GMC(1,N) Power Model Based on Cluster Analysis and Its Application in Predicting Ultimate Bearing Capacity of Single Pile

In view of nonlinear characteristics between ultimate bearing capacity of single pile and its influencing factors, in this paper, GMC(1,N) power model is established by introducing nonlinear parameters into GMC(1,N) model, and the parameters are solved by particle swarm optimization. To improve the model precision further, GMC(1,N) power model based on cluster analysis and the corresponding algorithm are put forward. Finally the model is used to model and predict ultimate bearing capacity of single pile, the result indicates that the prediction accuracy of GMC(1,N) power model based on cluster analysis is better than that of neutral network and GM(1,N) model. Therefore, the new model is efficient and effective.