Assessment of CO2 corrosion damage: Likelihood of failure in deviated wells by deterministic and stochastical modelling
Conference Paper
Overview
Identity
Additional Document Info
View All
Overview
abstract
Damage due to CO2 corrosion related processes is a cause of major concern regarding the structural reliability of subsurface installations used in oil and gas production. As to well completions, the effective control of CO2 corrosion includes the complete characterization of the operating environment since corrosion rates and CO2 corrosion morphology are closely related to variables such as the presence of multiphase flow, temperature and produced water composition. In the present paper an a priori damage assessment methodology is used to analyze casing failures observed in deviated wells with electro submergible pumps (ESP) in the Oritupano oilfield (located in eastern Venezuela). The study analyzes the assessment of CO2 corrosion damage and determines the subsequent probability of failure in relation to the structural reliability of the well. The method proposed is based on the mechanistic CO2 corrosion and multiphase flow modelling in order to deliver a comprehensive quantitative description of the expected corrosion rates within the context of the actual well operating environment and completion design. The probability of failure is quantified by assessing two independent probability events: damage distribution and the fracture mechanics associated failure probability. The corrosion related damage distribution is calculated by using stochastical simulation of the CO 2 corrosion mechanisms, linked to the deterministic corrosion study. The fracture mechanics associated probability is quantified by defining the main operating parameters (such as temperature and bottom hole pressure) as stochastically sensible independent variables within the context of a Level 2 Failure Assessment Diagram (FAD), as defined in API 579. The results obtained explain the difference observed in the reported work over rates between vertical and deviated wells in the specific oilfield (Oritupano A) using the same extraction methodology. It is also possible to predict consistently the reliability curves obtained from actual field data and discriminate among the possible causes of failure. Copyright 2007, Society of Petroleum Engineers.
