- Reduce CAPEX with fewer test separators
- Gain accurate and frequent well-by-well flow rate intelligence to optimize production
- Reduce OPEX of manual well testing workflows
US oil shale assets execute frequent well testing in order to understand the performance of a well and associated flow rates. During the testing operation, that may take up to several days, a single well is typically routed to a test separator at one point in time. The production volume may or may not be stabilized during this period. The sequential testing pattern causes a constraint on the number of feasible tests run and may influence test results due to conditional changes. In addition, the captured data is typically archived in spreadsheets. This raises the following challenges:
- An accurate assessment of the oil, gas and water flow in individual wells under daily varying conditions becomes challenging.
- Installing several test separators for simultaneous testing is costly, but not uncommon on US land based operations. On average, one test separator can roughly equate to a quarter of a million USD capital expenditure and an average of 40k USD/pa on operational costs.
- Production volumes may need to be reduced during test runs, reducing maximum production volume.
- Data is often not stored in one single source, creating an unneeded complexity in accessing and analyzing historic data.
With our self-calibrating, real-time FLUX Virtual Flow Meter we can continuously mimic well test runs and achieve optimal flow rate predictions over time. FLUX VFM is based on a hybrid technology, combining the predictive capabilities of physical models with the self-adjusting capabilities of machine learning over time. First, the physical models in our transient multiphase flow simulator (FLUX Simulator) are calibrated based on available reference and historic test data. Examples are fluid data, temperature and pressure, in the well, on the surface or on the topsides. The available historical well test data is utilized to develop an automated model that adjusts over time, with new well test data being routinely captured in the system in real-time. Here, the tool automatically detects when a test run is started and is completed and calibrates predictions moving forward.
The Value Gain
With this, we can predict flow rates of the individual wells from data gained through the intermittent, remaining well test runs. In other words, we need a certain amount of test runs to assure our models are accurately self-calibrated. With our FLUX VFM we aim to reduce well testing frequency by up to 70%. In a scenario where every well has its own test separator, this could reduce test separator equipment investments and operational costs to well test runs by 40-60%. That said, our system takes into account many more inputs than what one would typically find in a curve-fit type of analysis within a spreadsheet. We can therefore aim to predict the individual wells’ flow rates more accurately than through the traditional spreadsheet-based system, in addition to reducing the complexity of accessing and analyzing the data for the end user.
Lastly, the FLUX VFM includes a scenario mode, in which the operator can experiment with the flow of every well, for example by reducing the separator pressure or lift gas rate, and with that analyze well flow rate behavior and optimize the production accordingly. Our online system can be accessed through the operator’s system or through our own user interface, FLUX Applied.