How to build a Hybrid Physics-Data Virtual Flow Metering Solution for Production Monitoring with Cognite Data Fusion®

A global energy company wanted to improve production allocation across six producing wells. The wells are not equipped with multiphase flow meters and therefore, Cogntie Data Fusion® was used to create an indication of production through a virtual flow meter (VFM). The results are a near real-time, cloud-based VFM solution that provides Operators with a quantitative tool to justify a new well test or change injection strategy. The VFM can be used to improve backallocation, or monitor and alert for water breakthrough.

Outdated and biased data

Current industry practice for VFMs is to tune the model using the latest well test and use correction factors to calibrate frictional and gravity terms, which leads to predictions biased towards the latest well test. Additionally, for this particular field, well tests are performed as infrequently as semi-annually. Therefore, developing an accurate VFM is difficult.

Using Cognite Data Fusion®, the company was able to combine well measurements with fundamental flow equations through the tubing and choke. Well test data was used to calibrate these equations by tuning parameters such as the friction factor and choke flow coefficient. The models are auto-tuned based on the incoming and past well test data. Additionally, the models require a fluid properties table and wellbore deviation survey. When performing predictions, typical real-time measurements such as downhole and well head pressure/temperature and choke position are utilized.

A near real-time, cloud-based VFM solution

Using Cognite Data Fusion®, the operator developed a near real-time, cloud-based VFM solution. The approach is based on combining well measurements with fundamental flow equations through the tubing and choke. Well test data is used to calibrate these equations by tuning parameters such as the friction factor and choke flow coefficient. The models are auto-tuned based on the incoming and past well test data, and also utilizes a fluid properties table and wellbore deviation survey. When performing predictions, typical real-time measurements such as downhole and well head pressure/temperature and choke position are utilized.

The approach was tested on a well in a nearby field which had more frequent well tests and then extended to the considered field, which had very few calibration data points (~2 well tests/year). The results were validated using left-out well tests and demonstrated  mean absolute percentage (MAPE) errors consistenly within a 10% range.

The solution was then deployed on Cognite Data Fusion’s  scalable data platform to templatize calculations. The VFM was deployed to the first well within 3-4 weeks, then scaled to all six wells in less than two weeks. The solution can:

• Run calculations in near real-time, using relevant sensor and production data for the asset
• Effiently scale solution  to new wells through the use of templatized calculations
  Provide  full access control to the solution owner
• Minimize the need for on-premise modifications to the asset

Justify new tests, change injection strategy,  improve backallocation, or monitor for water breakthrough

The solution offers capability to perform  calculations in near real-time using relevant sensor and production data for the asset and minimizing  on-premise modifications  to the asset. Operators now have a quantitative tool to justify a new well test, change injection strategy,  improve backallocation, or monitor and alert for water breakthrough.