RBDO MODEL APPLIED TO WELL CASING SELECTION CONSIDERING RANDOM LOAD SCENARIOS

Resumo

This paper addresses the coupling of reliability-based models with optimization routines in the
selection of casing tubulars, regarding the failure modes defined in API/TR 5C3 code. The casing system has
great relevance in the construction and operation of oil and gas wells, considering the high costs associated with
its material selection and execution. In this context, regulatory agencies have become more rigorous and are
moving towards the adoption of probabilistic criteria for well integrity evaluation. Methodologies based on
structural reliability, taking into account uncertainties associated with the design variables, have been
disseminated in normative documents and adopted in the industry, with a positive impact on the design practice.
The proposed solution comprises integer programming routines combined to a probabilistic casing design model,
dealing with uncertainties associated with the tube manufacturing, as variations in geometrical and mechanical
properties, and the variability in load parameters, as pore pressure gradient and fracture gradient. The First Order
Reliability Method (FORM) is applied to the reliability index evaluation. The case study presented shows that
the probabilistic optimization strategy can provide interesting solutions on the composition of casing strings,
saving investments but complying with the casing integrity for the relevant load scenarios.