A CPSO-SVM Model for Ultimate Bearing Capacity Determination

In this study, the CPSO-SVM models, which combine chaotic system, particle swarm optimization (PSO) and support vector machine (SVM), are presented and applied to predict the ultimate bearing capacity of shallow foundations. Chaotic mapping enjoys certainty, ergodicity and the stochastic property. Chaotic PSO (CPSO) increases the convergence rate of PSO and precision of the results through introducing chaos mapping into the particle swarm optimization algorithm. Since the selection of parameters for SVM is crucial to its performance of prediction, the CPSO is adopted to search for the optimal parameters. The proposed methods are used to predict the ultimate bearing capacity of shallow foundations based on data of load tests. Results indicate that the proposed methods can appropriately describe the relationship between ultimate bearing capacity and its affective factors, and make good predictions.