Application of Linear Programming to Reservoir Development Evaluations
- K.K. Lo (Arco Alaska Inc.) | G.P. Starley (Arco Alaska Inc.) | C.W. Holden (Arco Alaska Inc.)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Engineering
- Publication Date
- February 1995
- Document Type
- Journal Paper
- 52 - 58
- 1995. Society of Petroleum Engineers
- 0 in the last 30 days
- 292 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 35.00|
A production forecasting model using linear programming concepts has beendeveloped to generate composite forecasts from multiple streams producedthrough a common facility. The model treats petroleum processing as a linearoptimization problem to maximize oil production under a set of independentconstraints. The model represents production streams competing for limitedfacility capacity on a consistent basis. This approach allows resolution ofproduction streams of varying character (GOR, GLR or WOR) into a composite rateforecast while optimally utilizing facility capacity. In particular, the modelmay be used to forecast production of several unrelated fields producingthrough a common facility. The model was applied to forecast production forthree Alaska North Slope reservoirs (Kuparuk, Lisburne and Point McIntyre) fordevelopment and facility expansion evaluations.
To validate the method, production forecasts generated with the linearprogram (LP) model were compared with results from full-field reservoirsimulations. The linear program model forecasts for each field comparedfavorably with simulator predictions. Validation demonstrated that the modelcould be used in place of fullfield simulations to evaluate impacts on totalproduction of changes in facility capacities or field development projects.Also discussed in the paper are different applications in productionforecasting for which the model may be used.
The linear program model runs quickly on a workstation or mainframecomputer, saving considerable computer time relative to full-field simulation.The model may also be effectively used as a case screening tool. Multiplescenarios can be rapidly evaluated using the model, and at a lower cost thanrunning repeated full-field simulation cases.
Full-field reservoir simulations are frequently used for evaluating benefitsand impacts of various development scenarios or changes to processingfacilities. Field-scale, black-oil simulations are quite effective inforecasting production from waterflood, immiscible gas injection, developmentdrilling, facility expansions, and production acceleration resulting frominfill drilling. However, they are severely limited when trying to forecastproduction from multiple, separate fields producing through a common facility,miscible gas injection projects, "net" incremental reserve benefitsfrom infill drilling, or impacts of sharing producing facilities among multiplereservoirs. These limitations can be critical when trying to estimate netproduction benefits from these projects and impacts to primary and/orwaterflood production under facility constrained producing conditions. Further,simulator development costs and model run times for repeated testing andapplication can become prohibitive when trying to extend full-field simulationcapabilities. A simplified technique was sought to reconcile forecasts ofcomplex depletion processes under constrained producing conditions whileobtaining a reasonable approximation of more rigorous simulations.
For these reasons, a production forecasting tool based on linear programmingconcepts has been developed to overcome some of the shortcomings of using ablack-oil, full-field simulator for multiple sensitivity studies.
|File Size||532 KB||Number of Pages||7|