Developed in-house to capture the complex physics that underpin improvements in oil recovery factors, simulations of Enhanced Oil Recovery (EOR) processes provide the vital link that allows us to pursue a consistent chain of reasoning from small scale to large scale. How? By using state-of-the-art computational codes that achieve more reliable simulations and, in turn, help us optimize our developments.
EOR: The Complexity of Simulating Physical Phenomena
The implications of EOR can be huge. Whether based on chemical injections, better gas injection techniques or thermal processes, Enhanced Oil Recovery can potentially boost oil output by 2 to 20% above the average recovery factor of just 37% achieved using conventional production techniques (such as water injection).
However, achieving a reliable estimate of the additional reserves potentially associated with an EOR deployment on any given field is a major challenge that demands robust expertise. Indeed, these processes rely on a number of physical phenomena that take place at the microscopic scale within the reservoir, typically by causing localized alterations in the intrinsic properties of the oil, or in the way the different fluids (oil, water, gas) interact. Simulating these advanced physical phenomena requires first fully understanding their complexity at microscopic scale, then selecting the most significant parameters for implementation in our larger-scale simulations.
On the strength of our experimental resources, proprietary tools and workflows, we are in the vanguard of this integrated simulation chain, which extends from the laboratory through to the field development studies. Mobilizing a wide range of expertise in numerous disciplines (geology, petrophysics, physics, chemistry, geochemistry, numerical analysis), it is supported by the computational power of our supercomputer Pangea III.
Simulating EOR From the Laboratory to the Field
Our simulation workflow for EOR processes consists in three main phases. The sequencing of these phases enables us to identify and calibrate the key parameters that describe the physical phenomena involved in the various EOR techniques.
1) Small-scale simulation – Small-scale simulations provide a detailed and accurate understanding of the physics involved in the EOR process at the microscopic scale, by modeling complex EOR experiments performed on core specimens a few centimeters long by TotalEnergies’ petrophysics laboratory. The aim of this simulation is to reproduce the experimental results with the highest possible fidelity. To do so, we deploy a high-performance pair of in-house tools developed through several years of R&D and unmatched on the market:
- TPP (Total Prototyping Platform), an R&D reservoir simulation tool designed to test new physical models developed by our team of physicists and numerical analysts. The TPP has the capability to integrate extremely complex algorithms and numerous simulation functionalities;
- COREMATCH, an automatic history-matching tool for core-sweep experiments to calibrate the physical parameters that best model the experimental data by making use of the TPP simulation results via an optimization algorithm.
Thanks to the extraordinary computing power of Pangea, the hundreds or thousands of simulations needed to history-match our EOR prototype codes can be launched simultaneously and run in parallel.
2) Upscaling – An extremely detailed formalization of physical, physical-chemical and even geochemical phenomena was used for the initial simulations, which were performed on grid cells covering a few millimeters. This formalization cannot be applied for field-scale reservoir simulations, in which each cell represents several tens of meters. Although the physical model must sometimes be simplified, these parameters must always be extrapolated for upscaling purposes and to avoid systematic bias.
3) Large-scale reservoir simulation – The large-scale model is built using INTERSECT, a new-generation reservoir simulator of which TotalEnergies has been a co-owner (alongside Chevron and Schlumberger) since 2012. Having access to the source code of this model means we are able to enrich our in-house version with the new EOR functionalities developed by our R&D even before they are implemented in the commercial version of the tool. This gives us a decisive advantage, namely, the ability to quickly meet the needs of studies to design large-scale pilots or developments.
Higher-Fidelity Simulations of EOR to Optimize Our Developments
We have been using this workflow, enriched with our prototype codes, for several years now, and it has given us a lead in the field of EOR simulation. By sustaining the momentum of continually implementing better and better physical equations into our computation codes, we can achieve more reliable predictions of the incremental reserves that EOR processes will yield. For example, our workflow demonstrated some simple EOR techniques to be economically efficient, whereas the classical physics approach showed that they would not be economically viable.
This has led to a number of noteworthy advances in recent years. For example:
- A functionality allowing a very complete simulation of Surfactant-Polymer (SP) chemical injection in TPP, including simulation of the microemulsion phase generated in situ by the SP mixture, even in the presence of gas, using a four-phase formalism currently non-existent in any commercially-available simulation tool. This advanced code allowed us to design the pilot to test SP chemical EOR – successfully – on ABK (Abu Dhabi) in 2014;
- New functionalities developed in the TPP then implemented in INTERSECT to simulate early-stage tertiary polymer injection (i.e., after brief waterflooding) by installing several formalisms specific to the presence of polymers, namely a hysteresis formalism to model the non-monotone change in relative permeability that is observed when a polymer remobilizes an oil, and a dependency function relating relative permeabilities to local concentrations of polymer in the model. We are currently taking advantage of these functionalities to design future pilots and deployments of the process on oil fields in Africa;
- Simulating Water Alternating Gas (WAG) injection, an EOR technique already deployed for fields in the Middle East and Brazil in particular, has also been made possible through the implementation of a hysteresis formalism extended to oil-wettable contexts, to faithfully reproduce the cyclic changes in residual water and gas saturations. Implemented in TPP then in INTERSECT, this functionality has successfully matched the production history of the WAG pilot trial run since 2011 on Al Shaheen (Qatar). Today, the functionality is being used to model the possible extension of the process to this giant field, along with other innovative simulation functionalities such as SWIM® (Smart Water Injection Method, consisting in injections of water with low salt concentrations or modified ionic composition), and polymer injection.
An Essential Link in the Exploration & Production Chain
Chemical EOR: an Ever-Growing Field of Application