Your account already exists. Please login first to continue managing your settings.
The right combination of technologies can shorten development cycles, increase production, and extend reservoir life.
As operators look to optimize production from increasingly marginal assets, extend the lifecycle of their reservoirs through sustainable reservoir management targets, and make effective decisions over the allocation of capital and resources, the models and analytical processes that define reservoir simulation and history matching have rarely been more important.
It is reservoir simulation that sits at the interface between the reservoir model and the economic evaluation of fields and production strategies, and it is history matching that ensures that the static geological models are in sync with production data to predict future performance and create uncertainty profiles of the reservoir.
The result is models that are not only consistent with all the available static data – such as well log and seismic, and dynamic data like production and 4-D seismic – but also are able to reproduce historical field performance.
Forecasts generated from these models play a crucial role for operators in shortening development cycles, increasing production, extending reservoir life, and enhancing ultimate oil and gas recovery.
Rising to the challenge
The rise in computer power, such as 64-bit multicore chip clusters, parallel processing, and computer-led automation, has ensured that reservoir simulation is now practiced via the desktop and across asset teams, enabling faster decisions and a greater ability to determine the important “what-if” scenarios . Today, multimillion cell reservoir simulation models are the norm.
The same is the case with history matching technologies. Manual history matching has been replaced by robust and automated algorithms that allow the reservoir engineer to focus on developing a clearer understanding of reservoir mechanisms and their relative impact on production behavior and creating simulation models that are fully consistent with their underlying geological interpretation . In this way, operators can generate more accurate information on the operational production decisions that need to be made to extend reservoir life.
Furthermore, the last few years also have seen significant advances in uncertainty prediction tools where, through the analysis of multiple plausible realizations and uncertainty parameters, operators can better quantify the effects of uncertainties on volumes and cumulative production.
As with any fast-changing technologies, challenges remain, particularly in the need for improved ease-of-use and functionality in simulation and history matching and the need for ever more sophisticated (but easy to use) economic evaluation tools. Reservoir simulation today still remains a complex process, and it is incumbent on today’s vendors to guide users through the entire process – from preparing and analyzing original data through to the economic evaluation of results.
Similarly, history matching also can be a time-consuming and cumbersome process, with history matches often achieved through different configurations, thereby making it difficult to determine which model is correct.
Few history matching packages allow the reservoir engineer to include all of the uncertainty parameters that form part of the history match into the prediction phase: it is with this in mind that Emerson has made ease of use and functionality central to its reservoir simulation and history matching software.
The Roxar Tempest simulation software guides users through the entire simulation process. The modeling of complex wells also has been improved in the latest versions through a new segmented well approach, that allows for a detailed well model that represents the underlying physics more realistically. The company also has introduced a Todd-Longstaff solvent feature that allows for efficient simulation of CO2 floods. Fractured wells also can be simulated.
Improved functionalities for economic forecasting are vital. Aggregate rates now are calculated and displayed in the software, making it easier to interact with other programs such as economic evaluators and spreadsheets. The software also includes a new feature where lift model parameters can be regressed against observed data such as friction, fluid gravity, or gas-liquid slips.
New load-on-demand features also are key to optimizing memory usage and letting engineers simultaneously analyze many more models with complex wells and fine grids. Derived data such as group totals and well ratios can be processed on demand, greatly speeding up model loading. This increased functionality and ease of use ensures that multiple simulation runs can be loaded and managed together with observed data, with comparisons between runs and with historical data greatly speeding up history matching and sensitivity studies.
Furthermore, while static data can provide information on the reservoir framework and fluid saturations at well positions, for example, it is the dynamic data that is so crucial to extending reservoir life by charting how fluid is moving during production.
To this end, pressure, volume, temperature, relative permeability curves, and well lift curves can be graphically edited within the simulation model. Historical measurements can be entered as a table, well trajectories can be input directly as 3-D xyz files, and dynamic data entered in tables as “events,” which can be interactively edited, sorted, filtered, and viewed on a timeline. In this way, operators can track the performance of their reservoirs in real time and ensure they produce at an optimal level to extend reservoir lifecycle.
The company’s Roxar EnABLE history matching software is playing a key role in extending history matching further into the predicting of uncertainties – crucial in providing tools to the operator to extend reservoir life.
At present, this tool is the only one available that provides total uncertainty assessment. Powerful statistical techniques are used to determine multiple matches of the reservoir-to-production history and to model the reservoir’s uncertainty. All of the uncertainty parameters that form part of the operator’s history match are included during the prediction stage (there is no need, for example, to narrow the minimum and maximum range ranges), and a proxy model is used during the history match, allowing the simulation model to be easily extended into predictions to help calculate uncertainty.
These results are used with the simulator to predict how a field will perform in the future and provide measures of uncertainty about these predictions – crucial information for extending reservoir life.
The improved usability, functionality, and integration of history matching and simulation workflows are central to risk mitigation in reservoir management today.
Through a complete understanding of production ranges, simulation models that are fully consistent with the underlying geology, and effective uncertainty quantification tools, operators can look forward to highly reliable production forecasts, increased reservoir performance, and an extension of reservoir productivity.