On The Probabilistic Assessment of Top-Hole Casing Design

Abstract

The casing system plays a crucial role in the integrity of oil and gas wells throughout their life cycle, providing tightness, stability, and support to external loads. This paper applies reliability-based models to the design of top-hole casing sections, taking into account uncertainties associated with soil behavior and casing tubulars manufacturing. Typical load scenarios are addressed to estimate the probability of occurrence of different soil-casing system failure modes. Reliability-based techniques stand out as powerful solutions for structural analysis and design. This work assesses soil characterization data from piezocone tests (CPTu) to statistically describe some mechanical parameters used for conductor and surface casing design. Random variables associated with the material and geometrical properties of tubulars are also considered, based on tubular manufacturing data presented in API/TR 5C3 (2018). The probabilistic models are developed by using the First Order Reliability Method (FORM), an expedited and accurate optimization-based procedure, and applied to various load scenarios to estimate failure probability in the context of top-hole casing design. Finite Element (FE) modeling is employed for the integrity analysis of the soil-casing system. Analyses have been carried out considering the variability associated with undrained soil strength evaluated from CPTu data, as this soil strength is expected to be the most relevant random variable due to its spatial heterogeneity. Other random variables taken into account are the outer diameter and wall thickness of casing tubulars, resulting from the variability in manufacturing process. Results indicate the feasibility and relevance of the proposed FE-FORM analysis in estimating the probability of occurrence of relevant failure modes defined following the oil company’s internal regulations regarding conductor casing load capacity and surface casing triaxial stress in the non-cemented region. For the specific case studies presented, failure probabilities ranged from the order of magnitude of 10-9 to inadmissible values approaching 50%. Concerning how random variables affect the probabilistic response, it is observed that the outer diameter is not significant due to its low dispersion. The novelty consists of considering both in-situ soil information and casing manufacturing data in a reliability-based framework that enables a more robust structural integrity analysis, supporting the decision-making process in top-hole design. This solution was implemented in the operator’s internal software and uses real data. Quantifying the soil-and-casing uncertainties by using a robust statistical-based methodology brings new information, enhancing knowledge about the variability of design parameters and its influence on the structural response.