A CO2 based indicator for severe driving? (Preliminary investigations - For discussion only)

A CO2 based indicator for severe driving? (Preliminary investigations - For discussion only)

A case study Diesel vehicle tested with PEMS on the JRC test routes : Route 1: Rural-Motorway Route 2: City -Rural Route 3: Mountain-Rural Based on their average characteristics (usual speed profiles, road grade, ), the Routes 1 & 2 are considered as normal whereas the mountain part of Route 3 is considered as very severe

Steps of the data analysis Emissions @ cold start and during DPF regeneration events are calculated separately Moving averaging window calculations, based on the vehicle CO2 emissions [kg] over the NEDC, calculating window emissions in [g/km]

Vehicle CO2 emissions on NEDC CO2 (g/km) Average speed (km/h) ECE 15 200 19 EUDC 120 63 NEDC 160 33

Tentative definition of normal driving through window CO2 [g/km] CO2 [g/km] Application of the principles to the case study: - 2 speed classes low and high speed - CO2 central values based on ECE15 and EUDC values - Upper and lower bounds at ±20% from the central value in each speed class Low Speed High Speed Average Speed

Data analysis strategy Marking cold start & DPF Regen 160.00 140.00 Resulting data for MAW analysis 2.00 1.50 120.00 100.00 1.00 0.50 80.00 60.00 Cold Start DPF Regen 0.00-0.50-1.00 Vehicle Speed [km/h] MAW ANALYSIS YES [1] / NO [0] DPF REGEN YES / NO COLD START YES / NO 40.00-1.50 20.00-2.00 0.00-2.50 0 1000 2000 3000 4000 5000 6000 7000 8000

Window CO2 Emissions [g/km] -- Vehicle Speed [km/h] Step 2: Moving averaging window calculations 250 200 150 100 Vehicle Speed [km/h] MAW CO2 [g/km] MAW NOx [g/km] 50 0 0 1000 2000 3000 4000 5000 6000 7000 8000

CO2 Step 3: Identification of severe driving using averaging window CO2 [g/km] 600 500 Uphill mountain driving (Severe to very severe!) 400 300 Urban driving area CO2 centered on ECE15 value Upper and lower CO2 at ±20% Extra-urban driving area CO2 centered on EUDC value Upper and lower CO2 at ±20% 200 100 0 0 20 40 60 80 100 120 140 160 Window Average Speed [km/h] Downhill mountain driving (Soft to very soft!)

Distribution of averaging window results NOx, All windows Route 1&2 Route 3 3500 1200 3000 1000 2500 2000 1500 ROUTE1 - RURAL/MOTORWA Y ROUTE2 - CITY/RURAL 800 600 ROUTE3 1000 400 500 200 0 0

Distribution of averaging window results CO2, All windows Route 1&2 Route 3 Normal CO2, normal idling 2500 100 1200 100 90 90 2000 1500 80 70 60 50 ROUTE 1 / RURAL- MOTORWAY ROUTE 2 / CITY- RURAL 1000 800 600 80 70 60 50 ROUTE 3 / RURAL- MOUNTAIN 1000 500 40 30 20 10 ROUTE 2 / CITY- RURAL (Cumul.) ROUTE 1 / RURAL- MOTORWAY (Cumul.) 400 200 40 30 20 10 ROUTE 3 / RURAL- MOUNTAIN (Cumul.) 0 0 0 200 400 600 0 0 0 200 400 600

Resulting NOx Emissions ALL CO2 CO2 ID10 R1-Test1 R1-Test2 R2-Test1 R2-Test2 R3-Test1 R3-Test2

General considerations

Boundary conditions & RDE procedures BC elements Data flagging > categorizing the data, depending on the vehicle, aftertreatment and driving situation PEMS tests: Defining normal driving through a composite CO2 indicator

BC elements (1) Ambient temperature range JRC proposal: alignment with 582/2011 requirements (roughly from -7 C to + 30 C) Vehicle conditioning Basis for the definition of cold / hot vehicle Based on engine condition (JRC proposal: based on engine coolant temperature, with time and/or temperature criteria) Needs to be complemented by after-treatment condition, at least for the DPF

BC elements (2) Driving conditions Speed Acceleration (based on RPA or better) Idling Road grade Vehicle payload Difficulty: to define appropriate indicators and ranges for some of the above elements: Defining normal driving through a composite (CO2?) indicator?

Composite driving severity indicator? For RCG testing, the driving severity is dictated only by the acceleration characteristics of the cycle (currently through RPA) For PEMS testing, the driving severity is dictated by a series of elements (acceleration, road grade, vehicle payload, head wind, driver, traffic, ) occurring simultaneously or not Difficulty for the PEMS data evaluation: To develop a set of appropriate indicators to define severe windows; To understand which elements contribute to the severity and to what extent.

CO2 emissions as driving severity indicator? Simple physics: Fuel consumption and therefore CO2 are influenced by: Speed Acceleration (based on RPA or better) Idling Road grade Vehicle payload High distance-specific CO2 emissions (fuel consumption ) should be a good indicator of severe vehicle use.

Data flagging strategy Possible status for the different elements Element Possible status A: Vehicle condition 1: Hot 2: Cold B: After treatment 1. DPF Functional 2. Regenerating C: Driving 1. Normal 2. Severe (*Based on a composite indicator, refer to the following slides) Comments: Cold vehicle (A2) and DPF regeneration (B2) (normally) do not occur simultaneously DPF Regeneration (B2) not always present Hot normal driving (A1 + C1) represents the largest part of a test (time or distance wise)

Data flagging strategy Mark the following test sections and calculate the emissions: Cold vehicle (A2) After-treatment (B2) Using the remaining unmarked data, perform the moving averaging window calculations: for each window, the composite (CO2) severity indicator is calculated to determine whether the windows belong to normal (C1) or severe driving (C2)

CO2 [g/km] Vehicle CO2 @ type approval Option 1 Defining normal driving using: - Speed classes, using the vehicle CO2 emissions [g/km] @ type approval, in each speed class - Intervals in each speed class - Lower and upper bounds in for the normal CO2 range in each speed class Very-Low Speed (PEMS only) Low Speed Upper bound MS CO2 for the medium speed range Lower bound MS Medium Speed High Speed Extra-high Speed Average Speed

CO2 [g/km] Vehicle CO2 @ type approval Option 2 Defining normal driving using: - Speed classes, using the vehicle CO2 emissions [g/km] @ type approval, in each speed class - Vehicle curve fitting the type approval values - Lower and upper bounds not necessarily at constant distance (only for the illustration). Could also be with ± x% from the vehicle CO2 curve Low Speed Medium Speed High Speed Extra-high Speed Upper bound Vehicle CO2 curve Lower bound Average Speed

Further remarks (1) Objective setting: The different data categories (cold start, DPF regeneration, normal and severe driving) can be given different NTE limits and/or weighing factors. Link between the high CO2 emissions and the characteristics of the corresponding driving situations to be documented in the development phase of the PEMS procedure

Further remarks (2) Requirements for the trip composition are needed, expressed in % of the different speed classes A sufficient number of windows should be needed in the different speed classes A minimum percentage of windows could be used to define a valid test (i.e. valid route and driving characteristics)

Further remarks (3) Idling indicator required to complement the CO2 indicator Averaging scale shall remain in the order of the averaging scale used for the type approval cycle or cycle phases. Time based windows could be investigated Hybrid vehicles? Representative CO2 values? Need to define a CO2 equivalent for the driving on electricity?