Project of quantitative risk analysis for designs of permanent abandonment of wells

On the Norwegian Continental Shelf, thousands of wells are approaching the end of their productive lifetime during the coming decades. Permanent well abandonment requires establishing cross-sectional barriers of a certain length that prevent cross-flow of fluids and do so indefinitely. As the condition of the original well cement can be uncertain or difficult to establish, limitations in current technology implies that expensive installations are often required for a considerable part of the plug and abandonment (P&A) operation. Finding more efficient, cost-effective and safer methods for well abandonment is of great importance for the industry and the society. Ideally, a P&A well design should lead to a leakage risk that is as low as possible, while being as time- and cost-efficient as possible. A risk-based approach to P&A should reflect the well-specific risk of leakage in the abandonment design, covering both barrier verification and barrier length, thereby allowing optimized abandonment design and the possibility of evaluating alternative P&A well designs and technologies. Risk-based approaches to P&A are under development, but to fully enable such an approach there is a need to better understand well barrier performance, limitations and capabilities of current verification methods, and also a need to define acceptance criteria that consider both performance, verification and possible environmental consequences of leakage. The main objective of this project has been to further strengthen the risk-based approach to P&A by addressing these knowledge needs. With this objective in mind, we have performed an evaluation of consequence modelling of leakage from plugged wells in a risk perspective, developed a model that enables well log results to be included in leakage modelling, and performed full-scale experiments that provides a foundation for describing permeability variation in cement that has not had challenges with displacement or interaction with formations. The results contributed to improved risk based assesssments of leakage risk from plugged wells. The methodology have been tested and demonstrated on field cases in Norway and Brazil.

This project was the first part of the P&A Innovation Program JIP (Joint Industry Project) and was an initiation for many of the activities performed therein which has since been expanded on based on the results from this project. As such, this project has been very successful for NORCE in terms of increased research activity on P&A and has benefitted from frequent dialog with the operator companies participating in the project. The full-scale experiment performed for this project was the first application of the P&A laboratory built as part of the Norwegian P&A Laboratories national infrastructure project. Further experiments are being performed with industry funding to expand on the knowledge gained from that experiment, and strengthens NORCE as a discussion partner to the industry and as a provider of test facilities. Parts of the models developed in the project have been implemented into the P&A Leakage Calculator developed at NORCE in an earlier project partly financed by the Research Council, improving the capabilities of the tool. Further implementation is ongoing with discussion with financing partners on how to adapt the implementation to their internal procedures. Implementation was not part of the project, but when finished the results from the project will be more easily exploited through this tool. The project also provides results that should influence the development of standards for plugging of wells. However, such changes take time to accept and adopt. The project has also resulted in the development of proposals for follow-up projects that close further knowledge gaps and missing capabilities of existing tools for consequence analysis.

Risk based approaches to permanent plug and abandonment (P&A) are by many perceived as the way forward, as the existing rule-based approach used in guidelines such as NORSOK D-010 across the world requires updating as technology develops and exceptions when experiencing difficult conditions. There have been several suggestions for risk-based approaches to P&A, but in this discussion more knowledge related to the reasoning behind what is acceptable is needed. This project, together with its connected project in Brazil, will develop models supporting a risk-based approach, and demonstrate and evaluate the applicability of such an approach on cases. The Norwegian activities will include full scale experiments to investigate how the permeability of a cement plug varies along its length, analysis of what leakage rates could go unnoticed by the common verification methods, and how the acceptance of P&A designs and verification methods should be based on risk analysis and quantification of possible consequences.