Estimation of Uplift Capacity of Helical Pile Resting in Cohesionless Soil

In this research article, the finite element method was used to investigate the effect of embedment depth (H), friction angle (phi), the ratio of the diameter of the helical blade to the diameter of the shaft (D-p/D-s), number of helices (n), and inter-helix spacing ratio (S/D-p), on the uplift capacity of the helical pile (Q(u)). The helical pile was modeled and validated with the experimental results obtained from past investigations. The numerical data was then processed through an artificial neural network to develop a prediction model between the uplift capacity and varied parameters. An increment in Q(u) was observed with the increases in phi, n, H, and S/D-p. However, a decrease in Q(u) for single and double-helical piles was noted with an increase in D-p/D-s ratio. The outcome of the artificial neural network revealed that the best prediction of Q(u) was given by the SIGMOID activation function. Additionally, the sensitivity analysis showed that Q(u) was strongly influenced by S/D-p and least influenced by Dp/Ds ratio.