Life Cycle Well Integrity and Reliability

Abstract

This thesis presents a method for reliability based casing design. The method is inspired by the model described by Das et al. (2015) - A model for Well Reliability Analysis throughout the Life of a Well Using Barrier Engineering and Performance (1). The purpose of this model is to show how reliability based casing design, based on statistical data, can be performed in practice. Monte Carlo simulation conducted in MATLAB is the basis of the approach. Statistical data of load and strength simulate a casing burst scenario for a life cycle period of 20 years. Degradation factors associated with casing wear and corrosion are accounted for. Based on the result from this simulation, the underlying life time distribution was identified using Nelson-estimator. Maximum Likelihood estimation was used to calculate parameters for the identified life time distribution. The results from the simulated data showed that the underlying lifetime distribution could be represented by a Weibull distribution. From this distribution, failure rate, failure function and survival function was found. The presented approach show how it is possible to quantify the reliability of a given design. Some of the identified challenges in using a reliability based design approach in practice, is related to the lack of qualitative statistical input data of load, and proper estimation of degradation factors related to the casing strength. To get an overview of the most important aspect of well design an introduction to the regulations, well integrity and load cases are given. The different design approaches used in well design is also explained and compared to get a full overview of how reliability based design differs from other design approaches.