Perforation Performance Simulation in Complex Multicasing Scenarios
- Thomas Burky (Halliburton) | G. G. Craddock (Halliburton) | Joshua Lavery (Halliburton)
- Document ID
- Society of Petroleum Engineers
- SPE Symposium: Decommissioning and Abandonment, 3-4 December, Kuala Lumpur, Malaysia
- Publication Date
- Document Type
- Conference Paper
- 2018. Society of Petroleum Engineers
- 2.2 Installation and Completion Operations, 1.6 Drilling Operations, 1.6.12 Plugging and Abandonment, 2.2.2 Perforating, 2 Well completion, 3 Production and Well Operations, 2.1.3 Completion Equipment
- plug and abandon, over penetration, plug failure, leak path, perforation penetration
- 2 in the last 30 days
- 149 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 9.50|
|SPE Non-Member Price:||USD 28.00|
A common procedure within plug-and-abandonment (P&A) application is to use explosive shaped charges deployed by hollow steel carrier perforating gun systems to perforate holes into single or multiple concentric casing strings to allow a flow path for isolation fluids such as cement or resin to be pumped from surface into annular spaces to provide a permanent pressure seal to prevent the flow of fluids. When planning these selective perforating operations, there are two key performance results which are crucial to predict: (1) determine whether or not the perforation tunnel will penetrate through the wall of the outer most casing string, thereby risking a release of fluids/preventing pressure containment; (2) determine the entrance hole diameter (of each hole created around the circumference of the well) into each casing string – including through multiple concentric casing strings – to ensure that adequate flow area (and flow velocity) are present for the successful pumping of cement in a squeeze operation, or for a washing operations to prepare an annular space for a cement squeeze. With the need to consider the effects of all casing characteristics and positioning of annulus materials, the penetration estimate is far outside of the scope of published API RP-19B data. A calibrated and field-usable simulation method has been developed to help bridge the gap between known penetration data and novel scenarios.
The calculation engine is built on wellbore materials categorized into six classes; each class having three parameters for each perforator charge: maximum penetration depth, a power operator for efficiency of penetration, and a power operator for efficiency of hole size. Further, this simulation process has been able to leverage the very complex behavior of big hole (BH) charges in long gun clearances to predict limited penetration results in very challenging P&A scenarios. Together, 24 coefficient values for each perforator charge are calibrated using a database of nearly 2,000 historical perforator charge tests conducted with actual wellbore materials.
The resulting tool can predict penetration depth and hole size for scenarios ranging from one to six layers of steel of varying thickness and strength. In addition, the tool can handle fluids or concrete in annular spaces of up to several inches thick. The use of power operators for these functions enables the realistic replication of nonlinear penetration events in multiple layers, as well as hyperbolic functions for BH performance in long clearances. Although generic coefficients derived from a charge's design type were observed to produce reasonable performance estimates in these complex perforation scenarios, calibration by actual testing using varying materials helps to improve these predictions.
This paper provides details about a novel approach used to estimate perforator performance in highly complex, multicasing scenarios that are similar to those often found in P&A operations. The approach leverages the understanding that penetration results are extremely nonlinear along the jet path and vary by charge design and materials. The calibration of the method using a large set of actual test data adds valuable realism to the results.
|File Size||1 MB||Number of Pages||9|