This is the first of a series of posts related to reservoir engineering topics in Kraken’s Blog. The reader can expect articles related to reservoir engineering and various simulation software applications. Other themes such as petrophysics, management, and curve analysis will also be discussed. This first post provides an overview of Reservoir Simulation and highlights key features and characteristics of a reservoir simulator and its role in reservoir management. In addition, some examples of properties visualization using Kraken are presented below.
According to the dictionary definition, simulation consists in imitating a situation or process. It can be applied to a range of technical fields, including geology and reservoir characterization which are our primary targets of interest. Over the past decades, interest in simulation has increased in areas of very large computational effort and investment demand. The exploration and appraisal of oil and gas fields fit both criteria, and reservoir simulation plays a vital role as a diagnostics tool to better understand and predict a reservoir’s behavior.
The earliest reservoir simulators were described as physical models in which sand, oil and water interacted. Also, electrical simulators characterized by the flow of electrical current and reservoir fluids were frequently used. With the advent of digital computers, reservoir modeling has reached new heights, advancing to simulators that numerically describe the dynamic behavior of reservoirs.
Initially, computer power was challenged by small memory storage capacity which in turn limited the size of the reservoir model and allowed only simulations with a small amount of data input. As computer power has swiftly increased with innovations in computing technologies, reservoir engineers are now able to create larger and more realistic models to describe what happens in a reservoir more effectively. In addition, today’s computational efficiency allows for powerful simulators to run larger models faster, and an also a combination of parametric cases (virtual experiments) to understand the reservoir behavior under various conditions.
As aforementioned, the main purpose of running a reservoir simulation is to forecast reservoir performance under different production conditions. That is, attempting to set an optimal production scheme to economically optimize hydrocarbon reservoirs. Thus, reservoir simulation is able to answer what the availability of hydrocarbons in the reserves is, how much of it can be recovered, and how quickly. Several important considerations must be taken into account to select the best development plan: number of wells and locations, surface facilities and the application of EOR methods are among them.
The tool necessary to conduct a reservoir simulation study is called simulator. Its development requires vast knowledge of petrophysical properties and physical processes in the reservoirs along with a mature knowledge in mathematical modeling and programming. Additionally, the engineer is the key player during a reservoir simulation study, once it will be responsible to carefully analyze and interpret the enormous amount of information generated during the process and refine them to reach a better reservoir description. Once it is established a good level of description, the reservoir simulator can be used to perform optimization tasks.
Reservoir simulators are created by subdividing a reservoir into finite volume elements, in a process called “discretization.” The reservoir is divided into a series of interconnected blocks and numerically solve for the flow between them. Each volume contains petrophysical properties representing the reservoir to be modeled and analyzed, and the set of these elements is denoted by the reservoir grid. The simplest grids consist of a number of equal, cube-shaped cells. Since most real reservoirs feature a complex internal structure and elements which are difficult to model- such as faults and fractures – a good approximated model will require a large number of cells.
The figure below illustrates the reservoir grid representation of the Permeability I distribution which is useful for analyzing how this property varies throughout the reservoir layers:
The equations used to describe reservoir models derive from fundamental principles, such as thermodynamical equilibrium, mass conservation, heat transfer and fluid flow, which is governed by a well-established equation known as Darcy’s Law. These flow-equations are expressed in terms of partial differential equations written in a finite-difference form, which discretizes the problem in time and space. This allows the simulator to compute the fluid flow through the entire reservoir.
The available range of simulators include black oil to compositional, and streamline simulators to finite volume methods, where the black oil simulator is the most commonly used in the industry. A suitable reservoir simulator is selected based upon the goals, type of reservoir, and production mechanisms involved.
After a series of simulations, the results can be visualized through specialized analysis tools. These tools are generally used to plot production variables such as oil and water production rates, as shown below:
Additionally, other types of data manipulations can be performed through the use of plots and grid views. The figure below illustrates the use of well selections and their representation within the grid. It allows the user to analyze well locations along with them neighborhood.
Powerful reservoir data manipulation tools such as Kraken can routine perform advanced analyses such as streamlines visualizations. Streamlines represent a snapshot of the instantaneous flow and also indicate drainage and irrigation areas associated with producer/injector wells. Streamlines are considered an important post-processing technique in reservoir simulation as they are used to evaluate the efficiency of producers and injectors as well as to support other reservoir engineering assignments such as sensitivity analysis and history matching.
In summary, reservoir simulation plays a crucial role in an oil field development, as it drives the production strategy that should be applied to maximize the operation’s profitability. Besides, the use of tools such as Kraken tied to reservoir simulations is very useful as it provides an efficient data visualization and manipulation while reducing downtime.