Abstract
Leakage pathways may develop in wellbores during construction, production, or during and after plug and abandonment (P&A). These pathways are created due to events and conditions during cementing operations, or because of physical and chemical changes after cementing such as changes in temperature and wellbore pressures, and deterioration of the cement. Common leakage pathways develop inside the cement sheath, or as microannuli along the cement-tubing interface. Numerous evidence exists showing that wellbores leak, but there is no verified method to determine if a well will leak or not. To ensure long term wellbore integrity, leakage risks need to be evaluated for plugged and abandoned wells.
To evaluate leakage risks from plugged and abandoned wells, numerical finite element models have been developed and used to investigate leakage scenarios during the life of the well. Currently, little work has been done to verify finite element numerical models with experimental data regarding flowpath size in cement sheaths. The aim of this paper is to model previously published experimental data to determine if the finite element models can accurately predict leakage potentials.
Two lengths of cemented annuli were modeled, each with conventional and expanding cement to replicate the Aas et. al. [1] experiments. The numerical results show that the simulated microannuli overestimate flow rate compared to experimental data, indicating that flow path dimensions and/or fluid friction factor does not accurately represent the fluid flow in the experiments.
