Please use this identifier to cite or link to this item: http://hdl.handle.net/10125/62258

Estimation of Clear-Water Local Scour at Pile Groups Using Genetic Expression Programming (GEP) and Multivariate Adaptive Regression Splines (MARS).

File Size Format  
2017-12-ms-truce.pdf 1.35 MB Adobe PDF View/Open

Item Summary

Title:Estimation of Clear-Water Local Scour at Pile Groups Using Genetic Expression Programming (GEP) and Multivariate Adaptive Regression Splines (MARS).
Authors:Truce, Bryan A.
Contributors:Civil Engineering (department)
Date Issued:Dec 2017
Publisher:University of Hawaiʻi at Mānoa
Abstract:The physical process of scour around pile groups is complex. Due to economical and geotechnical reasons, group piles have become more common in bridge designs. Various empirical models have been developed to estimate maximum scour depth at pile groups. However, these models are mostly based on the conventional statistical regression approaches, and are not able to adequately capture the highly nonlinear and complex relationship between scour depth and its influential factors. In this study, Multivariate Adaptive Regression Splines (MARS) and Genetic Expression Programming (GEP) were used to estimate clear-water local scour depth at pile groups from the current, sediment, and pile characteristics. Two combinations of data were used to train the GEP and MARS models. The first combination includes mean flow velocity, flow depth, mean grain diameter, pile diameter, distance between the piles, and the number of piles normal to the flow and in-line with the flow. The second combination contains seven non-dimensional parameters. Results indicated that MARS and GEP can generate accurate scour depth estimates. Both models yield better results when the dimensional data were used. Also, it was found that the MARS model (with root-mean-square-error (RMSE) of 0.0110 m and correlation coefficient (R2) of 0.969 outperformed the GEP model (with RMSE of 0.0187m and R2 of 0.911). The performance of GEP and MARS models was compared with that of the existing empirical methods. The comparison showed that both models perform better than the regression-based empirical equations. Finally, a sensitivity analysis showed that the pile diameter in dimensional combination and ratio of pile spacing to pile diameter in non-dimensional combination have the most significant impact on scour depth.
Description:M.S. Thesis. University of Hawaiʻi at Mānoa 2017.
URI:http://hdl.handle.net/10125/62258
Rights:All UHM dissertations and theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission from the copyright owner.
Appears in Collections: M.S. - Civil Engineering


Please email libraryada-l@lists.hawaii.edu if you need this content in ADA-compliant format.

Items in ScholarSpace are protected by copyright, with all rights reserved, unless otherwise indicated.