Prediction and evaluation of remaining useful life of casing with corrosion defects

ObjectiveAiming at the problem of severe corrosion damage to casing during oilfield development, which compromises safe operation, a method for predicting the remaining useful life of casing was established to realize accurate prediction of the remaining useful life and safety assessment of casing with corrosion defects.MethodsBased on the von Mises equivalent stress yield criterion, a 3D coupled finite element model of the casing-cement sheath-formation system was established. The influence of corrosion pit depth and angle on stress distribution and collapse external load was systematically analyzed. Furthermore, a particle swarm optimization-Gaussian process regression (PSO-GPR) model was developed to predict casing corrosion rates using field data.ResultsThe results indicate that the Mises stress of the casing increases with corrosion depth and decreases as the corrosion pit angle enlarges due to the mitigation of stress concentration. The PSO-GPR model demonstrates superior prediction performance for corrosion rates, with an average fractional error of 0.78%. By integrating numerical mechanical analysis with artificial intelligence prediction, a comprehensive remaining useful life evaluation framework for corroded casings was established, providing a foundation for risk prevention in oilfield operations.