Testing and Verification of Well Barriers: Verifying cement plug using borehole seismic
Abstract
The aim of this thesis is to demonstrate how seismic waves propagate in wells, and to investigate the potential of using borehole seismic as an alternative to conventional cement plug verification. An alternative verification method might be necessary to perform full plug and abandonment (P&A) operations from riserless light well intervention (RLWI) vessels.Testing, verification and monitoring of technical barrier elements are important to maintain well integrity. The barrier situation has to be kept under control in all phases of a well's lifetime. The focus on long-term integrity is increasing as more wells are getting ready to be plugged and abandoned. New and more effective P&A solutions are desired by the industry. Rigless P&A operations would save time and cost, but current technology does not allow RLWI vessels to do the entire operation. The work in this thesis is based on previous work where it is suggested how to place cement plugs in a well using a vessel. To act as a part of the barrier envelope, the plug needs to be verified and tested according to the NORSOK D-010 requirements. Tagging is often used for cement plug verification. Conventional tagging operations are not possible from RLWI vessels, and this thesis suggests borehole seismic as an alternative solution for cement plug verification.The alternative verification method is based on generating pressure waves using a seismic source, and detecting the reflection from the cement plug using receivers run on wireline. By doing this it might be possible to verify both the top of the plug and the cement integrity. To investigate the wave propagation and reflections from the plug, well models were created in MATLAB. The models were imported to the Madagscar software, and a 2D finite difference method was used to model the wave propagation. The wave response from different cement plug scenarios are simulated by adjusting the models and input parameters. These scenarios include plugs with varying degree of cement settling and mud contamination.To confine the simulation time and to avoid numerical instabilities, the model had to be simplified. This has increased the uncertainty in the results. Using the simulation result, the depth of the top of the plug and the wave velocity in the cement were calculated with small errors. Knowing the cement velocity might make it possible to determine the grade of cement hydration. When mud contaminated the top of the plug, the top was not easily identified. The precision of the results is dependent on the frequencies used in the simulations. Tube waves are expected to have great importance when using borehole seismic for plug verification. The result could not identify these types of waves. A reason for this might be that the modeling is in 2D while the well is 3D.To make final conclusions on the potential of the alternative verification method, new simulation tools fitted for well modeling should be developed. Field experiments could also give valuable results. The optimal frequency range for cement plug verification has to be determined through a more detailed study. Solving the verification challenge is only one step on the way towards completely rigless P&A operations. Remaining challenges include problems related to tubing removal, logging behind several casing strings, and control line identification and removal.