On the use of nonlocal regularisation in slope stability problems

File Description SizeFormat 
1-s2.0-S0266352X16302579-main.pdfPublished version5.85 MBAdobe PDFView/Open
Title: On the use of nonlocal regularisation in slope stability problems
Authors: Summersgill
Kontoe, S
Potts, D
Item Type: Journal Article
Abstract: The present study examines the use of nonlocal regularisation in a coupled consolidation problem of an excavated slope in a strain softening material. The considered boundary value problem allows for a thorough evaluation of the nonlocal regularisation approach, as it does not entail any kinematic restraint on the slip surface development. The examined nonlocal strain softening constitutive model requires the specification of one additional parameter, the defined length DL, which essentially modifies the rate of softening. In addition, the optional radius of influence parameter, RI, can be specified to reduce the number of local strains referenced in the nonlocal strain calculation and thus increase the efficiency of the analysis. The nonlocal strain softening constitutive model reduces significantly the mesh dependency of cut slope analyses for a range of mesh layouts and element sizes in comparison to the conventional local strain softening approach. The nonlocal analyses are not entirely mesh independent, but the predicted time to failure and horizontal displacement over time are much more consistent compared to analyses that employ the local strain softening constitutive model. Further investigation, computing the Factor of Safety of various mesh arrangements showed that for drained conditions the nonlocal regularisation eliminates the mesh dependence shown for the same analyses by the conventional local strain softening model. The impact of the two nonlocal parameters, DL and RI, on the numerical predictions is also parametrically examined. The parameter DL modifies the softening rate of the soil and therefore its selection should be based on simulating a realistic softening rate for the examined soil material. The RI parameter was found to reduce significantly the computational cost at the expense of affecting the development of the secondary slip surfaces. Overall though, the results and the critical slip surface were similar for all considered values of RI.
Issue Date: 24-Oct-2016
Date of Acceptance: 16-Oct-2016
URI: http://hdl.handle.net/10044/1/41703
DOI: https://dx.doi.org/10.1016/j.compgeo.2016.10.016
ISSN: 1873-7633
Publisher: Elsevier
Start Page: 187
End Page: 200
Journal / Book Title: Computers and Geotechnics
Volume: 82
Issue: 2
Copyright Statement: © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http:// creativecommons.org/licenses/by/4.0/).
Sponsor/Funder: Engineering & Physical Science Research Council (EPSRC)
Engineering & Physical Science Research Council (EPSRC)
Engineering & Physical Science Research Council (EPSRC)
Engineering & Physical Science Research Council (EPSRC)
Engineering & Physical Science Research Council (EPSRC)
DONG Energy Wind Power A/S
Engineering & Physical Science Research Council (E
Technology Strategy Board
Geotechnical Consulting Group
Funder's Grant Number: GR/M79462
GR/R67408/01
EP/D505488/1
EP/D506387/1
EP/G063486/1
030-13-0133
BH101595 RES0540/7539
101968
CISM_P60087
Keywords: Civil Engineering
Resources Engineering And Extractive Metallurgy
Interdisciplinary Engineering
Publication Status: Published
Open Access location: http://www.sciencedirect.com/science/article/pii/S0266352X16302579
Appears in Collections:Faculty of Engineering
Civil and Environmental Engineering



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

Creative Commons