Multivariate adaptive regression splines and neural network models for prediction of pile drivability
Piles are long, slender structural elements used to transfer the loads from the superstructure through weak strata onto stiffer soils or rocks. For driven piles, the impact of the piling hammer induces compression and tension stresses in the piles. Hence, an important design consideration is to chec...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2016
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/82347 http://hdl.handle.net/10220/39982 |
| _version_ | 1811678411446288384 |
|---|---|
| author | Zhang, Wengang Goh, Anthony Tech Chee |
| author2 | School of Civil and Environmental Engineering |
| author_facet | School of Civil and Environmental Engineering Zhang, Wengang Goh, Anthony Tech Chee |
| author_sort | Zhang, Wengang |
| collection | NTU |
| description | Piles are long, slender structural elements used to transfer the loads from the superstructure through weak strata onto stiffer soils or rocks. For driven piles, the impact of the piling hammer induces compression and tension stresses in the piles. Hence, an important design consideration is to check that the strength of the pile is sufficient to resist the stresses caused by the impact of the pile hammer. Due to its complexity, pile drivability lacks a precise analytical solution with regard to the phenomena involved. In situations where measured data or numerical hypothetical results are available, neural networks stand out in mapping the nonlinear interactions and relationships between the system's predictors and dependent responses. In addition, unlike most computational tools, no mathematical relationship assumption between the dependent and independent variables has to be made. Nevertheless, neural networks have been criticized for their long trial-and-error training process since the optimal configuration is not known a priori. This paper investigates the use of a fairly simple nonparametric regression algorithm known as multivariate adaptive regression splines (MARS), as an alternative to neural networks, to approximate the relationship between the inputs and dependent response, and to mathematically interpret the relationship between the various parameters. In this paper, the Back propagation neural network (BPNN) and MARS models are developed for assessing pile drivability in relation to the prediction of the Maximum compressive stresses (MCS), Maximum tensile stresses (MTS), and Blow per foot (BPF). A database of more than four thousand piles is utilized for model development and comparative performance between BPNN and MARS predictions. |
| first_indexed | 2024-10-01T02:52:50Z |
| format | Journal Article |
| id | ntu-10356/82347 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T02:52:50Z |
| publishDate | 2016 |
| record_format | dspace |
| spelling | ntu-10356/823472020-03-07T11:43:40Z Multivariate adaptive regression splines and neural network models for prediction of pile drivability Zhang, Wengang Goh, Anthony Tech Chee School of Civil and Environmental Engineering Multivariate adaptive regression splines Pile drivability Computational efficiency Back propagation neural network Nonlinearity Piles are long, slender structural elements used to transfer the loads from the superstructure through weak strata onto stiffer soils or rocks. For driven piles, the impact of the piling hammer induces compression and tension stresses in the piles. Hence, an important design consideration is to check that the strength of the pile is sufficient to resist the stresses caused by the impact of the pile hammer. Due to its complexity, pile drivability lacks a precise analytical solution with regard to the phenomena involved. In situations where measured data or numerical hypothetical results are available, neural networks stand out in mapping the nonlinear interactions and relationships between the system's predictors and dependent responses. In addition, unlike most computational tools, no mathematical relationship assumption between the dependent and independent variables has to be made. Nevertheless, neural networks have been criticized for their long trial-and-error training process since the optimal configuration is not known a priori. This paper investigates the use of a fairly simple nonparametric regression algorithm known as multivariate adaptive regression splines (MARS), as an alternative to neural networks, to approximate the relationship between the inputs and dependent response, and to mathematically interpret the relationship between the various parameters. In this paper, the Back propagation neural network (BPNN) and MARS models are developed for assessing pile drivability in relation to the prediction of the Maximum compressive stresses (MCS), Maximum tensile stresses (MTS), and Blow per foot (BPF). A database of more than four thousand piles is utilized for model development and comparative performance between BPNN and MARS predictions. Published version 2016-02-16T06:33:40Z 2019-12-06T14:53:48Z 2016-02-16T06:33:40Z 2019-12-06T14:53:48Z 2014 Journal Article Zhang, W., & Goh, A. T. C. (2014). Multivariate adaptive regression splines and neural network models for prediction of pile drivability. Geoscience Frontiers, 7, 45-52. 1674-9871 https://hdl.handle.net/10356/82347 http://hdl.handle.net/10220/39982 10.1016/j.gsf.2014.10.003 en Geoscience Frontiers ©2014, China University of Geosciences (Beijing) and Peking University. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/). 8 p. application/pdf |
| spellingShingle | Multivariate adaptive regression splines Pile drivability Computational efficiency Back propagation neural network Nonlinearity Zhang, Wengang Goh, Anthony Tech Chee Multivariate adaptive regression splines and neural network models for prediction of pile drivability |
| title | Multivariate adaptive regression splines and neural network models for prediction of pile drivability |
| title_full | Multivariate adaptive regression splines and neural network models for prediction of pile drivability |
| title_fullStr | Multivariate adaptive regression splines and neural network models for prediction of pile drivability |
| title_full_unstemmed | Multivariate adaptive regression splines and neural network models for prediction of pile drivability |
| title_short | Multivariate adaptive regression splines and neural network models for prediction of pile drivability |
| title_sort | multivariate adaptive regression splines and neural network models for prediction of pile drivability |
| topic | Multivariate adaptive regression splines Pile drivability Computational efficiency Back propagation neural network Nonlinearity |
| url | https://hdl.handle.net/10356/82347 http://hdl.handle.net/10220/39982 |
| work_keys_str_mv | AT zhangwengang multivariateadaptiveregressionsplinesandneuralnetworkmodelsforpredictionofpiledrivability AT gohanthonytechchee multivariateadaptiveregressionsplinesandneuralnetworkmodelsforpredictionofpiledrivability |