Finite Element Analysis Of Single Pile Under Static Loads And Prediction Of Bearing Capacity Using ANFIS

The prediction of vertical ultimate bearing capacity of single pile is an important problem in civil engineering. In this paper, the vertical ultimate bearing capacity of single pile has been studied using ANSYS software system. Based on a certain engineering example in Ningbo, the static simulation of single pile has been presented using numerical method. The inverse analysis is showed for some results analyzed by FEM(Finite Element Method) with GA-ANN(Genetic Algorithm-Artifical Neural Network) method.In order to improve of the accuracy of prediction, ANFIS (Adaptive Network Based Fuzzy Inference System) is introduced into the prediction of vertical ultimate bearing capacity of single pile. Because of its strong learning ability and fuzzy logic function, an ANFIS model for predicting vertical ultimate bearing capacity of single pile is established in this paper. By training and predicting from static tests, it is demonstrated that this model has high precision and strong application.Basic theory and classical algorithm of reliability analysis is presented, and the accuracy of classical algorithm is analyzed. In order to solve the problem of method choice of structural reliability theory in engineering application, the combined method of optimization and equivalent normalization is introduced to caculated the structural reliability index. Consequently, a research on reliability theory to predict the vertical ultimate bearing capacity of single pile using ANFIS is presented. The reliability analysis of bearing capacity of single pile is made by utilizing the limit state equation of dimensionless random variables, considering the statistics data of static test in Ningbo. The index of reliability of the single pile is calculated with optimization method. Finally, a reliability design method of single pile is given.