Centrifuge modelling of mono-pile under cyclic lateral loads

Abstract

Mono-pile foundations are widely used for wind turbines at present. They are always subjected to significant cyclic lateral loads due to wind and wave actions. In order to better understand the performance of mono-piles under cyclic lateral loads, a series of cyclic lateral load tests was conducted on a stainless steel mono-pile in the centrifuge. The axial and lateral responses of a large diameter mono-pile under one-way force-controlled cyclic lateral loads are described. Accumulated permanent pile shaft lateral displacements caused by cyclic lateral loads are discussed, and a function is utilized to estimate these permanent displacements. Also the influence of cyclic lateral loads on the pile lateral secant stiffness is investigated. These results offer an insight to further research to optimise designs of mono-pile foundations to resist live loads in service. (1994) improved the p-y approach to consider the effect of the number of cycles. Nevertheless, only 50 or less cycles of lateral loads are executed in most of the tests considered. Moreover, the use of p-y curves often fails to account for the permanent deformation that accumulates with increasing cycles (Moss et al. 1998). Therefore, additional cyclic load tests need to be conducted with many more cycles than previously to better understand the influence of cycle numbers. Not only the largest pile deflection in each cycle, but also the accumulated permanent displacement and secant stiffness of the pile should be investigated. In field tests, it is difficult to exert cyclic loading on mono-piles with large diameters due to the limitations of test facilities and high costs. However, centrifuge modelling offers an effective way to understand the influence of cyclic loads on piled foundations. Compared with field tests, centrifuge tests are more convenient, efficient and cheaper. In this research, a series of cyclic lateral load tests were conducted in the centrifuge. The axial and lateral response of a large diameter mono-pile under one-way force-controlled cyclic lateral loads is described. The accumulated permanent pile shaft lateral displacements caused by cyclic lateral loads are discussed, and a function is utilized to estimate these permanent lateral displacements. Also, the influence of cyclic lateral loads on the pile lateral secant stiffness is investigated. 2 EXPERIMENTAL METHODOLOGY 2.1 Acceleration and scaling in centrifuge tests A centrifuge test on a 1/100 scale mono-pile is proposed to be conducted in saturated sand at 100g to model the behaviour of mono-pile foundations for offshore wind turbines. In saturated sand, the effective vertical stress of soil is expressed in the following equation 1: z Gsat v ρ σ ′ = ′ (1) where Gsat is the acceleration in a centrifuge test in saturated sand, i.e. Gsat = 100 g; ρ’ = the buoyant density of saturated soil (1137 kg/m); and z = the depth of soil in a 1/100 scale model. For the purpose of an initial study, the fully drained cyclic response was required. Since excess pore pressures were to be avoided, it was realised that dry sand could conveniently be used. In order to achieve the equivalent soil conditions, the vertical stress of dry sand expressed in equation 2 should be equal to the effective vertical stress of saturated sand in equation 1: z Gd d v ρ σ = (2) where Gd is the acceleration to be imposed in a centrifuge test on dry sand; ρd = the dry density of the sand (1826.5 kg/m), and z = the depth of soil in a 1/100 scale model. Thus the d G value is 62.3 g, obtained by equation 3: