Abstract Flowlines and pipelines installed in deep-sea waters are submitted to axial and lateral loads due to the effects of flow stoppages and starts, thermal influences and internal pressure. To study the phenomenon of soilpipeline interaction, a physical model was used and special emphasis was given to the application of large horizontal loads coupled with vertical loads. This paper focuses on the experimental results of two sideswipetype tests with a pipeline model in a very soft soil with undrained shear strength of ~3kPa. Experimental yield envelopes are also included. The experimental tests revealed that for very shallow pipe embedments the maximum horizontal load is obtained for a value of V/Vmax = 0.5, and that for larger embedments this value is in the order of 0.2. 1. Introduction The design of pipelines installed in deep-sea waters is still a challenge for offshore geotechnical engineering. Flowlines and pipelines can be submitted to combined vertical and horizontal loads, thermal expansion and internal pressure, among other loadings. The pipeline laying installation is not a guarantee of their penetration embedment and, consequently, their stability. The problem of untrenched pipelines has been studied and reported in literature (Murf et al., 1989; Brennodden and Stokkeland, 1992; Cassidy, 2004; Fontaine et al., 2004; Cathie et al., 2005; Zhang and Erbrich, 2005; Cheuk and Bolton, 2006; Dendani and Jaeck, 2007; Bruton et al., 2008; Tian and Cassidy, 2008; among others). A physical model was used to understand the mechanisms of interaction between pipe and soil through a lateral visualisation, and then to simulate the pipe response under combined vertical and horizontal loads. This paper concentrates on the results obtained using sideswipe tests with large and short horizontal displacements, in order to obtain an experimental yielding envelope of the pipe.

Abstract A study of the break-out factor ( Nc ) for plate anchors is presented in this paper. Comparisons among finite element method (FEM) analysis and laboratory results were performed. A soil corresponding to a soft normally consolidated clay was considered. Numerical FEM analyses were performed with the Plaxis® code, using an elasto-plastic model with a Mohr-Coulomb criterion. Undrained soil parameters and an adhesion factor α = 1 were used. Values of Nc factor for axisymmetric and 2D FEM were obtained. Two geometries considering a perpendicular load applied in the anchor area and horizontal anchor position to different depths were studied. In the same way, anchor plates to a reduced scale were tested in a tank containing a soft soil. It was verified that the factors Nc reach constant values beyond a determined depth of the soil. Finally, the experimental and numerical values were compared with suggested methods reported in the literature. 1. Introduction Plate anchors are frequently used as foundation solutions for offshore structures to transmit forces to surrounding soils at various depths. The capacity estimation of the anchor is nevertheless uncertain insofar as there are factors that have a large influence in its behaviour. Examples of such factors include the installation process, soil characteristics, geometry of plates and large numbers of methods used to calculate the holding capacity of the anchors. In this context, this paper presents a study of Nc values, obtained from axisymmetric and plane strain finite element model (FEM) simulations done in Plaxis® code. In addition, it outlines the results of scaled-down plate anchors tested in the laboratory. In both cases, the soil considered was soft clay with deepwater sediment characteristics. The description of specific FEM analyses and laboratory test characteristics taken into account are described. Comparisons with expressions found in literature were made.