The computer models generated for visualisation take the vast heterogeneous reservoir's information and based on various attributes accurate forecasts of stock and production levels can be calculated. A conceptual model of an oil reservoir is one which is made up of a fractured grid where the un-fractured sections of rock are cubed in shape. The reservoir's dimensions can be hundreds of meters across and although they are usually far shallower than they are wide they can still reach depth of over 100 meters.
Computer processing power and memory restricts the 3D visualisation of precise details at very small block sizes as some reservoirs have over 10 billion of such grid-blocks. In order to allow for computer visualisation, reservoirs are broken into larger grid blocks which can be processed by the computer, sometimes over 200m3. The use of uniform grid blocks allow engineers to simplify the structure of these large stochastic reservoirs by omitting a lot of the boundary and fault detail resulting in a simplified and less complex 3D model. The space between the grid-blocks which are sometimes referred to as " mortar joints ", allow for isolation of particular blocks where various simultaneous calculations can be carried out such as pressure, temperature or time-of-flight readings. There are three main types of fractures which can be found in oil reservoirs and referred to as:
- Dilating Fractures
- Shearing Fractures
- Closing Fractures
Dilating Fractures
Dilating fractures, as shown below in Fig 1, occur when sections of matrix rock have separated at 900 one another and is sometimes referred to as a "joints".
Figure 1
Shearing Fractures
Shearing fractures occur when sections of the rock matrix slide in a parallel fashion to one another and are sometimes referred as " faults ". There are instances where both dilating and shearing occurs simultaneously. This is illustrated in Fig 2 below.
Figure 2
Closing Fractures
Closing fractures have closed due to the pressures of surrounding matrix blocks and gravity. During flooding when injecting with sufficient pressure these closed fractures can be pushed apart to allow fluid flow between the re-instated fracture. This is illustratedd below in Fig 3.
Figure 3
Computer performance has improved a great deal in later years and this has given birth to more detailed and precise 3D simulators visually detailing such things as fault lines and surface textures at high resolution. These fractured reservoirs consist of both the porosity of the rock matrix and the porosity of the fractures. Biot-Barenblatt in 1960 developed the idea of a 'dual-porosity' model. A realistic view of a section of reservoir rock matrix and its conceptual computer grid structure is illustrated using this model and is shown below in Fig 4 below.
Figure 4
Reservoir grid models can also be simulated using a " corner point " system where the grids follow the contours of blocks and fault lines instead of blocks of uniform dimensions. The irregular grid shapes produced in these models have more sides and have obtuse angled corners and can allow the simulator to compute fluid flow in more directions which can results in a more precise simulation. As reservoir models are heterogeneous when it comes to their various characteristics, such that the boundaries of permeability, they are normally different to the boundaries of porosity, temperature and pressure throughout the length, breadth and depth of the reservoir. In order for simulators to perform numerous calculations homogeneous regions of averages of these values can be used to form a simulator grid structure. The macroscopic grid model designed by E.J.Warren & J.P.Root is suitable for less precise calculations based on dual porosity simulators but each of these grids can be isolated whereby it is broken down again into another grid block system depicting a higher resolution of each of the previous levels of the grid block structure accommodating the simulation of greater levels of detail and precision. An example of this model can be seen in Fig 5 below.
Figure 5
They later improved this macroscopic grid model by introducing the concept of the passaged of hydorcarbons by means of capilary contact through each of the grid blocks within the model. This is shown in Fig 6 below.
Figure 6
