Combined with the stress characteristics of the bridge deep-water pile foundation during the operation period, this paper studied the general principles of the mixed model for pile axial force in deep-water pile foundation. The mixed model was firstly presented during the operation period based on the monitoring data of pile group foundation considering the influence of multiple environment factors. The mixed model of Sutong bridge pile group foundation was built by using the finite element simulation and PSO-SVM statistical method. In order to facilitate comparison, the Radial Basis Function (RBF) artificial neural network model was built, whose results were compared with the prediction results of the mixed model. The results showed that the mixed model had higher prediction accuracy and more robust predictive ability for the three piles under different loading conditions, and it exhibited better generalization ability. The mixed model could be applied to the prediction of the axial force in the deep-water pile group foundation during the operation period.