PREDICTION OF UPLIFT CAPACITY OF SUCTION CAISSON IN CLAY USING FUNCTIONAL NETWORK AND MULTIVARIATE ADAPTIVE REGRESSION SPLINE

Abstract

Suction caissons are extensively used as anchors for offshore foundation structures. The uplift capacity of suction caisson is an important factor from effective design point of view. In this paper, two recently developed AI techniques, functional network (FN) and multivariate adaptive regression spline (MARS), have been used to predict the uplift capacity of suction caisson in clay. The performances of the developed models are compared with other AI techniques; artificial neural network, support vector machine, relevance vector machine, genetic programming, extreme learning machine and group method of data handling with harmony search (GMDH-HS). The model inputs include the aspect ratio of the caisson, undrained shear strength of soil at the depth of the caisson tip, relative depth of the lug at which the caisson force is applied, load inclination angle and load rate parameter. Comparative analyses are made with the results of the above AI techniques, using different statistical performances criteria; correlation coefficient (R), root mean square error, Nash-Sutcliffe coefficient of efficiency, log-normal distribution of ratio of predicted to observed load capacity, with a ranking system to find out the best predictive model. The FN and MARS models are found to be comparably efficient and they outperform other AI techniques.