Back-propagation neural network modeling on the load–settlement response of single piles
Zhang Wengang,Anthony Teck Chee Goh,Zhang Runhong,Li Yongqin,Wei Ning
DOI: https://doi.org/10.1016/b978-0-12-816514-0.00019-9
2020-01-01
Abstract:Abstract Pile foundations are commonly adopted to transfer the loads from the superstructure through weak strata onto the stiffer soils or rocks. Therefore, the reliability of pile-supported structures depends largely on the behavior of the piles. An important design consideration is to check the load–settlement characteristics of piles. Owing to its complexity, pile settlement lacks a precise analytical theory or understanding of the phenomena involved. In situations where measured or numerical data are available, various soft computing methods have shown to offer great promise for mapping the nonlinear interactions between the system's inputs and outputs. In this study, the back-propagation neural network (BPNN) was used to assess pile settlement subjected to the axial load, under influencing factors including pile geometry, pile material properties, soil properties and applied load, as well as additional factors including the pile installation methods, the type of load test, and whether the pile tip is closed or open. A database of approximately 500 data sets of piles, obtained from the published literature, is used for model development. The predictive accuracy, model interpretability, and parametric sensitivity analysis of the developed BPNN pile settlement model are demonstrated. The proposed BPNN model gives reasonable predictions for pile behavior and will be useful for pile design.