PAVDAM - probabilistic and volumetric design of asphalt mixtures

Smit, Andre de Fortier (2002-12)

Thesis (PhD)--University of Stellenbosch, 2002.

Thesis

ENGLISH ABSTRACT: The dissertation presents a design procedure for HMA mixes based on probabilistic and volumetric approaches, hereafter referred to as PAVDAM. Central to PAVDAM is the use of an analytical model for estimating the voids in the mineral aggregate (VMA) of asphalt mixes. Validation of the mix design procedure was done through accelerated pavement testing (APT) with the Model Mobile Load Simulator (MMLS3) at the National Center for Asphalt Technology (NCAT) test track in Opelika, Alabama. In addition, the semi-circular bending (SCB) test was evaluated to serve as an analysis tool to augment the proposed mix design procedure. Non-linear finite element analyses using a quasi-isotropic material model were done using the CAPA-3D finite element system developed at the Technical University of Delft in the Netherlands to better characterise the tensile strength properties of specimens tested with the SCB. PAVDAM is partly based on performance related and analytical procedures such as the Stategic Highway Research Program (SHRP) Superpave and the Belgium Road Research Centre (BRRC) design method. The mix design system was developed based on gyratory compaction procedures. In this regard, the criteria relating gyratory compaction to design traffic as specified in Superpave are used. PAVDAM differs from other mix design methods in that a probabilistic approach is used to account for the variabilities associated with mixture components and properties during the manufacturing stage. It serves as a subset of the volumetric optimisation stage of the mix design process. The development of an analytical model to estimate the VMA of an asphalt mix was the central theme of the dissertation. The analytical model developed is based on modified Aim and Toufar (MAT) packing models developed under SHRP research and used in the concrete industry. The MAT packing models are based on the theory underlying the packing of monosized spheres and the combination of successive binary systems. These models were further refined for use in the asphalt industry. A probabilistic procedure based on the BRRC PRADO packing model is used to account for the influence of size ratio of the successive monosized binary systems. The result was a model that allows an estimation of the VMA of a mix from the gradation of the mix, the voids in the filler and the porosities of the individual aggregate fractions making up the mix. Research was undertaken to couple VMA estimates with gyratory compaction levels. This allowed estimates to be made of the optimum binder contents of mixes for different design traffic levels. The VMA of a mix is difficult to estimate accurately since it is difficult to quantify the factors influencing VMA such as gradation, particle shape, angularity, texture and rugosity. Furthermore, the influence of binder content and compaction must be taken into account. The MAT packing model underestimates the VMA of mixes compared to measured values. For this reason it is necessary to calibrate the model to allow more accurate estimations. More sophisticated models are required to more accurately estimate the VMA of mixes. It is recommended that the development of these be explored further. Asphaltic materials are inherently heterogeneous and there are a large number of factors that influence their volumetric properties. Because of this, Monte Carlo simulation techniques are used in PAVDAM to evaluate the combined effect of the variabilities of significant material properties. The dissertation expands on the different variabilities and the effects of variability on mixture volumetrics and mix design verification. The dissertation outlines the algorithms and procedures used in PAVDAM to estimate the binder content of a mix. In order to validate the PAVDAM model, analyses were done to determine the reliability of specific NCAT MMLS3 test section mixes in terms of densification in the field. A comparison of PAVDAM estimated and field binder contents allowed a ranking of the reliabilities of the different section mixes in terms of field densification at the design traffic level. This ranking compared favourably with that obtained from an analysis of actual densification trends monitored in the field under full-scale trafficking. Initial FEM analyses of the SCB using linear elastic isotropic modelling allowed the development of equations to characterise the tensile strength and modulus characteristics of specimens tested using the SCB. It was emphasized that these equations do not provide a realistic assessment of the strengths or moduli of asphaltic materials. The strengths and moduli of these materials are influenced by strain rates within the materials that cannot be assessed using a simple linear elastic approach. To address this, an alternative FEM analyses using CAPA-3D was undertaken. An approach was adopted to account for the influence of tensile and compressive strain rates on modulus. The analyses made use of a quasi-isotropic material model and it was shown to better characterise the tensile strengths of HMA materials using the SCB. The analyses also indicated that the tensile strengths determined using the equations initially developed based on a linear elastic approach result in strengths that are unrealistically high. It is recommended that further finite element research be done using non-linear material modelling to evaluate the very complex stress-strain conditions within an SCB specimen to better characterise fracture response. It is also recommended that the fatigue characterisation of HMA be explored based on strength tests using the SCB. PAVDAM represents a rational approach to mix design, a shift from experimental empiricism towards scientific fundamentalism. PAVDAM can be used to define the spatial composition of asphalt mixes. The influence of mix component variability may be addressed and reliability assessments of candidate gradations are possible during volumetric optimisation. Furthermore, changes in the volumetric properties of asphalt mixes may be investigated. As such, PAVDAM is a mix design management tool and can only be effective when used as part of a system that closely monitors variability and systematically refines the underlying packing model.

AFRIKAANSE OPSOMMING: Die verhandeling lewer 'n asfalt mengselontwerp metode gebaseer op statistiese en volumetriese metodes, genaamd PAVDAM. PAVDAM gebruik 'n analitise model om 'n skatting te maak van die ruimtes in die aggregaat (VMA) van asfalt mengsels. Stawing van die mengselontwerp metode is gedoen deur vesnelde plaveisel toetse met die Model Mobiele Las Simuleerder (MMLS3) op die National Center for Asphalt Technology (NCAT) se toetsbaan in Opelika, Alabama. Die Semi-Circular Bending (SCB) toets is ook geëvalueer om die voorgestelde ontwerp metode te analiseer. Nielineêre eindige element analise met gebruik van 'n kwasi-isotropiese materiaal model is gedoen met die CAPA-3D eindige element (FEM) stelselontwikkel deur die Tegniese Universitieit van Delft in Nederland om die treksterkte van monsters getoets met die SCB beter te karakteriseer. PAVDAM is deels gebaseer op gedragsverwante en analitise prosedures soos die Strategic Highway Research Program (SHRP) Superpave en die Belgium Road Research Centre (BRRC) ontwerp metodes. Die mengselontwerp stelsel is ontwikkel gebaseer op girator verdigtingsmetodes. Die girator verdigtingskriteria soos voorgeskryf in Superpave is gebruik. PAVDAM verskil van ander mengselontwerp metodes omdat daar gebruik gemaak word van statistiese metodes om die veranderlikhede geassosieer met mengsel komponente en ontwerp prosedures gedurende die vervaardigingsfase. Dit dien as 'n komponent van die volumetriese optimeringsfase van 'n mengsel ontwerp proses. Die sentrale tema van die verhandeling was die ontwikkeling van 'n analitisie model om 'n skatting te maak van die VMA van asfalt mengsels. Die analitise model wat ontwikkel is is gebaseer op gewysigde Aim and Toufar (MAT) pakkingsmodelle ontwikkel as deel van SHRP navorsing en wat gebruik word deur die betonindustrie. Die MAT pakkingsmodelle is gebaseer op die teorie van die pakking van eenvormige groote sfere en die kombinasie van binêre sisteme. Die modelle is verder ontwikkel vir gebruik in die asfaltindustrie. 'n Statistiese prosedure gebaseer op die BRRC se PRADO pakkingsmodel is gebruik om die invloed van groote verhouding (size ratio) van opeenvolgende binêre sisteme in ag te neem. Die gevolg was 'n model wat gebruik kan word om 'n skatting te maak van die VMA van 'n mengsel vanaf die mengsel gradering, ruimtes in die vulstof en die porositeit van die individuele aggregaat fraksies wat die mengsel saamstel. Navorsing is gedoen om die VMA te koppel aan girator verdigtingsvlakke. Die gevolg is dat skattings gemaak kan word van die optimum bindstof inhoude van mengsels vir verskillende verkeer ontwerp vlakke. Dit is moeilik om 'n akkurate skatting van die VMA van mengsels te maak omdat dit moelik is om faktore wat VMA beïnvloed te kwantifiseer, soos byvoorbeeld gradering, partikel vorm, hoekigheid en tekstuur. Die invloed van bindstof inhoud en verdigting moet ook in ag geneem word. Die MAT pakkings modelonderskat die VMA van mengsels in vergeleke met gemete waardes. Daarom is dit noodsaaklik dat die model gekalibreer word om meer akkurate skattings van mengsel VMA te maak. 'n Meer gesofistikeerde model is nodig om meer akkuraat die VMA van mengsels te skat. Dit word voorgestel dat die ontwikkeling van hierdie modelle verder ondersoek word. Asfalt materiaale is inherent heterogeen en daar is 'n groot aantal faktore wat die volumetriese einskappe van die mengsels beïnvloed. Om hierdie rede word gebruik gemaak van Monte Carlo simulasie in PAVDAM om die gekombineerde effek van veranderlikheid in ag te neem. Die verhandeling brei verder uit op verskillende veranderlikhede en die effek van veranderlikheid op die volumetriese einskappe van mengsels en die stawing van die model. Die algoritme en prosedures wat deur PAVDAM gebruik word om 'n skatting te maak van die optimum bindstof inhoud van 'n mengsel word in die verhandeling uiteengesit. Vir stawing van PAVDAM is analises gedoen om die betroubaarheid van NCAT MMLS3 toets seksies in terme van verdigting in die veld te bepaal. Vergelykings tussen PAVDAM geskatte en veld bindstof inhoude het gelei tot 'n rangorde van die betroubaarheid van die verskillende seksie mengsels in terme van verdigting in die veld onder die ontwerp vervoer. Hierdie rangorde het goed vergelyk met die wat gekry is van analise van die werklike verdigtingstendense soos in die veld gemonitor is onder volskaalse verkeer. Voorlopige FEM analise van die 8CB met gebruik van lineêr isotropiese modelering het gelei tot die ontwikkeling van vergelykings om die treksterke en styfheidseienskappe van monsters getoets met die 8CB te skat. Dit word beklemtoon dat hierdie vergelykings nie 'n realistiese skatting van die treksterkte of styfhede van asfalt materiaale gee nie. Die treksterkte en styfhede van die materiaale word beïnvloed deur verplasingstempo's binne die materiaale wat nie deur lineêr elastiese benaderings bepaal kan word nie. Om dit aan te spreek is alternatiewe FEM analise gedoen deur gebruik te maak van CAPA-3D. 'n Benadering is gevolg om die invloed van trek en druk verplasingstempo's op styfheid in ag te neem. Die analise het gebruik gemaak van 'n kwasi-isotropiese materiaal model en dit het 'n beter karakterisering van die treksterkte van asfalt mengsels tot gevolg gehad. Dit word voorgestel dat verdere eindige element navorsing gedoen word met gebruik van nie-lineêre materiaal modellering om die breekgedrag van 8CB monsters beter te kan karakteriseer. Dit word ook voorgestel dat die vermoeiings eienskappe van asfalt mengsels ondersoek word gebaseer op treksterketoetse met gebruik van die 8CB. PAVDAM is 'n rasionele benadering tot mengselontwerp, 'n paradigma skuif weg van empiriese metodes tot wetenskaplik fundamentele metodes. PAVDAM kan gebruik word om die ruimtelike samestelling van asfalt mengsels te definieer. Die invloed van veranderlikheid kan in ag geneem word en betroubaarheidsskattings van kandidaat mengsels is moontlik gedurende die volumetriese optimiseringsfase. Verder kan veranderings in die volumetriese eienskappe van mengsels ondersoek word. As sulks is PAVDAM 'n hulpmiddel vir die bestuur van mengselontwerp en kan net effektief wees wanneer dit gebruik word as deel van 'n stelsel wat die mengselveranderlikheid monitor en sistematies die onderliggende pakkings model verbeter.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/52987
This item appears in the following collections: