An Energy-Based Pipe-Soil Interaction Model

Harald Brennodden, Jon Tore Lieng, T. Sotberg, Richard Verley

Offshore Technology Conference · 1989 · 52 citations · 0 references

Concepts

Abstract

ABSTRACT This paper describes a research project into pipe-soil interaction conducted by SINTEF on behalf of the American Gas Association (A.G.A). An empirical model has been developed which predicts soil resistance to lateral motion of untrenched pipelines. Full scale pipe-soil interaction tests were conducted for three soils providing data for the empirical relationships in the model. The tests confirm previously reported results; that soil resistance is strongly dependent on pipe embedment and soil condition (shear strength of clay and relative density of sand). Pipe embedment is 'estimated empirically from the energy dissipated in the soil and soil resistance is split into two components; a sliding component and a penetration dependent component. In this manner the soil model accounts for the effects of cyclic pipe movements on soil resistance. The new soil model represents a significant improvement compared to the Coulomb friction estimation traditionally used in pipeline stability calculations. Especially on soft clay and loose sand and in general at high lift forces, the predicted soil resistance is larger than that predicted using friction coefficients typical in design. INTRODUCTION The A.G.A.'s motivation for this study was to complete a handbook with practical design procedures for submarine pipeline stability design. The pipe-soil interaction project focuses on the description of the interaction between an untrenched pipeline and the seabed where the pipeline is exposed to environmental loading. To meet this goal, full-scale pipe-soil interaction tests were conducted to collect sufficient information from various loading histories for pipes resting on three different soils:Loose medium/coarse sandDense medium/coarse sandSoft clay. The understanding and description of soil resistance for an on-bottom pipeline has benefited considerably from work carried out by Lyons [1], Lambrakos [2], Karal [3], Brennodden et al. [4], Wagner et al. [5], Morris et al. [6] and Palmer et al. [7]. These pipe-soil interaction studies have shown that a pipeline moving cyclically accumulates embedment, thereby leading to an increased lateral soil resistance. Experience from an earlier project, the PIPESTAB project, both in running full-scale experimental tests, [4], and describing the soil resistance by two components [5], has been beneficial to the present study. The test facilities and the test and soil handling procedures from PIPESTAB have been applied with minor modification. Furthermore the same procedure has been used to analyze test data in order to capture the penetration dependent soil resistance; by separating the total soil resistance into two terms: (Equation available in full paper) FF is equal to a sliding resistance coefficient multiplied by the net vertical pipe load (accounts for the lift force), whereas FR varies with the pipe embedment and the amount of soil piled up on each side of the pipeline, but is independent of the lift force.