1The overall effects of road works on global warming gas emissions

TransportationResearchPartD14(2009)576–584

ContentslistsavailableatScienceDirect

TransportationResearchPartD

journalhomepage:/locate/trd

Theoveralleffectsofroadworksonglobalwarminggasemissions

PhilippeLeperta,*,FrançoisBrilletb

a

bDivisionofRoadMaintenance,SafetyandAcoustics,LaboratoireCentraldesPontsetChaussées,44340Bouguenais,FranceFormerlywiththeDivisionofRoadMaintenance,SafetyandAcoustics,LaboratoireCentraldesPontsetChaussées,44340Bouguenais,France

articleinfoabstract

Roadcharacteristicsin uencetheamountofpollutiongeneratedbytraf c.Theygovern

therollingresistance,whichincludes:theviscoelasticbehaviourofthepavementstruc-

ture,surfacetexture,andpro leunevenness.Roadworksareperformedonaperiodicbasis

tomaintainthesecharacteristicsatproperlevelsofservice(e.g.safety,comfort,andnoise).

Whilesuchworksproduceadditionalpollution(duetocongestionaroundworksitesor

extrafuelconsumptionalongthedetourroute),theystillresultinlesspollutionoverthe

longterm.Thisarticleassessesthebalancebetweenanincreaseingasemissionsduring

roadworksandthereductioninemissionsfromtraf concetheworksarecompleted.It

isdemonstratedthatwhenroadworksareintroducedtocorrectlongitudinalpro les,

ratherthantexture,theemissionsbene tissubstantial.

PublishedbyElsevierLtd.Keywords:AtmosphericpollutionGasemissionsRoadcharacteristicsRoadtransport

1.Introduction

Overthelastdecadeorso,environmentalissueshavebecomemorewidelyaddressedinR&Dprogrammesfocusingontransport.Thisemphasisincreasedwhenitbecameapparentthattheeconomiccostofclimatechangewashigherthanmanyearlierstudieshadindicated(Stern,2007).IthasbeenfoundthatinEurope,thetransportoperationsrepresentsthesecondlargestsourceofgreenhousegas(GHG)emissions(23.4%),justbehindtheenergysector(28.9%).Further,between1990and2005,theshareinGHGemissionscontributedbyenergy-relatedindustriesfellby6%,whereasthetransportsector’sshareroseby35%(EuropeanCommission,2007/2008).Moreover,theshareofroadmodesisabout72%ofemissionsreleasedbytransport.Foragiventypeoffuel,thegasemissionsfromvehiclesaredirectlycorrelatedwithenergyconsumption,basedonanumberofphenomena,whichmaybegroupedintothreecategories:

Phenomenarelatedexclusivelyeithertodriverbehaviour(e.g.a‘‘nervous”drivingstyleincreasesdieselconsumption)

ortovehicleperformance(engineef ciency,etc.).

Phenomenathat,tosomeextent,dependonroadinfrastructureproperties:slopes,curves,pavementstiffness,surface

unevenness,surfacetexture,etc.

Roadworks,byaddingtotraf ccongestion,alsoimpactfuelconsumptionpatterns.

TheRoadManagingAuthorities(RoadDirectorate,Roadsupervisors,etc.)areincreasinglyinclinedtocontributetoreduc-ingGHGemissions.Their eldofinterventionisclearlyrestrictedtothesecondandthirdcategoriesabove,i.e.thosedealingspeci callywithroadinfrastructurecharacteristics.Takingintoaccounttheenvironmentalissuesinherentinroadworkmanagementrequiresgoodknowledgeabouttheexactin uenceofvariousroadfactorsonemissions.1

*Correspondingauthor.Tel.:+33240845820;fax:+33240845992.

E-mailaddress:philippe.lepert@lcpc.fr(P.Lepert).

1LaboratoireCentraldesPontsetChausséeshasbeeninvolvedsuchwork(LaganierandLucas,1990).

1361-9209/$-seefrontmatterPublishedbyElsevierLtd.

doi:10.1016/j.trd.2009.08.002

P.Lepert,F.Brillet/TransportationResearchPartD14(2009)576–584

Table1

Emissionsgenerated,expressedinkilogramsperlitreoffuel(kg/l).

HCCO

11.7Â10À3

5.2Â10À3NOx$044.9Â10À3Particles0.56Â10À30.75Â10À3577CO22.392.64ai

bi7.0Â10À31.3Â10À3

2.Impactsofroadmaintenanceonairpollution

Weusethetermairpollutiontorefertobothparticleandgas(CO2,CO,NOx)emissions.Becauseoftraf cconditionsandadverseweather(rain,frost,snow),roadsarecontinuallydeteriorating.Inanefforttomaintainnetworksatanoperationallevel,roadmanagingauthoritiesperformmaintenancework,2whichintermsofairpollutionexertsseveralimpacts,somenegative,somepositive:

Duringexecutionoftheseworks,equipmentpartsisrunningtoprepare,transport,spreadandcompactpavement

materials,andisgeneratingairpollution.

Improvedworksiteperformanceoftenrequiresimplementingtraf crestrictionmeasures,whichattimesmaycause

congestionandwithitpollution.

Onceworksarecompleted,pavementsurfacecharacteristicsareimproved,andthismayreducefuelconsumptionand

theairpollutiongeneratedbyvehicles.

Themainobjectiveofroadworkshasbeentopreservethestructuralstrengthofpavements.Formorethantwentyyears,improvingroadsafetyhasbeenaddedasasecondmajorobjective.Morerecently,environmentalissueshavebecomeapressingconcernforroadauthorities.Takingthe‘‘greening”objectiveintoaccountinmaintenancemanagementassistancesystemsobviouslyrequiresmodelsthatprovidereliablerelationshipsbetweenroadworks,roadoperations,vehiclecon-sumptionandgasemissions(TaylorandPatten,2006;BrilletandVadepied,2007a).Anumberofresearchprojects,carriedoutinlocationsthroughouttheworld,havebeenlaunchedinthelastfewyearstodevelopsuchmodels.InEurope,the7thFrameworkProgramme(EuropeanCommission,2008)isclearlyfocusingontransport-relatedenvironmentaleffects.InFrance,theNationalResearchAgency(ANR),agovernmentbodyresponsibleforsupervisingpublic-sectorresearch,initiatedin eldsincludingroadtransporttwoprogrammesin2008.Inthiscontext,theresearchconductedonreducingroadtrans-portpollutionispursuingthreemainapproaches:

Reductionofgasemissionsonroadworksitesbyoptingfor‘‘lowenergyconsumption”and‘‘long-lasting”maintenance

actions,throughuseofappropriateimplementationdevices.

Reductionoftraf ccongestioncausedbyroadworks,sincesuchbottlenecksgeneratemorepollutionthanthesame

leveloftraf cunderfree owconditionsatthesamesite.

Reductionofvehicleconsumptionviaanoptimalselectionofpavementcharacteristics.

Thechoiceof‘‘green”maintenanceactionsintendedtoreduceemissionscausedbytheworksiteaswellastheirultimateef ciency( rsttopic)havebeenaddressedinanumberofpapers(FederalHighwaysAdministration,1993;JullienandVen-tura,2007)andconstituteslegalprescriptions(SETRA,2004;CalifornianEnvironmentalProtectionAgency,2007).Thefol-lowingsectionswilldiscussthespeci cresearchdealingwiththelasttwoapproaches.

Wedonotmeasurevehicleemissionsdirectly.Theinstrumentationrequiredforsuchmeasurementswouldhavebeentoocostlyandcomplextooperate.Itwasthereforeassumedthatgasemissionsarecorrelatedwithfuelconsumption.Theserelationshipsareintricateandtakeintoaccount:engineef ciency,thecatalyticexhaustpipe,vehiclespeed,enginerate,etc.(BrilletandVadepied,2009)howeveradaptedagroupofmodelsinitiallydevelopedinhighwaydevelopmentandmanage-ment(HDM)torecentdatapublishedbytheEuropeanprojectARTEMIS(André,2005).Asetofsimplerlinearrelationshipsarederivedfromtheirworkforoperationalconditionsclosetothoseconsidered:

Qe;i¼aiCpetrolþbiCdieselð1ÞwhereQe,iisthequantityofemittedgases,ithetypeofemission,andCpetrolandCdieseltheconsumptionquantitiesofbothtypesoffuel.Table1providesestimatesforaiandbi.

3.Reductionofemissionsfromtraf ccongestionduetoroadworks

Toperformroadworksundersafeandef cientconditions,theclosureofoneorseverallanesisoftennecessary;suchsituationstemporarilydisturbtraf c owandmaycausecongestionduringpeakperiods.Toevaluatetheextraemissionscausedbythiscongestion,twoquestionsmustbeanswered:

Theterm‘‘maintenance”willbeusedforthoseworksdealingwithroadsurfacerenewal(resurfacing).Worksthatmodifyeitherthestructureorgeometryoftheroadway,suchasradiusofcurvatureorslope,willbereferredtoas‘‘rehabilitation”andarenotconsideredhere.2

578P.Lepert,F.Brillet/TransportationResearchPartD14(2009)576–584

Whatistheprobabilityandmagnitudeoftheresultingcongestion?

Whatistheamountofadditionalcongestion-relatedconsumption,onaper-vehiclebasis?

Researchhasbeenconductedoverthelastdecadetobuildreliablepredictionmodelsofcongestionatroadworksites(Ober-SundermeierandZackor,2001;Leper,2006).Theprinciplebehindthesemodelsissimple(Fig.1):whenandwherethetraf cdemand,q(t),exceedstheroadcapacityalongtheactualworksite,thencongestion,CR,appearsupstream.The‘‘magnitude”ofthiscongestion(numberofvehiclesinvolved)continuestoincreaseaslongasthesituationremainsun-changed.Oncethesituationhasbeenreversedhowever,congestiondiminishesandeventuallydisappearsaltogether.Suchamodel,developedduringthePAV-ECOEuropeanproject(HildebrandandLepert,1999),isappliedandre nedaspartofboththeFORMATEuropeanproject(Lepertetal.,2004)andFrenchPROPICEproject(BrilletandVadepied,2009).

Indeterministicmodels,traf cisdescribedthroughitsdailydistribution,whichdisplays2or3peaks.Inmostcaseshow-ever,theonsetofcongestionduringrushhourisaveryunstableprocesssinceitdepends,toacertainextent,onhumanbehaviour.Toovercomethisdif culty,semi-probabilisticmodelshavebeendeveloped(Brillet,2007),distributingtraf cthroughoutthedayandallyearlongconstitutesaprobabilisticinput.Congestioncalculationsareperformedforeachprob-abilitylevel,asexplainedinthedeterministicmethod;thesecalculationsprovidethedistributionofcongestionmagnitudeandthusoftheextraconsumptionduetocongestion.Themathematicalexpectancyofthisresultoffersthemostprobablevalueforextraconsumption.

Onagivenroadsegment,aslongastraf cdemandremainslessthanroadcapacity, owsareunrestrictedatrelativelyhighspeeds.Oncethedemandnearscapacity,traf c owbecomesdisturbedandmayevolveintoa‘‘stop-and-go”situation,atwhichpointthemostminorrandomeventmaytriggercongestion.Theseconditionswereexperimentallyidenti edona2Â2-lanehighway(Brillet,2007)witha90km/hspeedlimit.Theextraconsumptionofavehicletravellingthroughawork-siteandexperiencingeither owdisturbanceorcongestion,isdisplayedinTable2.

Toextendtheassessmentofextraconsumption,let’snowconsidertheexampleofonlyonelanebeingopentotraf c owineachdirection.Thedailytraf cvolumeiscomposedofNvehicles,distributedinto‘‘k”differentcategories.Letfi(i=1,k)bethepercentageofvehiclesincategory‘‘i”.Weassumethatatanytimet,thedistributionoftraf centeringthecongestedzone,q(t),isthesameasthatinthedailytraf cdistribution.Thedistributionofvehiclesinvolvedinthecongestionevent,atanytime,isthereforealsothesame.Theextraconsumptionofacategory‘‘i”vehicleincongestionisthengivenby(Fig.1):

Dcwsi¼cwsi

105ÁkX

i¼1fiÁlmið2Þwhere¼QðtÞÀCRtisthenumberofvehiclesinvolvedinthecongestioneventattimet;lmiisthelengthoccupiedbythe‘‘representative”category‘‘i”vehicle,inmetres;cwsiistheextraconsumptionincongestion,expressedinlitresper100km;and105isaconversionfactorbetweenlitres/100km(l/100km)andlitres/m(l/m).

Theextraconsumptionincongestionstartingattimet0andendingattimet1isexpressedas:

DCtot( #)"#Z Zt1"kkkkt1XXXXcwsicwsi¼fiÁqðtÞÁÁfilmidt¼fi5ÁfilmiÁqðtÞ:½QðtÞÀCRÁt dtÁ10510i¼1t0t0i¼1i¼1i¼1ð3Þ

Whennecessaryandpossible,themanagingauthorityimplementsoneormoredetours,inwhichcasethedivertedtraf cvolumemustbeestimatedovervarioushourlyperiods.Theextraconsumptionalongthesedetourswillthenonlybegen-eratedbythedifferenceinroadlengthandotherdetourcharacteristics.Undercertainconditions(lowerspeedlimitalongthedetour),theextraconsumptionmaybenegative(e.g.asindieselsavings).Asoftwarewasspeci callydeveloped(Brillet,2007)toapplythe

approach.

Fig.1.Beginning,evolutionanddisappearanceofroadcongestion.

P.Lepert,F.Brillet/TransportationResearchPartD14(2009)576–584

Table2

Speed,accelerationandfuelconsumptioninvarioustraf c owconditions(Peugeot406sedan).

DataFlowcondition

Free

Meanspeed(km/h)

Standarddeviationonspeed(km/h)

Meanacceleration(m/s2)

Standarddeviationonacceleration(m/s2)

Meanconsumption(l/100km)a

Meanextraconsumption(%offree ow)

a579Disturbed748.900.46.9+21%Stopandgo4623.000.47.4+70%Congestion43.700.337.1+555%901.100.15.7–ForaPeugeot406sedan.

Table3

Testsectionsurfacecharacteristics.

Section

E1

E2

M2

L1

aTypeRecentdenseAC0/10Traf ckeddenseAC0/10VerythinAC0/6PolishedresinMTD0.81a0.931.300.20SectionL2FGTypeSandAsphaltSpecialtechnique‘Lowfriction’AC(0/10)MTD0.500.770.62Thisreductionintexturemaycomefromthepresenceofathinbitumen lmovertheaggregates;this lmwilldisappearundertraf c.

4.Relationshipbetweenpavementconditionandfuelconsumption

Thelevelofconsumptioniscorrelatedwithrollingresistance.Threetypesofpavementcharacteristicsin uencerollingresistance:

-Nonlinearmodelparametersthatdescribepavementstructurebehaviourunderloadingalsogoverntheamountofenergylostasaresultofstructuraldeformation.

-Wearingcoursetexturedeterminestheamountoffrictionbetweentyresandpavementsurface.

-Longitudinalpro leevennesscontrolsthequantityofenergylostinsuspensionandtyres.

Thepavementstructureisdeformedintermsofaxleloading.Wheelmovementcreatesdepressionsreferredtoas‘‘de ec-tionbasins”.Ifthestructurewerepurelyelastic,thentheamountofenergyabsorbedfromvehicles(mainlylorries)duetothesedeformationswouldbeentirelyreturnedtothevehicles.Yetmostpavementmaterialsexhibitnonlinearbehaviour;thisissueiscurrentlycentraltoadebateovertheenvironmentalimpactsofvarioustypesofpavements(Perriot,2008)andliesoutsideourscope.

Atthetyre–pavementinterface,theinelasticdeformationoftyregumtreads,alongwithinelasticwearingcoursedefor-mationand‘‘stick-and-slip”friction,servestodissipatesomeenergy.Pavementsurfacecharacteristics,especiallywearingcoursetexture,partiallygovernthedissipation.Thephysicalphenomenainvolvedinthisdissipativesystemarenumerousandhighlycomplex;moreover,nomodelsaresuf cientlyre nedandreliabletoassess,byeithertheoreticalornumericalapproach,therelationshipsbetweenroadtexture,tyrecharacteristicsandenergydissipation.ExtensiveexperimentationhasbeenconductedontheLCPCtesttracktoestimatethisrelationship.Thetesttrackconsistsofalongstraightawaycontainingseveralevensections3mwide(Table3),100–250mlongandfeaturingdifferentwearingcourses.Thesesectionshavenocurvesorslopes,withjustasmallcrossfall,meaningthatvehiclescansafelytravelonthematspeedsupto130km/h.

APeugeot406sedanwasusedtoaccuratelymeasuredieselconsumptiononthevarioussurfaces.Thistestcarwasequippedwithacompletesetofsensors,acquiringposition,speed,acceleration,wheeltorqueandwithaninstrumentedaxle.Twosetsofmeasurementchannelsprovedespeciallyrelevantforthisexperiment:the rstone(wheeltorquesensors+instrumentedaxle)measuredthehorizontalaxialforceexertedbythevehicleonthepavement(rollingresistance),whilethesecond(twofuel owmeters)waspreciselyrecordinginstantaneousfuelconsumption.Severaltestsequenceswereperformed(BrilletandVadepied,2007b).Fig.2showstheevolutioninrollingresistance,3FR,asmeasuredataspeedof70km/hondifferentsections.

Rollingresistanceincreaseswithmeanpro ledepth(MPD)androughrelationshipcouldcanbederivedfromtheexperiments:

dFR 183dðMTDÞð4ÞCertaintypesofwearingcourseshoweverprovideexcellentskidresistancewithrelativelylowrollingresistance(closetotheskidresistanceofapolishedresin).Replacingthewearingcoursemaythuscontributetodecreasingrollingresistancewithoutnecessarilydecreasingskidresistance.

Moreprecisely,thedifferencebetweenrollingresistanceonthewearingcourseandrollingresistanceonaspecialpolishedresin,consideredhereinasreference0,underthesameconditions.3

580P.Lepert,F.Brillet/TransportationResearchPartD14(2009)576–584

Fig.2.RelationbetweenrollingresistanceFRandpavementtexture,underdryconditionsataspeedof70

km/hr.

Fig.3.Relationbetweenvehiclefuelconsumptionandspeedfordifferenttextures,underdryconditions.

Theimpactoftextureonfuelconsumptionisnotasstraightforward.Sinceconsumptiondependsheavilyonenginerate,thein uenceoftextureonfuelconsumptionasafunctionofspeedwasmeasuredinthemostappropriategear(i.e.theoneminimisingconsumptionatthegivenspeed).Ataround70km/h,consumptionofthe406sedanonanactualpavementvar-ies(Fig.3)fromaminimumof5.5l/100km(recentAC,FR=78N)toamaximumof6.1l/100km(olderAC,FR=250N),thusindicatingthatatapprox.70km/h,a172-Nreductioninducesasavingsof0.6l/100km.Forthissedantravellingatthisspeed,eachadditionaljoulerequiredtoovercomerollingresistanceimpliesanextrafuelconsumptionof:

dq¼0:5¼2:9Â10À3l=100km=Nð5ÞFig.3howeversuggeststhatthisrelationmustbeconsideredcarefully.Therathersmallimpactofpavementtextureonconsumptioniseasilyunderstood,giventhattheenergydissipatedtocompensateforrollingresistanceisonlyonecompo-nentoftheenergyrequiredtomovethevehicle,withmostofthisbeingusedtocompensateforinternalmechanicalfriction,especiallyatlowspeed,andaerodynamicresistance,athigherspeed.

Onunevenpavements,vehiclesuspensionissubjecttodifferentialmotionbetweenthewheelsandthevehiclebody.Thesemotionsarelimitedbythedampersthatdissipateenergywhenheated.Asopposedtowhatwasobservedforenergydissipatedatthetyre/pavementinterface,itisnoteasytoconductexperimentsinwhichpavementunevennesscanbepre-ciselycontrolledanddistinguishedfromotherparametersin uencingenergydissipation.Fortunately,thephenomenain-volvedinvehiclesuspensioncanbeapproximatedwithdecentreliabilitybymeansofsimpletheoreticalmodels.

ThisstudywascarriedoutonamodelofthePeugeot406sedan.Onlyaquarterofthevehiclewasmodelledtoworkwithjustasingledegreeoffreedom.Thevehiclewasassumedtorunatagivenconstantspeed‘‘V”,withthequarter-vehicleas-signedamass‘‘M”.Thelowersuspensionextremityliesincontact(throughthetyre)withtheroadpro le,whichhasbeenassimilatedtoapuresinefunctionofamplitude‘‘y”andwavelength‘‘L”.Accordingtoasimpledynamicequation,theuppersuspensionextremityfollowsasinefunctionofamplitude‘‘x”andwavelength‘‘L”(Fig.4

).

P.Lepert,F.Brillet/TransportationResearchPartD14(2009)576–584581

Fig.4.Dynamicquarter-vehiclemodel.

Theroadpro leischaracterisedbyaspeci cindicator.InFranceforinstance,threeindiceshavebeende nedaspropor-tionaltothelogarithmofthepro lesignalenergycontentineachofthreewavelength(WL)ranges(Martinetal.,2000):shortWL(0.707–2.828m),mediumWL(0.282–11.312m),andlongWL(11.312–45.248m).Theamplitude‘‘x”canbeeasilyderivedfromthesystemequations:

_þKx¼Cy_þKyM

本文来源：https://www.bwwdw.com/article/feb4.html