable reserves out of the grid blocks used to form the channel. In reality, taking into account large sizes of propant granules and relative weakness of capillary effects, the breaking through the fracture happens quickly and can be more properly described by hydraulic rather than filtration mechanism. The reserves in model grid blocks intersected by the fracture get slowly (not instantly!) drained through the fracture surface. Such fundamental differences in fluid dynamics causes development of systematic biases in the distribution of the remaining reserves and eventual deterioration of model's forecasting qualities.
|File Size||655 KB||Number of Pages||4|