Effect of Large Deformation Analysis for Site Investigation Tool - CPT in Layered Soils

Qiang Xie,Yuxia Hu,Mark J. Cassidy
DOI: https://doi.org/10.1115/omae2020-19099
2020-08-03
Abstract:Abstract Cone penetration test (CPT) is regularly used during offshore site investigations to interpret soil stratification and soil characteristics due to its continuous penetration resistance profile. However, its use could be improved if better numerical methods to simulate its penetration could be developed. Finite element (FE) analysis, for instance, has the potential to provide insightful information on soil response and soil flow mechanisms. However, it is challenging to simulate CPT in layered soils, as the soil experiences extremely large strains around the cone and the simulation costs are high. In this study, the efficiency of using a partial large deformation FE (LDFE) approach was explored to examine the pre-embedment depth allowed for saving LDFE analysis cost. The LDFE analysis was conducted using the remeshing and interpolation technical with small strain (RITSS) method to model the large strain problem. Both soft-stiff-soft clays and clay-sand-clay soil were considered to study the thin stiff layer effect when it was sandwiched in soft clay. The LDFE/RITSS analysis compared a CPT penetrating from the soil surface with penetrations from a pre-embedded depth above the stiff layer. Pre-embedded small strain analysis was also conducted for comparison. The results show that the small strain analysis underestimated the resistance in both clay and sand. For the partial LDFE analysis with pre-embedment in the top clay layer, the CPT response in the middle stiff clay layer could be well captured regardless of the initial pre-embedment depth. However, for the middle medium dense sand layer (ID = 60%), the pre-embedment depth needs to have sufficient distance above it (10D, D is cone diameter) to capture the soil response in the sand layer correctly.
What problem does this paper attempt to address?