document

Abstract

Existing single-string analysis methods are inadequate for solution of heat-up problems, which require a global analysis of the whole well system (all casings together). This paper presents a method for such a technique, termed global service life analysis, and describes an FE program (ADHOC) developed to implement it. The formulation is fully general, and is applicable to a wide range of casing design problems.

Introduction

An earlier paper co-written with BP described a recent study into annulus heat-up stresses in subsea production wells. A new and rigorous theoretical model was developed, and used for the computer analysis of several case studies. The results showed that:

- annulus heat-up causes high burst/collapse stresses; - conventional design methods were inadequate for predicting these stresses.

However, limited space prevented any attempt to explain the analysis method. This paper presents the theoretical basis for the earlier work, and explains how this theory is applied in practice.

This introductory section reviews existing design techniques, and outlines the objectives laid down for development of the new analysis model.

Review of existing design methods

The existing design methods are:

- worst case checking - single string service life analysis

and these are discussed briefly below.

Worst case checking. This is usually a hand check of some given worst case condition - for example, casing burst during drill stem test (DST). Buckling and thermal effects are not usually considered, and the stresses calculated generally do not include those already existing before the chosen load condition.

Single-string service life analysis. This does consider all the loads so far applied, but only for one given string in simple cases (eg, tubing stress-check for production loads). It can include for helical buckling and axial thermal effects, but is only applicable for subsea wellheads, because it assumes independence between strings. It therefore cannot model annulus fluid heat-up or platform well behavior.

It was clear that a much more general analysis method was required, and the development objectives were defined as follows.

Objectives

1. Develop a rigorous method for service life analysis of the whole well system (global service life analysis), including all possible loadings and a full treatment of the interaction behavior.

2. The method should permit the analysis of completely general well models, with no limits on well geometry, use of combination strings, section changes, etc.

3. Write and validate a program to implement the method.

P. 529