Abstract

Abstract Two distinct concepts have been developed in general structural engineering and are now being applied to pipelines. Limit state thinking focuses on conditions that directly threaten the security or operability of a system, and have already been adopted into several codes applicable to offshore pipelines. Reliability theory attempts to quantify the safety of a system by statistical arguments. The paper examines how far limit state methods can be exploited in marine pipeline engineering without underpinning them by reliability theory. It explores this application of reliability theory, and suggests that in this instance - as in many others - the data needed for confidence are not available (and in many instances can never be available). As in structural failures generally, a majority of pipeline failures are beyond the reach of reliability theory. Paradoxically, this fact may strengthen the case for the adoption of less conservative requirements on limit states, because the real increase in risk is often zero m extremely small. These arguments are then applied to a discussion of the allowable circumferential stress in pipelines where traditional design factors have often been copied from code to code. Introduction This paper explores the application of limit-state and reliability methods, primarily in the pipeline industry. These concepts are widely applied in other areas of civil engineering, notably in steel structure design and in concrete design, and in geotechnics. pipeline engineering has 10 some extent lagged behind in this development, at least in its more explicit forms, though it cart be argued that some of the methods customarily applied to pipelines apply limit state ideas implicitly. Limit state concepts have however influenced the Dutch code NEN3650, the Russian SNiP 2.05.06-85, the Canadian CSA Z662-94, ANSI B31 .8, the British BS 801 O: Part 3 and the 1995 Germanischer Lloyd rules. The two groups of concepts spoken of as 'limit state' and 'reliability' are separate and distinct. In a historical perspective, the two concepts have only relatively recently become intertwined and .sometimes confused, not always advantageously. Limit states Limit-state structural engineering are completely central, and far from new. A structure can become unserviceable in various ways. It can collapse, bits can off, deflections can damage internal finishes or make it impossible 10 operate, oscillations can be so large that the occupants feel uncomfortable, and so on, Each of these conditions defines a limit state in which the structure becomes unfit for purpose. Engineers have thought in these terms from the earliest times. The Roman engineers who designed aqueducts like the Pent du Gard must have asked similar questions how do 1 proportion the arches w that they do not be lapse as soon as the is removed, is there a risk of by crushing, is there a risk of progressive distress because of settlement of foundations, and so on.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call