·¬ÇÑÉçÇø

Topic

Load Tests and Numerical Modeling of CFA and Driven Piles in Marl in Savannah, Georgia

Department of Civil Engineering

Date & location

• Tuesday, August 27, 2024

• 2:00 P.M.

• Virtual Defence

Reviewers

Supervisory Committee

• Dr. Cheng Lin, Department of Civil Engineering, ·¬ÇÑÉçÇø (Supervisor)

• Dr. Min Sun, Department of Civil Engineering, UVic (Member) 

External Examiner

• Dr. Wenbo Zheng, Department of Civil (Geotechnical) Engineering, University of Northern British Columbia 

Chair of Oral Examination

• Dr. Irina Paci, Department of Chemistry, UVic

   

Abstract

Marl along the southeast coast of the US, especially in South Carolina and Georgia, is a stiff, calcium carbonate-rich, over-consolidated fine-grained soil commonly used as a bearing layer for pile foundations. However, two significant engineering challenges arise: unreliable CPT-based estimations of pile capacity and strain-softening behavior complicating the prediction of pile capacity, deformation, and load-transfer mechanisms under axial load.

Comprehensive pile load tests conducted in Savannah, Georgia, were used to improve CPT based capacity estimations and develop a numerical model using t-z and Q-z load-transfer curves for driven and Continuous Flight Auger (CFA) piles. For CPT-based capacity estimations, the accuracy of the LCPC and Eslami and Fellenius methods was first evaluated. Pile load test results were then used to back-calculate the coefficients of these methods, which were subsequently verified with additional pile load tests. The main strength parameters of the numerical model were correlated with the improved CPT methods coefficients. After calibration, the numerical model was verified and worked well. Additionally, the pile load test data were utilized to analyze the effects of soil densification during the installation of two types of Continuous Flight Auger (CFA) piles: full displacement CFA piles and partial displacement CFA piles. Moreover, a finite element (FE) analysis using the ‘OC Clay’ model in Plaxis 3D was conducted to capture strain-softening behavior and examine load-transfer mechanisms. The FE model was calibrated and validated against pile load test data, and a parametric study was conducted to investigate how soil strength parameter, pile dimensions, and embedment depth into marl influence pile performance.

Keywords Marl; Driven pile; CFA pile; Numerical simulation; t-z method; CPT correlation; Finite element modeling