Interpreting strength parameters in soft clays from a new free-fall penetrometer

Abstract

Free-fall penetrometers (FFP) are time-efficient and cost-effective tools for estimation of seabed strength parameters. However, the accuracy of strength parameters are coupled affected by the strain-rate effect, the strain-softening effect and the hydrodynamic drag on the FFP. A modified free-fall ball penetrometer with a booster (FFBPB) was put forward in the present study for the convenience of measuring the strength of soils with deep embedment depth. Numerical analyses based on the computational fluid dynamics (CFD) approach were conducted in the present study to investigate the complex ball-soil interaction, from which the expressions of the strain-rate factor, drag coefficient, and added mass were established. Then a back-analyses framework was developed to estimate the strength parameters, including the original undrained shear strength, strain-rate parameter, and strain-softening parameters, based on the measurements from the FFBPB. Finally, a series of model tests were performed to validate the back-analyses framework. The back-analysed strength parameters are considerably consistent with that from quasi-static tests, indicating the reliability of the back-analyses framework.

Introduction

Estimation of seabed strength plays an important role in geotechnical design, which is usually, however, hampered by the very high cost in respect of planning offshore site investigations. Free-fall penetrometers (FFP) are allowed to fall freely in the water column before impacting the seabed, during which the strength profile can be estimated. The FFP provides a cost-effect and time-efficient alternative for estimations of seabed strengths. Two types of FFPs are recently developed, the free-fall cone penetrometer (FFCP) with slender profile and the free-fall ball penetrometer (FFBP) with sphere geometry (Dayal et al., 1975, Chow et al., 2014, Morton and O′Loughlin, 2016a, Zhang and Liu, 2018). The interpretation of seabed strength is impeded by coupling effects resulted from very high strain rates, strain-softening behaviours and hydrodynamic drag during the dynamic penetration of the FFP.

Laboratory and field tests were conducted to rectify the effectiveness and reliability of the FFP (Dayal et al., 1975, Chow et al., 2014, Chow and O′Loughlin, 2017, Steiner et al., 2014, Stark et al., 2016, Morton and O′Loughlin, 2016a, Morton and O′Loughlin, 2016b). In experimental tests, the hydrodynamic drag acting on the FFP is usually estimated by using a constant drag coefficient, and the strain-rate effect is usually estimated by comparing the undrained shear strength determined from dynamic penetration tests with that from quasi-static penetration tests. Actually, the drag coefficient depends on the Reynolds number within the fluid mechanics framework, and varies with the non-Newtonian Reynolds number in soft clays (Zakeri, 2009, Randolph and White, 2012, Liu et al., 2015). In addition, the strain-rate effect of the soil surrounding the FFP experiences some difference with that of a soil element deduced by Einav and Randolph (2006). The soil elements located at different regions around the FFP are undergone different shearing modes (Randolph and Andersen, 2006, Zhang and Liu, 2018), hence the strain-rate behaviour of the soil estimated based on the measurements of the FFP is a macro performance differing from that of the soil element. Through experimental investigations, it is very difficult to de-couple the effects of hydrodynamic drag and strain-rate property, both of which enhance the seabed strength. Therefore, it is necessary to clarify the uncertainty which affects the accuracy of the estimated seabed strength and develop formulas to calculate each force on the FFP by conducting numerical simulations. Moreover, a more sophisticated back-analyses framework should be put forward to estimate the strength parameters including the undrained shear strength, strain-rate parameter and strain-softening parameters.

Compared with the FFCP, the FFBP is attractive due to its spherical structure and relatively large projected area. First, attributed to the spherical structure of the FFBP, there exists an exact solution for the undrained shear strength as a function of the end bearing resistance. Second, the FFBP usually has a relatively large projected area, meaning high soil resistances are mobilised in soft clay and high resolution of the measured undrained shear strength profiles. However, the final penetration depth of the FFBP is usually confined with several diameters (Morton and O′Loughlin, 2016a, Morton and O′Loughlin, 2016b), hence the undrained shear strength profiles of the soil with deep embedment cannot be estimated.

In the present study, a modified FFBP is introduced with the aim of widening the application of the current FFBP. Then relatively sophisticated formulas are established to calculate the forces acting on the FFBP based on numerical simulations. In the following, a back-analyses framework is proposed to estimate the strength parameters (including the original undrained shear strength, strain-rate parameter, and strain-softening parameters). At the end of the paper, the effectiveness of the modified FFBP and the reliability and accuracy of the back-analyses framework are verified by a series of model tests.