Reservoir Pattern Surveillance

Use studioSL's surveillance technology to make low-cost operational decisions on mature floods. Our data-driven surveillance technology provides ROIP distributions, ML-embedded forecasts of rate targets, shut-ins, reactivations, and/or producer-injector conversions. There is no tedious history matching, meaning that you can rapidly go from model inputs to model results.

For the 16th Matzenfield peripheral waterflood in Austria, surveillance along with our floodOPT workflow resulting in an oil production increase of 3% (35,000 stb) in one year, the Belridgefield in California showed a reversal in oil decline, the Wilmingtonfield in California reduce oil decline from 20% to 2% over one year, in the 8th Tortonian in Austria incremental oil production was 35,000 STB in 30 months, and in Argentina the oil decline for a water flood was reduced from 16% to 0% for the following 3 months.

More Info:Brochure,Pricing,SPE-Webinar, Surveillance Book, YouTube Videos

Support:Knowledge Base,Tutorials,Training

The Reservoir Surveillance Workflow

 

Well locations, production history, and geology. Import historical production, well locations, and perforation data from OFM, Petrel, geoSCOUT, or AccuMap. Import 2D Zmaps from Petrel or Petra to build a 2D or 3D grid, or directly import a 3D geomodel from Petrel, Skua-Gocad, or other geomodeling software using Eclipse-like grid formats.

 

Compute streamlines. With well locations, the injection and production history, and a grid defined, calculate the streamlines throughout history. The image to the left shows streamlines, as colored by originating injector, for one time interval only. Because historical well rates change, streamlines are computed at each timestep that history is defined at.

Surveillance does not involved the transport of fluids along the streamlines so calculations are fast and efficient.

 

Flux Pattern Map.Collapse the streamline bundles between each well-pair and represent this as a graph. This is Streamsim's patented (US Patent 6,519,531)Flux Pattern Map to quantify the strength and the allocation factors between wells. Since streamlines change with time, so too will the FPmap, giving a dynamic pattern picture of your flood. These maps can be combined for a range of timesteps to then represent average patterns and average allocation factors over the time range.

 

Injection Efficiency.Once well-pairs are identified and the well-rate allocation factors are calculated, studioSL can summarize the performance of the injectors via an injector efficiency plot. Each point on the plot represents an injector and is associated with an offset oil rate produced on the y-axis. The IE-plot is powerful diagnostic of the efficiency of each injector in a field.
   

ROIP Distribution.Given an Original Oil in Place (OOIP) distribution, apply Streamsim's unique streamline region based material balance (SLR-MB*) simulator on each dynamic well-pair to generate a Remaining Oil in Place (ROIP) distribution. This will highlight areas of low vs high ROIP as driven by the injection/production history.

Shown on the right is a comparison between the OOIP and the ROIP after integrating all the historical production into per-pattern material balance calculations. Red areas are high values, blue areas are low values, and purple areas are zero values where gas cap or water zones are located.

* SLR-MB is protected by US patent No. 10,726,173 B2

OOIPROIP
Well forecast Rapid Forecasting. Using regression techniques, historical well rates are correlated to local grid properties to compute individual well-level forecast of oil, water, and gas rates. This is useful for rapid forecasting of existing well base declines, shut-ins, reactivations, or producer-to-injector conversions. Forecasting of a surveillance model can also be combined with Streamsim's floodOPT workflow to forecast the impact of new well rate targets that promote sweep and demote fluid cycling.
Rapid Forecasting of Producer-Injector Conversions. Using a flow-based surveillance model we can test shutting in a producer and converting to an injector. For the picture on the right, we tried two active producer locations that each have +98% watercut.  For each case we shut in the producer and reopened the well as an injector but on each producer's last voidage rate.  Thus there are 3 simulation forecasts.  The basecase do nothing (blue lines), converting location 1 only (red lines), and converting location 2 only (green lines).  As seen in the right figure, the incremental oil response is better for location 1 than location 2.  The bullets are field historical rates. 

 

For a short white paper on field results with our technology, please see here. See our knowledge base item for more information on a Flow-Based Surveillance model vs a Flow Simulation model here.

See our April 2008 SPEREE paper or our JPT Jan 2010 Distinguished Authorarticle for additional technical details.

Contact in USA

Corporate Headquarters 
StreamSim Technologies, Inc.
865 25th Avenue
San Francisco, CA 94121
U.S.A.

Tel: (415) 386-0165

Contact in Canada

Canada Office 
StreamSim Technologies, Inc.
Suite 102A - 625 14th Street N.W.
Calgary, Alberta T2N 2A1
Canada

Tel: (403) 270-3945

Important Links