ABSTRACT

Analysis of the upper hulls of a semi submersible drilling platform by finite element approach shows that the conventional effective breadth method for box girder design has serious limitations when applied to the modern drilling vessels where complicated structural geometry and unusual loadings are often encountered. In some cases, it is recommended to vary the thickness across the flange according to the actual stress distribution which can be found by modern computing techniques.

INTRODUCTION

Box sections are used extensively in the construction of many offshore drilling vessels such as jack-up platforms and semi submersibles. For the purpose of structural analysis, the hull of a drilling ship may also be regarded as a simple girder or metrical box. The design process of a box girder can often be simplified by introducing an artificial effective breadth, usually somewhat less than the real breadth of the box girder, and assumption of uniform direct stress across this breadth. This conventional design concept is based on the theory of shear lag.

The upper hulls of two semi submersibles designed by the Offshore Company (SCP III Mark 2 and Chris Cheery) are of box sections. A finite element analysis was performed by The Offshore Company Engineering Department to find out:

  1. Stress distribution at the diagonal to upper hull joints,

  2. Effect of eccentric intersection of the tubular diagonal to the transverse upper hull,

  3. Effect of a large drilling slot in the transverse upper hull.

Stress curves from this analysis were plotted against that from the theory of shear lag. Based on the comparison of these stress curves, the author believes that the conventional design process has limitations when applied to the modern drilling vessels because of the following reasons: (1) There have been some basic changes in the geometry of today's vessel.

Box girders intersected by some other members or with large openings are often encountered in today's drilling rigs. (2) Reduction in weight and keeping stress below a certain level to avoid fatigue failure have become more important as drilling is getting into rougher and deeper sea.

This paper points out two cases in which the effective breadth concept is not valid. This analysis also demonstrates the importance of the finite element approach in obtaining information related to a special problem in structural design.

THE EFFECTIVE BREADTH CONCEPT

If the proportions of the box girder are such that they conform to those of a conventional beam (depth greater than breadth), the flexural stress across the flange breadth will be essentially uniform and of such intensity as is calculated by elementary beam theory. On the other hand, if the flange of the same girder were made wider and wider, the flange stress distribution would become progressively less uniform because of shear r la g. Thus, ins teed 0 f beam sections remaining plane after bending, the flange shearing deformations resulting from the highly localized manner in which the horizontal shearing stresses are transmitted to the flange cause the longitudinal strips of flange more remote from the flange-web juncture to lag farther back than those nearer the web s. Hence, the term" shear lag". Fig. 1 shows the stress pattern in flange for a single cell box. The effect of shear lag has made the maximum stress max greater than the average value obtained from the elementary formu la 6'=MC I I.

This content is only available via PDF.
You can access this article if you purchase or spend a download.