Uplift Capacity of Suction Caisson in Clay Using Artificial Intelligence Techniques

Abstract

Caisson anchor is the most common structure for the offshore structures. In this study artificial intelligence techniques artificial neural network (ANN), genetic programming (GP), support vector machine (SVM), and relevance vector machine (RVM) have been used to predict the uplift capacity (Q) of suction caisson in clay. The model inputs included the L/d (L is the embedded length of the caisson and d is the diameter of caisson), undrained shear strength of soil at the depth of the caisson tip (Su), D/L (D is the depth of the load application point from the soil surface), inclined angle(θ) and load rate parameter (Tk). A comparative analysis is made with the results of the above listed artificial intelligence techniques based on different statistical performance criteria like correlation coefficient (R), root mean square error (RMSE), Nash-Sutcliff coefficient of efficiency (E), log normal distribution of ratio of predicted to observed load capacity. Model equations based on the above techniques are discussed and a sensitivity analysis is made to identify the important parameters contributing to the uplift load capacity of caisson.