Predicting the nonlinear load response of caisson foundations is critical to the foundation design. Despite extensive studies aimed at developing models for predicting the combined V-H-M bearing capacity of suction caissons in clay, accurately predicting the three-dimensional (3D) deflection response of the foundation remains a significant challenge. In this paper, we present a novel solution by developing a fully connected (FC) neural network model that enables load-deflection prediction of suction caissons on clay. To train and evaluate the FC model, a series of 3D finite element simulations were performed covering caissons responses with an embedment ratio of up to 1. The effect of various model hyperparameters on the model's prediction accuracy and generalisation ability was systematically investigated. The results show that the proposed model achieves load-deflection response prediction with simplicity, efficiency and accuracy, demonstrating the significant potential of deep learning technology in the geotechnical design of foundations.

Predicting the non-linear loading response is the key to the design of suction caissons. This paper presents a systematic study to explore the applicability of deep learning techniques in foundation design. Firstly, a series of three-dimensional finite element simulations was performed, covering a wide range of embedment ratios and different loading directions, to provide training data for the deep neural network (DNN) model. Then, hyper-parameter tuning was performed and it is found that the basic Fully-Connected (FC) neural network model is sufficient to capture the non-linear response of suction caissons with excellent accuracy and robustness. Furthermore, the optimized FC neural network model was also successfully applied to a database of suction caissons in sand, demonstrating its broad applicability. By comparing three typical DNNs, i.e., FC, Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM), it was observed that the FC neural network model excels over others in terms of simplicity, efficiency and accuracy. More importantly, by looking into the model's generalization performance, the FC neural network model can also identify the change in foundation failure mechanisms. This study demonstrates the DNN's powerful mapping ability and its potential for future use in offshore foundation design.