Analysis of the cone penetration test in layered clay

Abstract

This paper presents an analysis of the cone penetration test in multi-layered clays using the commercial finiteelement code Abaqus/Explicit. The von Mises yield criterion and its associated flow rule are assumed to model the plastic behaviour of elastoplastic undrained clays. An arbitrary Lagrangian-Eulerian scheme and an enhanced hourglass algorithm are adopted to preserve the quality of mesh throughout the numerical simulation. Initially, the behaviour of the penetration resistance is examined in a soil with only two layers. The bottom layer is the weaker of the two and the behaviour of the penetration resistance when the cone approaches the lower layer is studied. The investigation is then extended to study the cone penetration test in a multi-layered clay by sandwiching a weaker clay layer between two stronger clay layers. The thickness of the weaker clay layer is varied and the behaviour of the penetration resistance is studied in relation to the thickness and relative strength of the soil layers. The results are discussed with respect to the soil mechanisms that are present when the cone moves past the relevant layer boundaries so that the position of these boundaries can be determined more accurately.KEYWORDS: clays; in situ testing; numerical modelling Cette communication présente une analyse de l'essai de pénétrabilité au cône dans des argiles multicouches, en utilisant la norme aux éléments finis commerciale Abaqus/Explicit. On a supposé que le critère de fluage de von Mises, et sa règle de fluage connexe, reflètent le comportement plastique des argiles élastoplastiques non drainées. On adopte un système Lagrangian-Eulerian arbitraire, ainsi qu'un algorithme à sablier renforcé, pour conserver la qualité de la maille tout au long de la simulation numérique. Au départ, le comportement de la résistance à la pénétration est examiné dans un sol à deux couches seulement. La couche inférieure est la plus faible des deux, et on étudie le comportement de la résistance à la pénétration lorsque le cône est proche de la couche inférieure. On poursuit ensuite l'examen pour étudier l'essai de pénétrabilité au cône dans une argile multicouche, en intercalant une couche d'argile plus faible entre deux couches d'argile plus résistante. On varie l'épaisseur de la couche d'argile plus faible, et on étudie le comportement de la résistance à la pénétration relativement à l'épaisseur et à la force relative des couches de sol. On discute des résultats relativement aux mécanismes du sol présents lorsque le cône se déplace au-delà des limites des couches concernées, de façon à déterminer avec plus de précision l'emplacement de ces limites.

Large deformation finite element analyses in geotechnical engineering

“…comm.). The EALE analysis predicts slightly higher resistance than the RITSS, whilst the resistances from the CEL agree with those in Walker and Yu [41]. The resistance-displacement curve from the RITSS analysis is remarkably smooth, while results of the explicit algorithms show some computational noise (the appearance of which depends on the sampling rate as well as parameters defining the analysis).…”

“…For cone penetration tests in soft clays, the undrained shear strength of soil, s u , can be deduced from the net penetration resistance by means of a theoretical or calibrated bearing factor N kt . The bearing factor has been investigated using various ALE approaches with implicit or explicit schemes [42,41,18,19,25]. Here, a benchmark case in Walker and Yu [42,41] was replicated: a standard cone with projected area of A = 1000 mm 2 (shaft diameter D = 35.7 mm) and apex angle of 60°was penetrated into weightless clay (though a density has to be specified for the CEL analysis as this influences the initial estimate of stiffness and hence the critical time step) under undrained conditions; the cone was assumed to be smooth; the soil strength was uniform with s u = 10 kPa; and the soil rigidity index G/s u = 100, where G is the elastic shear modulus.…”

“…The resistance obtained from RITSS is slightly higher than that of Walker and Yu [42], but moderately lower than Walker and Yu [41]. Both analyses by Walker and Yu [42,41] are based on the ALE function in Abaqus/Explicit, however, the penetration resistance was calculated using the averaged stress from the integration points below the cone face in the former and using the nodal forces at the cone face in the latter (Yu, 2014, pers.…”

Layered effects on soil displacement around a penetrometer

“…Vreugdenhil et al, 1994;Mo et al, 2017) and numerical approaches (e.g. Ahmadi and Robertson, 2005;Xu, 2007;Walker and Yu, 2010) have also been performed to investigate penetration problems in layered soils. Despite these valuable contributions, there is still a limited amount of data available on penetration induced soil deformations within layered soils.…”

Distinct element analyses of inclined cone penetration test in granular ground

“…Numbers of approaches have been used to explain the mechanism of CPT, including the bearing capacity theory [1] , the cavity expansion theory [2] , the strain path method [3] , the finite element method [4] , the distinct element method (DEM) [5] and the model test [6] . However, most of these studies are based on vertical CPT, which can be simplified as an axial symmetric boundary-value problem.…”