1、 by FEV all rights reserved. Confidential no passing on to third partiesPrepared forIAC 2016Long Qingyang, Business Unit Leader of Vehicles, FEV ChinaFEV GASOLINE ENGINE DEVELOPMENTMANAGING REAL DRIVING EMISSIONSShanghai, Nov 30th, 2016 by FEV all rights reserved. Confidential no passing on to third
2、 parties |FEV observes that RDE has significant impact on powertrain layout and that specific development tools are required22016-10-28_FEV-Gasoline-delivered-02.pptxStringent Legislation EC, “Comission Regulation (EU) 2016/427”, March 31st 2016; Post Euro 6Euro 6dEuro 6d TEMPEuro 6b2015 2016 2017 2
3、018 2019 2020 2021 2022 2023 2024 2025Particulate Number Emission LimitsNEDC CycleWLTP ProceduresWLTP Additions (-7, MAC, OCE, OBDelipsis)New Pollutants and GHGReal Driving EmissionsCO2 FleetMobile Air Conditioning6 x 1012 1/km Post Euro 66 x 1011 1/kmPhase-OutPhase-InConf. Factor Step 1: CFPN = 1.5
4、 (CFNOx = 2.1)NO2, N2O, NH3, Ethanol, AldehydesPhase-InEstimated Implementation Schedule130 PC (175 LDV) g/km on NEDC 95 PC (147 LDV) g/km on NEDCPhase-Intbd g/km on WLTP tbd g/km on WLTPMonito-ringConformity Factor Step 2:CFPN/NOx = 1 + 0.568-78Market DriversEmission Roadmap: Automotive Industry un
5、der PressureLead-time New MAC procedure3 by FEV all rights reserved. Confidential no passing on to third parties |Market Drivers:Emission Roadmap China: Automotive Industry under PressureSources: FEV research and analysis; BGCF_ChinaVI Pre-draft summary_17052016, Limits and Measurement Methods for E
6、missions from Light-duty Vehicles (CHINA 6) 2016-04-28Post CN6bCN6bCN6aCN5elipsis2019 2020 2021 2022 2023 2024 2025 2026 2027 2028Emission LimitsParticulate Number Emission LimitsNEDC CycleWLTP ProceduresWLTP Additions (-7, MAC, OCE, OBDelipsis)New Pollutants and GHGReal Driving Emissions_ Post CN6b
7、6 x 1011 1/kmPhase outAdopt WLTP ProceduresAddition of N2O limit. Estimated Implementation Schedule (dates not fixed yet)Implementation of combined NMHC+NOXOBD threshold, introduction of NOX-limit for cold conditionsConformity Factor:CFPN = 2.1, CFNOx = 2.1Adoption of EU6d methodology for RDE + Meas
8、ure and record of CO and cold start phase emissions6 x 1011 1/kmPost CN6bCO -50%, other asEU6CO, THC -50%, NMHC -48%, NOx -42%, PM -33%Equal toEU6Monitor and reportExtended boundary conditions to 2400 m altitude (3 categories)Beijing announced to start CN6b already by 2018Beijing 2018,Shanghai 2019#
9、 ANIMATED by FEV all rights reserved. Confidential no passing on to third parties |Source: FEV; NEDC: New European Driving Cycle; WLTP = Worldwide Harmonized Light-Duty Vehicles Test Procedures; RDE = Real Driving EmissionNew European regulation framework increases the complexity of powertrain devel
10、opmentVariable driving cycle: road gradient, road curves, compositionAdditional consumer (A/C, entertainment system,elipsis)Individual driving style (smooth, hectic,elipsis)Variable by temperature, altitude, wind, humidity, traffic conditionsRDEFixed driving cycleRoller load is set to individual veh
11、icle specification, increases load compared to NEDCDefined driving styleFixed within small tolerancesWLTP vs NEDCDriving profileVehicle loadDriving StyleEnvironmental conditionsRoller Dyno RoadCO2 emission target values are assumed to be determined in WLTP from 2021NOx emission limits are challengin
12、g regarding the required exhaust aftertreatmentsystem efficiency (Diesel)Particulate number emission limit is major challenge for gasoline engines in RDEkg2016-10-28_FEV-Gasoline-delivered-02.pptx 5 by FEV all rights reserved. Confidential no passing on to third parties |Challenge: conducted CO2-tar
13、gets in Europe are based on NEDC, the conversion to the WLTP test procedure implies a tightening of CO2-targetsSWITCH FROM NEDC TO WLTP TEST PROCEDURE;Source: UNECE, European Commission, FEVBoundary conditionssquare6 Adopted CO2 fleet targets are valid for NEDC onlyAssumption square6 CO2 difference
14、based on switch to WLTP driving cycle will be implemented into a NEDC barb2right WLTP correlation factorsquare6 The correlation factor will significantly reduce the CO2 difference between NEDC and WLTP test procedure square6 Target of the WLTP test procedure is also a more realistic consideration of
15、 weight and rolling resistancebarb2rightAdd. tightening of CO2-targets NEDC test procedure0204060801001201400 400 800 1200v eh i cl e s pe ed /k p/ htime / sECE EUDCWLTP test procedureMain differencessquare6 Weightsquare6 Rolling resistancessquare6 Aerodynamic features square6 Driving cyclesquare6 e
16、lipsis0204060801001201400 450 900 1350 1800v eh i cl e s pe ed /k p/ htime / sLow Extra HighHighMedium# BACKGROUND2016-10-28_FEV-Gasoline-delivered-02.pptx 6 by FEV all rights reserved. Confidential no passing on to third parties |Improvement WLTP TMLvs. NEDC 4 %23 %24 %36 %36 %68 %5 %33 %75 %+62 %8
17、 %82 %Most of the considered technologies for gasoline engine show much lower CO2-emission potential in WLTP than in NEDCE-SEGMENT: GASOLINE POWERTRAIN SINGLE TECHNOLOGY POTENTIALS1:1.6 l DI-TC gasoline, 4-cylinder, 6-speed manual transmissionSource: FEVExtended Start/Stop 1.1Electric water pump 1.3
18、Split cooling 2.02-step VVL (Intake) 3.6Optimized friction 4.0Enthalpy Storage 4.13-step VVL (Intake) 5.8Cylinder deactivation 10.42-step VCR 10.4Full variable VCR 12.9Miller 13.8Improvementin g/kmDownsized basis1CO2-emissionImprovementin g/km0.21.20.52.43.81.33.76.77.99.913.3Improvementin g/km0.21.
19、20.42.14.01.13.35.87.49.313.9NEDC156 g/kmWLTP TML167 g/kmWLTP TMH190 g/km# SINGLE TECHNOLOGY ANALYSISCooled LP-EGR 1.6 2.6 2.82016-10-28_FEV-Gasoline-delivered-02.pptx 7 by FEV all rights reserved. Confidential no passing on to third parties |RDE Scenario non-reproducable driving on theroadRDE TEST
20、CYCLES & PEMSAachenOEM 1OEM 2OEM 3elipsisRTS 95TimeV eh i cle Sp ee dRDE Boundariessquare6 On-road testing Portable Emission Measurement System (PEMS)square6 Driving in real traffic (non-reproducible driving conditions)square6 RDE-legislation boundary conditions: Testing time 90 120 minutes Max 145
21、km/h (short term 160 km/h) Ca. 1/3 of the total distance: Urban (0-60 km/h) Rural (60-90 km/h) Motorway (90-160 km/h) Extended ambient conditions Post processing of exhaust emissions after PEMS testing2016-10-28_FEV-Gasoline-delivered-02.pptx 8 by FEV all rights reserved. Confidential no passing on
22、to third parties |RDE Scenario engine operation extendsinto entiere mapENGINE MAP OPERATION NEDC, WLTP AND RDE RDE Boundariessquare6 Extended emission technologies required for PN reduction: DI Solenoid 350 bar & Calibration GPF Dual FIS DI + PFIsquare6 Advanced TC technologies required as scavengin
23、g is limited to Exh = 1 and by catalyst temperaturesquare6 In a next step component protection technologies may be required as alternative to enrichment Integrated Exhaust Manifold High Temperature TCs Cooled EGR & Water Injectionsquare6 Promotion of increased displacement (Rightsizing)2016-10-28_FE
24、V-Gasoline-delivered-02.pptx 9Engine Full LoadRDELoadSpeedWLTCNEDC by FEV all rights reserved. Confidential no passing on to third parties |TOPEXPERT Vehicle Test Automatization:VTA RDE online monitor1185%Successful testssquare6 Monitors continuously the compliance to allRDE driving requirements and
25、 comfortably visualizes respective conditionssquare6 Connects to calibration tools like INCA, CANape, ControlDeskor directly to the OBD CANsquare6 Prevents repeated tests and increases probability for test complianceLEGISLATIVE COMPLIANCE CHECK2016-10-28_FEV-Gasoline-delivered-02.pptx# Outlook by FE
26、V all rights reserved. Confidential no passing on to third parties |Changes in Legislation Lead to New Requirements for Longitudinal Dynamics Simulationsquare6 Fixed velocity profilesquare6 Limited and discrete boundary conditions, e.g. conditioned temperature -7 C / 22 Csquare6 Emission measurement
27、 is usually valid barb2righteasy comparison measured value to legislation limitsquare6 Limited engine map areasquare6 10 variants (for different markets)Conventional Driving Cycles Real Driving Emissionssquare6 Free driving / selected route with trafficsquare6 Arbitrary boundary conditions, in very
28、cold/hot conditions: higher limitsquare6 Post processing of emissions with statistic evaluation depending on driving severitysquare6 Operation in the entire engine map possiblesquare6 10 variants ( a2right )ModerateExtended2016-10-28_FEV-Gasoline-delivered-02.pptx 12VTARDE by FEV all rights reserved
29、. Confidential no passing on to third parties |Future emission legislation technology trends for RDE:High fuel pressure, dual fuel injection system, GPFRDE TESTING & PN EMISSIONSsquare6 Development focus of MHI injector development is towards PN reduction: Increased fuel pressure (350 bar +) Improve
30、d nozzle designs Small injection quantity control RDE calibrationsquare6 Injector nozzle coking, fuel sensitivity and cold start prime challenges Compliance with RDE under worst case conditions often border line Very strong OEM focus observed towards GPFsquare6 Additional PN-reduction technologies:
31、Dual FIS Piezo InjectorTime / s0 1000 2000 3000 4000 5000 6000C um p ar t i cl e e mi s si o ns0.01005.010131.010141.510142.010142.510143.01014 RDE - 7 C, 90 % gross vehicle weightSolenoid, multi-hole nozzlewithout first 5 minEU 6c limit atCF factor moderate = 1S pe ed /k m/ h060120GPF2016-10-28_FEV
32、-Gasoline-delivered-02.pptx 13 by FEV all rights reserved. Confidential no passing on to third parties |FEVs RDE simulation process:elipsis check out our presentation in Vienna 2016PT-Technology DefinitionIdentification of RDE Lead CycleEmission ModellingParametric Driver and Road Profilex, y, z, el
33、ipsisHCelipsisPNCO2NOXL oa dSpeedsV eh i cle Sp ee da, b, c, elipsissV eh i cle Sp ee dPNx, y, z, elipsis a, b, c, elipsisagressivedefensiveRDEDoEVehicle simulationEmission Sub-Models TWCTCOOL elipsisBase Maps 90 C, = 1, elipsis 2016-10-28_FEV-Gasoline-delivered-02.pptx 14 by FEV all rights reserved
34、. Confidential no passing on to third parties |FEVcal Design of Experiments (DoE) Model Allows Identification of Most Challenging Combination Regarding PNsquare6 Influence of parameters depends on parameter combination (interdependence) Average distance between two stops becomes more critical for lo
35、wer maximum vehicle speedsquare6 Temperature influence At start: more enrichment, colder combustion chamber walls Also higher PN emissions for warm engine barb2rightintake air influence3101241012510126101271012-505101520253035500 1000 1500 2000 2500 30008101271012510124101231012210126101220304050607
36、0Driver BehaviourPN Emission / (1/km)A mb i en t T em pe r at u re / CAverage Distance Between Two Stops / mA ver a ge Ve hi cle Sp ee d/ (km/ h )PN Emission / (1/km)Defensive Normal Agile2016-10-28_FEV-Gasoline-delivered-02.pptxPN EMISSION DEPENDENCIES18 by FEV all rights reserved. Confidential no
37、passing on to third parties |0.01003.010126.010129.010121.210131.51013100 200 300 400 500Results DoE for RDE indicate a large variance of emissions and CO2 of one powertrain (example gasoline PT)square6 300 simulationresultssquare6 Max. PN result 10 timesabove min., max. CO2 3 timesabove min.square6
38、 Overall trend: CO2 a2upbarb2rightPN a2upsquare6 High variance in PN for identical fuel consumption (CO2)square6 Identification of parameters with high impact on PN results required, e.g. Route parameters Driver behavior Calibration, injection strategiesRDE Scenarios (Simulated)NEDCWLTCP NEm is si o
39、 ns /( 1/ k m)CO2Emissions / (g/km)DoE ModelRDE compl.(+EMROAD)Max. PNEmissionMin. PNEmissionAssumptions and boundary conditions: PN-CF = 1.5 RDE free realistic scenarioVehicle specifications: Inertia weight class: 1.590 kg 2,0 l TCDI gasoline: 180 kW 8-speed ATEU 6d-TEMPLimitSOURCE: FEV, 37TH INTER
40、NATIONAL VIENNA MOTOR SYMPOSIUM 2016# Outlook2016-10-28_FEV-Gasoline-delivered-02.pptx 19 by FEV all rights reserved. Confidential no passing on to third parties |Optimization Strategies Tested Not Only for RDE-LC, but for All Investigated Parameter Combinationssquare6 No beneficial reductions of em
41、issions considered, assumption: worst case Postprocessing (EMROAD, CLEAR) can further reduce emissions or lead to an invalid RDE test Correction factor/higher limit for extended ambient conditions neglectedsquare6 High number of stops especially critical for this vehicle regarding PN Improvement pos
42、sible via Calibration strategy New injection system layoutAPPLICATION OF A CALIBRATION AND NEW INJECTION SYSTEM REDUCES PN EMISSIONS BELOW LIMITRDE leg.Assumption:2016-10-28_FEV-Gasoline-delivered-02.pptx 200.01003.010126.010129.010121.210131.51013100 200 300 400 500NEDCWLTCP NEm is si o ns /( 1/ k
43、m)CO2Emissions / (g/km)Euro 6d-TEMPRDE Scenarios (Baseline)RDE Scenarios (Opt. Cal., Inj. System) by FEV all rights reserved. Confidential no passing on to third parties |Full and Highly Automated Toolbox & Testing Equipment Available at FEVICEModel-basedCalibrationMORPHEEAutomation SystemxMODCo-Sim
44、ulationPlatformFEVALYSPost-Processing ReportingFEVFlex Test Field Information Management CoolConCoolantLubConOilAirConAirFuelConFuelDyno controllerTOM / DCUDynamometer Throttle ActuatorI/O ModulesMIOIndicating SystemOSIRIS / FEVISFuelRate AirRateBlowbyRate IntercoolerEmission AnalyzerFEVER / FTIR201
45、6-10-28_FEV-Gasoline-delivered-02.pptx 21 by FEV all rights reserved. Confidential no passing on to third parties |RDE requires emission testing on the road (PEMS)-but also bringing road, driver & vehicle to test bench2016-10-28_FEV-Gasoline-delivered-02.pptx 22Road trialChassis dynoPowertrain test
46、bedEngine test bed HiL MiLReal EnvironmentVehicleEnvironment prescribed or assumedReal Vehicle SimplifiedVehicle Model: ForcesVehicle ModelRealDriver checkbld Driver ModelReal Trans-mission/PTcheckbld checkbld Transmission/Powertrain & TCU ModelRealEngine checkbld checkbld checkbld EngineModelReal E
47、CU checkbld checkbld checkbld checkbld ECUModelIncreasingmodel depthMORPHEE xMOD by FEV all rights reserved. Confidential no passing on to third parties |Road-to-Rig-to-DesktopLongitudinal acceleration comparison between ViL and real streetDescription-15-10-5051015L on g. ac ce l er at i on /m/ s 02
48、04060801001200 50 100 150 200 250 300 350 400 450V eh i cl e s pe ed / k m/ hTest duration / sViL Bench Real streetsquare6 Real measured vehicle speed profile on a part of the FEV-Eifelrunde with a real/human driver (Smart C453). (black lines)square6 Using the real vehicle speed profile as the desir
49、ed value for the speed controller with the same car incl. a vehicle and driver model on a vehicle-in-the-loop test bench with fully automated driving. (red lines)square6 Good comparability between the desired speed value and the actual speed value of the car on the ViL.square6 Good comparability between the measured longitudinal acceleration on the real street compared to the cal
