A Case Study: Early Event Detection of Tubing Failure

Yesterday, we looked at why data quality is important. It leads to early detection, which we will explore here with an ROI Case Study.

Tubing Failure

Early Event Detection of a tubing failure can help you regain $3M or more in your oilfield.  In this example, we examine rod pump #10 and an example of its tubing failure. With OspreyData’s Production Intelligence solution, we are able to look at sensor data such as casing pressure and tubing pressure, and also calculated sensors such as fluid load. We are also able to see dynacards and dynacard descriptors.

When we look at an example of our solution above, the grey bar at the top indicates the time at which the operator detects the tubing failure, and also indicates the time at which the workover is being conducted. The blue bar indicates the AI model which confirms that there is a hole in tubing and the orange bar is a Tubing Failure Likely Model which was triggered earlier at a 65% confidence level. As we can see, the operator detected it on the 10th, while the AI model detected it on the 2nd, which gives you a lead time of 8 days.

Dynacard Descriptors

At the same time, another cool feature is that our solution will examine your dynacard descriptors.  As the downhole dynacards come in (demonstrated below), the AI model could describe the dynacard to be having full-fillage, some possible rotation and also slight downhole friction based on the skew in the dynacard. If we look at subsequent dynacard, it could say that there is a potential high fluid level. By the time we reach the time when the tubing has failed, the dynacard has collapsed. Any SME would be able to confirm that this is representative of a tubing failure.

The takeaway is that the AI-based model is able to automatically detect all of this: that this is flumping well, there is a hole in tubing, or there is also a possibility that the traveling valve is hung open. This is especially helpful when your SME is trying to process 6000 dynacards a month. Realistically, with 6000 wells, your operators are only sampling or prioritizing some of the wells or dynacards. They may be looking at all the dynacards for their high priority well, but may be missing out on dynacards from a lower priority well, or due to sampling, and therefore, some of the events may be detected late.

Increase ROI

Late detections cost money.  As you can see in the case example above, for a well (or field of wells) that are producing 95 bbl/day with an annual average failure rate of 12% and oil prices of $50, we estimate that OspreyData’s early detection can regain annual production of 1,140 bbl and revenues of over $57,000.  If we extrapolate this over a field or play of wells, then you could be looking at an increased ROI of $3M or over $10M respectively.

If you want to maximize your oil production and increase your ROI, contact us to set up an appointment with our sales team to discuss easy implementation on your oilfields. We say easy because Production Unified Monitoring is a cloud-based tool that can be accessed from any web browser without any hardware or software installs, meaning there are no capital expenditures. Contact Us Today!

Related Articles