Model-based determination of hydrogen system emissions.pdf

Available at vhch,ch,significantly cheaper than liquid fuels since it is left behind as2and additionally can be produced from waste 1. Hydrogen asoverpressure. Thus, it is of great interest to developers,theAll these situations can be simulated on chassis dynam-ARTICLE IN PRESSINTERNATIONAL JOURNAL OF HYDROGEN ENERGY 33 (2008) 863869C3Corresponding author: Tel.: +41448234679; fax: +41448234044.fuel for fuel cells as well as for I.C. engines is likely to play animportant role in future vehicle propulsion technology. Thedevelopment of hydrogen-powered vehicles is also driven byometers embedded in climatic chambers. However, it is onlypossible with enormous effort to keep such climatic cham-bers so airtight that the evaporative emissions can be0360-3199/$-see front matter & 2007 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rights reserved.doi:10.1016/j.ijhydene.2007.09.041E-mail address: martin.weilenmannempa.ch (M. Weilenmann).can be used as natural gas shows one of the highest field-to-wheel efficiencies and the best CO balance among biofuelscar subject to changing ambient conditions and emissions atsystem start, while running and at system stop 4,5.waste in the refinery process. Natural gas as fuel offersnotable CO2and exhaust emission benefits over gasoline anddiesel. In addition, the production of biogenic methane thatmanufacturers and legislators to be able to monitoroverall system emissions of these gaseous fuels underrealistic conditions. This holds for both the case of a parked1. IntroductionVehicles with gaseous fuels are becoming more and morecommon as they show many advantages compared to gaso-line or diesel fuelled cars 1. In some countries LPG isair quality factors, the problem of CO2and other greenhousegases and fossil fuel supply dependency 2,3.For all these gaseous fuels there are different fuel storagesystems such as high-pressure gas bottles, low temperatureliquidation, metal hydrides and others, operating at a certainArticle history:Received 21 March 2007Received in revised form28 September 2007Accepted 28 September 2007Available online 26 November 2007Keywords:Fuel cellGaseous fuelsHydrogenSystem emissionsTest cellConcentration measurementBecause of air quality problems, the problem of CO2related greenhouse gas emissionsand shortage of fossil fuels, many vehicles with gaseous fuels (CNG, biogas, hydrogen, etc.)are under research and development. Such vehicles have to prove that both their exhaustemissions and their overall system emissions (including running loss) remain belowcertain safety limits before they can be used in practice. This paper presents a cost-effective way of monitoring such system emissions from hydrogen or other gaseousfuel powered vehicles within an air-conditioned chassis dynamometer test cell, ascommonly used for low ambient emission tests on gasoline vehicles. The only additionalequipment needed is a low-concentration sensor for the gas of interest (e.g. hydrogen).The method is based on concentration measurements and a dynamic mass balance model.It is shown by a real experiment that very low emissions can be recorded. Additionally,error bounds and sensitivities on different parameters such as air exchange ratio arequantified.& 2007 International Association for Hydrogen Energy. Published by Elsevier Ltd. All rightsreserved.article info abstractjournal homepage: www.elseModel-based determination ofof motor vehicles using climate-cMartin Weilenmanna,C3, Christian Bacha, PhilippeaEMPA, Swiss Federal Laboratories for Materials Testing and ResearbEMPA, Swiss Federal Laboratories for Materials Testing and ResearCH-8600 Duebendorf, S/locate/ijhydeneydrogen system emissionshamber test facilitiesNovaka, Andrea Fischerb, Matthias HillbI.C. Engines Laboratory, CH-8600 Duebendorf, SwitzerlandAir Pollution/Environmental Technology Laboratory,measured by monitoring the increasing gas concentrationwithin the chamber directly. The methods to monitorevaporative hydrocarbon emissions from gasoline vehiclesdescribed in 6,7 allow either the measurement in an airtightchamber (SHED sealed housing for evaporative determina-tion) or the measurement with the so-called point sourcemethod. While the SHED-method works well for test on thestanding car, severe sealing problems occurred for running-loss tests, where a chassis dynamometer is neededwithin theSHED. For hydrogen in particular, the sealing of the housingposes even more of a challenge than for evaporative emissionmeasurement of hydrocarbons. In the alternative pointsource method all potential points of leakage as fuel tankcap or vent opening need to be equipped with funnels leadingto an analyser with an appropriate air flow, strong enough tocollect all emissions but small enough to keep concentrationsmeasurable. Thus, this method results in rather extendedequipment that needs to be adapted to each vehicle, and onecannot be sure that the total of evaporative emissions is reallymonitored.Alternatively this paper presents a method to monitorrunning loss and system emissions of gas-fuelled cars byapplying the mass balance method to a climatic test cell withventilation. It is shown what sensor equipment is needed andhow the source emissions are calculated. The method isvalidated by tests. A sensitivity analysis is also presentedshowing what limiting conditions have to be fulfilled to reacha certain quality of measurement.2. Methodology2.1. Mass balanceThe basic idea of this approach is the fact that atoms cannotvanish. Based on Fig. 1 showing a chassis dynamometer in aclimatic chamber, this leads to the mass balance of Eq. (1).The change in mass of a gaseous matter (subsequently calledgas G) inside the chamber is the difference between all massflowing into the chamber and all mass flowing out of thechamber. This assumes that no chemical reactions betweenthe gas concerned and other matter take place. This istypically true of hydrogen, methane or propane at ambientconditions and concentrations below 1ppm 8.qmGqtXmC3G;inC0XmC3G;out,qmGqt mC3G;vent_inmC3G;carC0XmC3G;out. 1qmGqtdenotes the change in mass of gas G within the cell,PmC3G;inthe sum of all mass flows of gas G into the chamberandPmC3G;outthe sum of all mass flows of gas G out of thechamber, mC3G;vent_inthe mass flow into the chamber fromventilation and mC3G;carthe source flow of interest. All variablesare functions of time.A mass flow of gas G into the chamber occurs if this matteris found in the ambient air. Thus, the mass flow of ventilationair and the concentration of gas G in the intake air need to beARTICLE IN PRESSINTERNATIONALJOURNALOFHYDROGENENERGY33 (2008) 863869864Drivers aidAirstream ventilatorMax. flow: 100000 m3/hInternal ventilation, airconditionning heat exchangerVintake*VleakageFig. 1 Sketch of climatic chambChassis dynamometerVair,vent_in*VexhaustVair,vent_outer with ventilation flows.Inchalocation.Naturbedeterminemair rairVair. (2)ARTICLE IN PRESSOGENThe contained mass flow of gas G then ismC3G cGrGVC3air, (3)where cGis the concentration of gas G and rGits density. SinceC3thedaduct of air density rairand volume flow Vair.C3prothe cell. It is thus sufficient to measure the air inflow.addition, since the concentration inside the climaticmber is homogenous, it needs to be measured at just oneally, a flow of gas G as the inflow mC3G;vent_incannotmeasured directly. By assuming ideal gases, it may bed as follows. Any mass flow of air mC3airis theC3intomeasured. The second mass flow into the chamber is theevaporation from the vehicle, which is of interest.There are different possibilities for flows of gas G out of thechamber:C15 Intended ventilation.C15 Leakage. The doors of the chamber as well as channels forcables and pipes are not airtight, so some air leaks. Mostclimatic chambers operate with a slight overpressure toensure that air flows out at all openings, since inflowinghumid air, when operating at low temperatures, wouldcause dangerous ice formation and additionally disturbthe humidity control of the chamber (Fig. 1).C15 If the vehicle is running and is propelled by a system thatconsumes air (engine or fuel cell system), either thecorresponding air supply can be from outside the chamberor air from the room is used. Since the exhaust gases aretypically led outside the chamber and measured there, thelatter case is also an outflow for the mass balance of gas G.It is obviously not possible to measure the mass flow andconcentrations of gas G at all the outflow locations, but thisproblem can be bypassed by the following approach.The chassis dynamometers for exhaust emission measure-ments are equipped with fans for the cooling of the vehicle.Together with the ventilation of the air conditioning of thecell, this can cause such high turbulence that the concentra-tion of gas G in the room can be considered to behomogeneously distributed. In other words the mixing timeconstant in the room must be significantly lower than the airexchange rate. It must be ensured that no dead zones whereventilation is poor exist inside the climatic chamber. In otherwords, in most cases where chassis dynamometers areinstalled in climatic cells, the rolls of the dynamometer aswell as the breaking electric motor are in an under-floorcompartment that is contained in the cell. It must thereforebe possible for this compartment to be ventilated intention-ally by opening covers and adding additional ventilators.If the concentration of gas G inside the chamber is indeedhomogenous and measured, this concentration also holds forall the outflows of the cell. As long as the pressure remainsstable within the cell, which is controlled by the ventilation,the total mass flow of air out of the cell is equal to the flowINTERNATIONAL JOURNAL OF HYDRtests take place in a climatic chamber and do not last forys, it may be assumed that both temperature and pressureremain stable, and thus that densities are constant. It is thussufficient to measure the volume flow of air and theconcentration of gas G to determine its mass flow. For thechamber it correspondingly holds thatmG cGrGVch. (4)The index ch stands for chamber. Assuming that the volumeflow of air out of the chamber is equal to the inflow and thatthe distribution of gas G in the chamber is homogenous,Eqs. (1)(4) giveqcG;chqtrGVch cG;vent_inrGVC3air;vent_in mC3G;carC0cG;chrGVC3air;vent_in. 5And this can be solved for the mass flow of the source, thusthe carmC3G;carqcG;chqtrGVchC0 cG;vent_inrGVC3air;vent_in cG;chrGVC3air;vent_in. 6So, the system emissions as mass per time unit can becalculated by knowing the chamber volume, the density ofgas G (thus temperature and pressure) and measuring thevolume flow of air into the chamber as well as the gasconcentration of G inside the chamber and in the air intake.As pressure and temperature in inflow and outflow are alike,densities for both flows can be considered to be equal.2.2. Measurement equipmentA commercial gas chromatograph (Reduction Gas Analyzer(RGA3), Trace Analytical Inc., California, USA) was used tomeasure H2inside the climatic chamber. The RGA3 is anultra-trace level gas detection system capable of monitoringlow ppb concentrations of reducing gases such as H2.The instrument consists of a microprocessor-controlled gaschromatograph which utilises method of reduction gasdetection.Synthetic air preconditioned by molecular sieve 5AandSOFNOCAT to remove H2O and reaction impurities (CO andH2) is used as carrier gas. Aliquots of air samples are flushedwith a rate of 20ml/min over a 1ml sample loop. Afterequilibration, the sample volume is injected onto thecolumns. Sample components of interest are separatedchromatographically in an isothermal mandrel-heating col-umn oven. The chromatographic precolumn (Unibeads 1S,60/80 mesh; 1=800C23000) is mainly used to remove CO2,H2Oand hydrocarbons. Subsequently H2and CO are separated bythe analytical column (molecular sieve 5A, 60/80 mesh;1=800C23000) and pass into the detector which contains aheated bed of mercuric oxide. Within the bed a reactionbetween mercuric oxide (solid) and H2occurs and theresultant mercury vapour in the reaction is quantitativelydetermined by means of an ultraviolet photometer locatedimmediately downstream of the reaction bed. The columnsare kept at 751C; the detector is heated to 2701C. The amountof H2in the air sample is proportional to the amount ofmercury that is determined.ENERGY 33 (2008) 863869 865During the quasi-continuous observations of the H2con-centration in the test chamber, measurements were takenevery 2min. At the beginning and end of each test cyclethe ambient air concentration (concentration of the ventila-tion inflow) was measured for 30min. Typically the concen-trations were very constant over the short time of one testcycle and in the range of the mean of 576C694ppb atDuebendorf 9.Two high concentration reference gases (50 and 100.2ppmH2; Messer Schweiz, Switzerland) were dynamically dilutedwith zero air to the range of interest by means of a dilutionunit (MKAL diluter, Breitfuss Messtechnik GmbH, Harpstedt,Germany). The dilution unit was indirectly referenced againstthe primary gas flow standard of the Swiss Federal Office ofchange substantially; thus average concentration during onesampling step is approximated by the mean of the valuesSo the mass emitted during the sampling interval k ismG;car;k rGcG;ch;kC0 cG;ch;kC01Vch VC3air;vent_in;kTC18C2cG;ch;k cG;ch;kC012C0 cG;vent_in;kC18C19C19. 9Mathematicallymore complexbut also more accurate is thediscretisation by solving the differential equation (5) analyti-cally for one time step, what needs certain assumptions.Here there is freedom to assume all input signals (i.e.VC3air;vent_int, cG;vent_int, mC3G;cart) as arbitrary functions of time.ARTICLE IN PRESSINTERNATIONALJOURNALOFHYDROGENENERGY33 (2008) 863869866measured at either end of it. The mass balance results inmC3G;car;kcG;ch;kC0 cG;ch;kC01TrGVchC0cG;vent_in;krGVC3air;vent_in;kcG;ch;k cG;ch;kC012rGVC3air;vent_in;k. 8*mG,car(t)tCase: early peak Case: constant*mG,car(t)Metrology. The different mixtures of the two high concentra-tion standards showed excellent agreement with each otherand the NOAA/GDM scale 10. Detection limit for H2was C610ppb and the standard uncertainty of the measure-ment 5%.2.3. Analysis methodologyAs described in the previous section the low concentrations ofthe gases of interest cannot be measured with high timeresolution, i.e. within seconds. The equipment describedabove allows a sampling rate of 2min. Thus Eq. (6) needs to besolved discretely.The most direct and simple approach of discretisation isreplacing the derivative of the chamber concentration by thedifference of the last two measured values. For time step kthis results inqcG;chqtt kTC25cG;ch;kC0 cG;ch;kC01T, (7)where T is the sampling interval 11. Since both ambientconcentration of gas G and ventilation air flow typicallychange very little over one time interval, it does not matter ifthe values at the beginning or the end of the samplinginterval are used. The chamber concentration, however, may(k-1)T kT (k-1)TFig. 2 Most extreme possibilities of time functiHence, if necessary it might be possible to measure theventilation air flow at high time resolution and use that timefunction for the calculus, but usually this flow is reasonablyconstant. The ambient concentration of the gas G typically isconstant too if not working downwind of a huge non-uniformgas source. Of course the time function mC3G;cart of how thevehicle emits the gas G is unknown. If the total mass emittedduring one time step mG;car;kis given, the most extreme casefor the calculus is if all of it is released immediately after thetime interval starts or immediately before the time intervalends (peak functions, Fig. 2). The average case happens ifthe vehicle is constantly emitting gas G. For benchmarkingthe quality of this methodology in Section 3.2, Eq. (5) is solvedsubsequently for all three assumptions.In the case of the early peak, the solution of Eq. (5) for thetime t kT iscG;ch;k cG;vent_in cG;ch;kC01mG;carrVC0 cG;vent_inC18C19eC0VC3=VT, (10)and thus, the mass of gas G emitted in one time period ismG;car;k rVcG;ch;kC0 cG;vent_ineC0VC3=VT cG;vent_inC0cG;ch;kC01!. (11)For the late peak case we obtaincG;ch;k cG;vent_incG;ch;kC01C0 cG;vent_ineC0VC3=VTmG;carrV, (12)mG;car;k rVcG;ch;kC0 cG;vent_incG;vent_inC0cG;ch;kC01eC0VC3=VT.(13)ttCase: late peak*mG,car(t)kT (k-1)T kTons of gas release for benchmarking.And for the average case of a constant emitting sourcemC3G;cartmG;car;k=T:cG;ch;k cG;vent_inmG;carrTVC3 cG;ch;kC01C0 cG;vent_inC0mG;carrTVC301AeC0VC3=VT,(14)mG;car;krTVC3VcG;ch;kC0 cG;ch;kC01eC0VC3=VT1C0eC0VC3=VTC0 cG;vent_in01A. (15)Even though Eqs. (11), (13) and (15) look rather different,their outputs remain similar as long as the sampling intervalTis small compared to the ventilations time constant V=VC3.Sothe quality of this method rises if both the sampling intervalvalues belong to this test equipment.heat exchange units, etc., are difficult to describe. Thus a testwhere a well-defined volume of helium was released im-3.2. Identification of volume flow and validationIf the volume flow of the ventilation is not possible to bemeasured directly, but is constant over time, it is possible todetermine it by the following test.As above, a certain volume of a measurable gas such ashelium is injected into the cell (with ventilation running).After the mixing phase in the cell the helium concentrationwill follow Eq. (5) or one of its solutions (10), (12) or (14) with azero source activity of the car mC3G;cart0. Measurements areshown in Fig. 3. Subtracting the background concentrationand building the logarithm of the He concentration result in astraight line for the first 2000s where concentrations arereasonably above the detectio