A new modified approach to evaluating the installation power of large-diameter helical piles in sand validated by centrifuge and field data

Helical pile (screw pile) is becoming increasingly popular in emerging pile foundation techniques for offshore structures. Its installation parameters (i.e., installation force, torque, and power) are significantly important for design and construction of helical pile foundation. However, there are very limited studies on the installation power of large-diameter multi-helix piles in sand particularly with the consideration of the installation effects of rotary press-in. This paper proposes a new modified method suitable for predicting the installation torque and force of single-helix and multi-helix piles, which involves four aspects of important factors: the installation effect of helical piles, the bidirectional force (i.e., vertical and tangential shear stress) on the side of the central shaft during rotary press-in, the interaction between helices for muti-helix piles, and the vertical installation force (crowd force) effect on torque generation. The proposed and the two existing methods (i.e., Ghaly and Hanna model and Al-Baghdadi model) are used to predict the installation toque and power of helical piles in sand with different relative densities (loose, medium dense, and dense) based on pile geometry characteristics. Predictions from aforementioned three methods are compared with measurements from the centrifugal model tests of various screw piles in sand with the proposed method giving better agreement with the measured data. The calculation from the proposed method also gives closer agreement with field measurements of the double-helix pile than those from existing methods. In addition, it is found that the installation power of helical piles increases nonlinearly with the increase of the relative density of sand (or peak friction angle) and installation force and torque account for about 10% and 90%, respectively, of the total power through the parameter sensitivity analysis.