RDE testing how to define NTE emission limits

RDE testing: how to define NTE emission limits? Nikolaus STEININGER nikolaus. steininger@ec. europa. eu Automotive Industries European Commission, DG GROWTH Brussels, 26 June 2015

Real Driving Emission (RDE) Testing of LDVs 1. Why? (short refresh) 2. How? (short refresh) 3. Where do we stand (1 st & 2 nd RDE package)? [Next steps: - RDE testing for PN: 3 rd RDE package) - In-service-conformity tests: 4 th RDE package)]

RDE testing: why? Ø Massive air-quality problems: NO 2 concentrations exceed regulatory limits mainly in urban hotspots Ø Main culprits: very high NOx emissions of LDVs exceeding regulatory emission limits by a factor 5+… Ø RDE emissions of Euro 3 – 5 vehicles more or less constant (demonstrated by many studies), despite nominal regulatory emission limits decreasing from 500 to 180 mg/km Ø Euro 6 vehicles peform slightly better but emit NOx still far above their emission limit of 80 mg/km. Huge differences between the performance of individual Euro 6 vehicles (ranging from "complying with reg. Euro 6 limits" to "not better than the worse Euro 5 vehicles" Ø Lab testing on a fixed test cycle is not sufficient for regulating NOx emissions!

RDE testing: why? Ø Euro 5/6 Regulation (EC) 715/2007 defines emission limits under "normal conditions of use", i. e. not in relation to a particular test cycle! => current test procedures fall short of legal requirements! Ø "Normal conditions of use" not explicitly defined in legislation => "normal is what actually occurs on the road"… Ø “Normal conditions of use”: collectivity of all driving situations an "emission vehicle type" encounters on the road weighted according to their statistical occurrence => regulatory emission limits = average emissions on the road : = "emission factors" NB: average emissions on the road ≠ emissions at "average driving"

RDE testing: how? Covering all “normal conditions of use" in one test impossible => Ø Randomized test conditions potentially covering “all driving situations on the road" Ø Normalisation: achieve good statistical correlation of "individual PEMS trip" "normal conditions of use" Ø For all possible RDE tests performed according to legal requirements: Emissions < Not-To-Exceed (NTE) limits Ø NTE emission limit = (regulatory emission limit) * (compliance factor) => CF > 1 accounts for variations of RDE test with respect to "normal conditions of use ", "benefit of doubt" for manufacturer NB: In general NTE emission limit ≠ Emission factor

RDE testing: how? Ø Testing with on-board "portable emission measurement systems (PEMS)" Ø Randomized test cycles (dynamics & boundary conditions) Ø Simulation methods (? ? ? ) •

RDE: where do we stand? Ø Starting in 2011 a PEMS test procedure for RDE testing has been developed Ø "Normalisation" methods for dynamcial driving conditions (velocity, accelerations): Power binning method (CLEAR tool) Moving averaging windows method (EMROAD) Ø Set of ambient boundary and trip conditions Ø Technical procedures for performing the test Ø "PEMS test family" concept to limit amount of PEMS testing at initial type approval Ø For the moment emissions during cold start and after-treatment system regenerations are excluded from RDE assessment => to be addressed by special procedure in the future => RDE test procedure for initial type approval voted by TCMV on 19 May 2015

RDE: where do we stand? Ø As from January 2016: monitoring of RDE testing for NOx (no mandatory requirements yet) Ø As from September 2017/18 (new emission types/all new vehicles) : 1 st step of NTE limits (NOx + PN) Ø As from … : 2 nd step of NTE limits (NOx)

RDE: where do we stand? Ø Now: preparation of 2 nd RDE package, to be adopted in 2015 Ø “Complementary boundary conditions” addressing “biased driving” for a whole PEMS trip: - Too dynamic/aggressive: limit 95%tile of v*a - Too smooth: RPA = ∫v • a+dt /D > minimum value Ø NTE emission limits for NOx to be defined…

RDE: where do we stand? Mandatory NTE emission limits for NOx: Ø Values for the 2 nd step to be defined via a strict interpretation of the existing regulatory Euro 6 emission limits (full redesign of vehicles possible) Ø Values for the 1 st step (alternatives): - account for “the better end” of existing Euro 6 “standard” technologies (vehicles with NOx after-treatment) + “reasonably achievable” improvements (? ? ? ) OR - "political" decision on the basis of 2 nd step values + existing situation?

RDE: where do we stand? Ø Regulatory decision on 2 nd step NTE emission limit not to be performed after the 1 st step! Work for the two steps can be performed in parallel. Ø Time schedule: - Various technical elements needed: to be discussed today - Possibly two task forces - Presentation of a Commission service proposal for first discussion at TCMV in September - Intended vote at TCMV still in 2015

RDE: where do we stand? Ø 2 nd step NTE emission limit, principles: - Uncertainties of measurement equipment, e. g. mass flow, pollutant concentrations, vehicle speed - Error propagation - Manufacturers' vs. regulator's risk Ø 2 nd step NTE emission limit, process: - Work by the JRC in progress - To be shared with stakeholders asap - Possibly task force?

RDE: where do we stand? Ø 1 st step NTE emission limit, technology route (brainstorming, approach not agreed yet by COM): - Consider "mainstream" Euro 6 vehicles, i. e. at least equipped with some NOx after-treatment - "Reasonable" short-term improvements, e. g. software recalibration - Different vehicles + improvements => Distribution of NOx emission performances - NTE emission limits to be decided upon short-term air quality requirements + "acceptable" share of existing diesel vehicle models to be replaced on the market within 2 – 3 years

RDE: where do we stand? Ø 1 st step NTE emission limit, technology route, information needed: - Definition of "mainstream Euro 6 vehicle' (industry stakeholders, Euro 6 impact assessment? ) - NOx emission performance distribution of existing "mainstream" Euro 6 vehicles => use PEMS data (? ) - Quantitative short-term improvements, e. g. via software calibration changes (industry stakeholders? ) Ø Time frame needed for developing this technology route …

Thank you for your attention!