Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/82347
Title: Multivariate adaptive regression splines and neural network models for prediction of pile drivability
Authors: Zhang, Wengang
Goh, Anthony Tech Chee
Keywords: Multivariate adaptive regression splines
Pile drivability
Computational efficiency
Back propagation neural network
Nonlinearity
Issue Date: 2014
Source: 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.
Series/Report no.: Geoscience Frontiers
Abstract: 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.
URI: https://hdl.handle.net/10356/82347
http://hdl.handle.net/10220/39982
ISSN: 1674-9871
DOI: 10.1016/j.gsf.2014.10.003
Schools: School of Civil and Environmental Engineering 
Rights: ©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/).
Fulltext Permission: open
Fulltext Availability: With Fulltext
Appears in Collections:CEE Journal Articles

Files in This Item:
File Description SizeFormat 
Multivariate adaptive regression splines and neural network models.pdf1.73 MBAdobe PDFThumbnail
View/Open

SCOPUSTM   
Citations 1

384
Updated on Jun 15, 2024

Web of ScienceTM
Citations 1

317
Updated on Oct 27, 2023

Page view(s)

419
Updated on Jun 20, 2024

Download(s) 20

209
Updated on Jun 20, 2024

Google ScholarTM

Check

Altmetric


Plumx

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.