Handling Real-Life Well-Production Instabilities and Uncertainties in Digital Fields: A Practical Application From Congo and Gabon
- Dennis Denney (JPT Senior Technology Editor)
- Document ID
- Society of Petroleum Engineers
- Journal of Petroleum Technology
- Publication Date
- May 2009
- Document Type
- Journal Paper
- 60 - 61
- 2009. International Petroleum Technology Conference
- 1 in the last 30 days
- 40 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||Free|
|SPE Non-Member Price:||USD 17.00|
This article, written by Senior Technology Editor Dennis Denney, contains highlights of paper IPTC 12545, "How To Handle Real-Life Well-Production Instabilities and Uncertainties Within Digital Fields: A Practical Application From Congo and Gabon," by Jacques Danquigny, SPE, Marc Tison, Guennole Ouaye, Emmanuel Segui, and Michel Vie, SPE, Total, prepared for the 2008 International Petroleum Technology Conference, Kuala Lumpur, 3-5 December. The paper has not been peer reviewed.
Digital fields use modeling tools to optimize production. Modeling is challenged by transient and instable flow regimes. Optimization requires a further level of accuracy: It implies the modeling of an envelope of production points. This study highlights parameters having the main influence on well-model outputs. It shows the accuracy of these models that can be used to carry out virtual flow metering. It also provides guidelines to keep well-flow models updated, which is key to sustaining real-time production optimization.
To address the production-optimization challenge on a daily or even real-time basis, many oil companies are developing digital-fields technology. Total has developed a fit-for-purpose tool, which was installed as a pilot in a mature field in the Congo. This performance-modeling tool provides data visualization that is used as an alarm system.
One workflow is automatic modeling of production tests. The modeling tool automatically detects new production tests in the corporate production-database-management system (PDMS). Downloading the averaged results of tests from PDMS, along with the reservoir static pressure and the productivity index (PI) of the well from the reservoir database, the test is modeled with two computation techniques, system calculation and quick-look.
Studied Set of Production Tests This study considered production tests of gas lifted wells in a mature Congo field between January 2007 and February 2008. This period was long enough to obtain a significant number of tests per well and short enough that static pressure and PI remained relatively constant for each well. Proper instrumentation and monitoring were required for all real-time inputs and outputs of the PI application. For all the statistical computations, these PI tags were resampled with a constant timestep of 120 seconds.
Some of the tests were not valid because operational events made them unrepresentative. Therefore, these tests were not considered in the study. On average, five tests were studied per well, with a total of 254 tests analyzed. Typically, test duration was 12 hours.
Flow-Production Regimes and Production-Parameter VariabilityAmong the 254 tests, some showed fairly constant conditions whereas others were unstable. It was expected that the unstable features affected the quality of the modeling. Dealing with unstable wells required more-accurate definition of the different flow/production regimes observed in gas lifted oil wells. An unstable periodic-production regime does not necessarily have a sinusoidal trend, and transient tests can undergo a change in both signature and trend. Once the flow/production-regime map was established, each production test of the survey was classified according to this map. Although this mapping was somewhat subjective, numerical criteria were defined to classify each test, as detailed in the full-length paper.
|File Size||119 KB||Number of Pages||2|