Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Modelling of surfacing seal binder ageing and rheology
Thesis (PhD)--Stellenbosch University, 2021.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | en_ZA |
| Published: |
Stellenbosch : Stellenbosch University
2021
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867614042602864640 |
|---|---|
| access_status_str | Open Access |
| author | Goosen, Elaine Simone |
| author2 | Jenkins, Kim Jonathan |
| author_browse | Goosen, Elaine Simone Jenkins, Kim Jonathan |
| author_facet | Jenkins, Kim Jonathan Goosen, Elaine Simone |
| author_sort | Goosen, Elaine Simone |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Thesis (PhD)--Stellenbosch University, 2021. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/123631 |
| institution | Stellenbosch University (South Africa) |
| language | en_ZA |
| last_indexed | 2026-06-10T12:45:45.384Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2021 |
| publishDateRange | 2021 |
| publishDateSort | 2021 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/123631 Modelling of surfacing seal binder ageing and rheology Goosen, Elaine Simone Jenkins, Kim Jonathan Stellenbosch University. Faculty of Engineering. Dept. of Civil Engineering. Machine Learning UCTD Bitumen Rheology Road surfacings -- Maintenance and repair Road surfacings -- Strains and stresses Thesis (PhD)--Stellenbosch University, 2021. ENGLISH ABSTRACT: This thesis aims to develop a model that accounts for in-service changes in seal binder rheology. South Africa is in a transitional period from penetration-viscosity to performance grade specifications. Current practices and ageing analysis are still prima- rily focused on empirical approaches. The study addresses the gap in engineering know- ledge by asking the extent to which fundamental performance parameters can explain binder ageing. Samples include extracted and recovered binders from seal specimens and original, artificially aged binders. Shift and master curve models are determined from DSR and BBR testing. Numerical model and performance parameters are analysed extensively to establish primary indicators of changes in rheology and ageing. Machine learning models are developed from the significant properties to predict the correspond- ing surfacing age. Adjustments are suggested to the extraction and recovery, and Mod- ified Kaelble shifting procedures. The regression efficiency of the Generalised Logistic and CAM master curve models are compared. Specific performance properties, e.g. G-R, ωc and Gvet , are notably capable of accounting for ageing-related changes in mechanical responses. Ageing ratios specifically indicate the relative ageing potential of binders in different behavioural ranges. Emphasis is given to machine learning model optimisation and feature selection strategies for smaller data sets. The final ageing model combines linear and non-linear, e.g. k-Nearest Neighbour and Decision Trees with bagging and boosting, machine learning models in a voting ensemble. Rheology indicators, e.g. Gvet , Tmax and C2, and crude estimators of climate and seal type are among the properties used to estimate surfacing age within an acceptable tolerance. AFRIKAANSE OPSOMMING: Die doelwit van die proefskrif is om ’n model te ontwikkel wat veranderinge in padse- ëlbindmiddel-reologie kan omskryf. Suid-Afrika is tans in ’n oorgangsfase vanaf penetra- sie-viskositeit tot uitkomsgebaseerde spesifikasies vir die klassifisering van bindmiddels. Praktyke en verouderingsanalises word steeds hoofsaaklik op empiriese benaderings ge- baseer. Die studie spreek ’n gaping in ingenieurskennis aan deur die beskrywing van veroudering deur fundamentele duursaamheidsparameters te ondersoek. Monsters uit herwinde seëlproefstukke en oorspronklike, kunsmatig-verouderde bindmiddels word in die studie geanaliseer. DSR- en BBR-toetsresultate word gebruik om gedragskurwes op te stel. Die numeriese model- en duursaamheidsparameters word geanaliseer om vas te stel wat die primêre aanwysers van veranderinge in reologie en veroudering is. Masjienleermodelle om padseëlouderdom te voorspel, word vervolgens vanuit die be- duidende eienskappe ontwikkel. Aanpassings tot die prosedures vir bindmiddelherwin- ning en gewysigde Kaelble-modellering, word voorgestel. Die doeltreffendheid van die veralgemeende logistiese- en CAM-modelle word vergelyk. Spesifieke duursaamheids- eienskappe, soos G-R, ωc en Gvet , is daartoe in staat om verouderingsverwante veran- derings in meganiese reaksies te verklaar. Verouderingsverhoudinge dui spesifiek die relatiewe verouderingspotensiaal van bindmiddels in verskeie styfheidsintervalle aan. Klem word geplaas op strategieë vir die keuse van veranderlikes en optimering van ma- sjienleermodelle en met kleiner datastelle. Die finale verouderingsmodel kombineer li- neêre en nie-lineêre, soos k-naaste-buurman en beslissingsbome, masjienleermodelle in ’n saamgestelde stemmodel. Reologiese aanwysers, soos Gvet , Tmax asook C2, en een- voudige beskrywings van klimaat- en seëlsoort, is van die eienskappe wat gebruik word om die padseëlouderdom binne ’n aanvaarbare toleransie te voorspel. Doctoral 2021-07-26T10:16:13Z 2021-12-22T14:13:06Z 2021-07-26T10:16:13Z 2021-12-22T14:13:06Z 2021-12 Thesis http://hdl.handle.net/10019.1/123631 en_ZA Stellenbosch University 279 pages application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Machine Learning UCTD Bitumen Rheology Road surfacings -- Maintenance and repair Road surfacings -- Strains and stresses Goosen, Elaine Simone Modelling of surfacing seal binder ageing and rheology |
| title | Modelling of surfacing seal binder ageing and rheology |
| title_full | Modelling of surfacing seal binder ageing and rheology |
| title_fullStr | Modelling of surfacing seal binder ageing and rheology |
| title_full_unstemmed | Modelling of surfacing seal binder ageing and rheology |
| title_short | Modelling of surfacing seal binder ageing and rheology |
| title_sort | modelling of surfacing seal binder ageing and rheology |
| topic | Machine Learning UCTD Bitumen Rheology Road surfacings -- Maintenance and repair Road surfacings -- Strains and stresses |
| url | http://hdl.handle.net/10019.1/123631 |
| work_keys_str_mv | AT goosenelainesimone modellingofsurfacingsealbinderageingandrheology |