Modelling the pullout capacity of marquee ground anchors using neurofuzzy technique

Abstract

Marquees are temporary light structures that are connected to the ground by tensile anchors to resist forces imposed by wind acting on the structure. Failures of such structures are not rare and have resulted in deaths and tens of thousands of dollars of damage. Consequently, an accurate estimation of the ultimate pullout capacity of ground anchors is essential; however, current methods for estimat-ing the ultimate pullout capacity of marquee ground anchors are inaccurate. This is attributed to the fact that the models used by current methods were originally developed to estimate the axial ca-pacity of a single pile rather than small size an-chors. The aim of this paper is to develop a more accurate model for predicting the pullout capacity of marquee ground anchors based on the neuro-fuzzy technique. The type of neurofuzzy networks used are the B-spline networks that are trained with the adaptive spline modelling of observation data (ASMOD) algorithm. A series of 119 in situ anchor pullout tests, conducted at six different lo-cations in the Adelaide Region, South Australia, are used for the neurofuzzy model calibration and validation. Statistical analyses that compare the ac-tual measured pullout capacity with those pre-dicted by the neurofuzzy model and three existing pile capacity prediction methods are conducted and discussed.