The method for deriving p-y curves based largely on the back-analysis of instrumented piles are highly empirical. Cyclic T-bar penetrometer can directly measure the soil strength parameters. A monotonic p-y curve model using in-situ cyclic T-bar test is first established according to the numerical simulation of pile in cohesive soil. On this basis, considering the softening of the soil under lateral cyclic loading, a cyclic p-y curve model based on in-situ cyclic T-bar test results is developed, and its correctness is verified by comparing with the p-y curve data obtained from the simulation of lateral cyclic movement of pile foundation.


Offshore pile foundation is often subjected to the lateral cyclic loads such as wind, wave and current. Therefore, the lateral bearing capacity, especially the bearing capacity under lateral cyclic load, must be considered in the design of offshore single pile (Randolph et al., 2005). In order to accurately evaluate the interaction between laterally loaded piles and seabed soft clay, Matlock (1970) proposed a pile-soil loadtransfer curves (p-y curve) based on field test to calculate the lateral bearing capacity of pile foundation. American Petroleum Engineering Association (API, 2014) modified the p-y curve model, making it widely used to evaluate the lateral load displacement curve of pile foundation in soft clay. However, many scholars pointed out that the ultimate soil resistance calculated by Matlock (1970) was lower than the actual value (Stevens and Audibert, 1979; Jeanjean et al., 2009; Zakeri et al., 2015), and the initial stiffness calculated by API (2014) model tended to be infinite when the displacement was infinitesimal because the curve model adopted the form of power function, which was obviously inconsistent with the actual situation. In order to evaluate the lateral bearing capacity of pile foundation more accurately, many new p-y curve model were developed. Jeanjean et al. (2009) proposed a hyperbolic p-y curve model through centrifuge model tests and finite element analysis. The functional form was simple and enabled calculation of the initial stiffness, but the model shear modulus Gmax was difficult to determine in the field analysis of soil properties. Zhang et al. (2017) found that the corresponding p-y curve of soil can be obtained by scaling the soil stress-strain curve obtained in the laboratory triaxial compression test, but the high sensitivity of the seabed cohesive soil led to the difficulty in sampling and the low accuracy of the sample.

This content is only available via PDF.
You can access this article if you purchase or spend a download.