Comparison of VECTO and GEM
Dr. Felipe Rodríguez22nd January 2018
G20 Transport Task Group:Deep Dive to Support Heavy-Duty Vehicle Efficiency Labeling and Standards Meeting #2
2
”Heavy-duty vehicle fuel-efficiency simulation: A comparison of US and EU tools” (2015)
Franco, V., Delgado, O., & Muncrief, R. (2015). Heavy-duty vehicle fuel-efficiency simulation: A comparison of US and EU tools. ICCT. http://www.theicct.org/heavy-duty-vehicle-fuel-efficiency-simulation-comparison-us-and-eu-tools
§ ICCT conducted a comparison of GEM 2.0 (Phase 1) and VECTO (2.0.3) in 2015
§ Since then both tools have been updated to GEM 3.0 (Phase 2) and VECTO (3.2.1), resulting in changes to the required model inputs and the resulting fuel consumption simulation.
§ Nevertheless, ICCT’s 2015 report is a good reference document for understanding:§ Forward vs. backward models§ VECTO’s shifting strategy§ VECTO’s look ahead coasting§ VECTO’s file structure
www.theicct.org
APRIL 2015WHITE PAPER
B E R L I N | B R U SS E L S | SA N F R A N C I S CO | WAS H I N GTO N
HEAVY-DUTY VEHICLE FUEL-EFFICIENCY SIMULATION: A COMPARISON OF US AND EU TOOLS
Vicente Franco, Oscar Delgado and Rachel Muncrief
3
”The Present of EU’s Vehicle Energy Consumption Calculation Tool (VECTO), and Recommendations for the Future” (2018)
Rodríguez, F. (2018). The Present of EU’s Vehicle Energy Consumption Calculation Tool (VECTO), and Recommendations for the Future (Publication pending). International Council on Clean Transportation.
§ A new comparison study of the latest releases of GEM and VECTO
§ Although focused on VECTO, it describes the model architectures of both GEM and VECTO
§ The results of this new study are the focus of this presentation
§ Publication expected in February 2018
4
Input comparison between GEM and VECTO
Component VECTO input GEM input
Engine
Displacement, idle speed, fuel consumption map, full load torque curve, motoring friction curve, brake-specific fuel consumption over the Worldwide Harmonized Transient Cycle (WHTC)
Displacement, idle speed, fuel consumption map, full load torque curve, motoring friction curve, fuel consumption over the ARB Transient Drive Cycle for 9 different vehicle configurations
TransmissionTransmission type, gear ratios, torque loss map as a function of torque and speed for each gear, maximum torque and speed per gear
Transmission type, gear ratios, and maximum torque per gear. Optional: Power loss map as a function of torque and speed for each gear
Axle Axle ratio and torque loss map as a function of torque and speed
Axle ratioOptional: Power loss map as a function of torque and speed
Aerodynamic drag
Air drag area as determined during the constant speed procedure. For rigid trucks, a standard box is used. For tractors, a standard trailer is used.
Air drag area as determined by the coastdown methodology. Standard trailers are used for tractor modeling.
Tires Tire dimensions, rolling resistance coefficient (Crr), and load applied during the rolling resistance test for each axle
Rolling resistance coefficient (Crr) for each axle, and drive tire revolutions per mile
Vehicle Curb vehicle weight, gross vehicle weight rating, and axle configuration
Vehicle weight reduction (sum of standardized weight reductions per component), vehicle regulatory subcategory (e.g., Class 8, sleeper cabin, high roof), and axle configuration
Other
Auxiliaries: Technology used for the following auxiliaries: cooling fan, steering system, electric system, pneumatic system, A/C system (whether it is present or not), and power take-off
Off-cycle technologies: Improvements through the application of the following technologies: Speed-limiter, neutral-idle, intelligent controls, accessory load reduction, extended idle reduction, tire pressure system, and other technologies.
5
GEM’s model architecture
Vehiclespeed
Target speed
100
200
300
400500
600700800
9001000
11001200
1300
1400
1500
Input
torq
ue/Nm
0
1x104
2X104
3X104
4x104
5x104
Input speed / 1/min0 500 1000 1500 2000
Torque loss / Nm
Powertrain
Trans.
Driveline
Vehicle
Driver
Drivingcycle
Accelerator pedal position andfeedback from the powertrain
Torqueloss maps
Fuel consumptionmap
Vehicle weightAir drag area
Rolling resistancecoefficients
Road grade andambient temperature
Force atthe wheel
Vehiclespeed
Advance indistance and time
New vehicle positionand speed
30.032.0
38.0
40.0
41.0
44.0
44.044.5
44.8
Torq
ue/Nm
0
500
1000
1500
2000
2500
Engine speed / rpm600 1000 1400 1800 2200
Brake thermal efficiency / %
FRR
Ambient
Engine
§ GEM does not feature a graphical user interface.
§ GEM was developed in Matlab Simulink as a forward-looking model: The simulation runs from the accelerator pedal to the wheels.
§ The GEM architecture is comprised of four main modules: Powertrain, Vehicle, Driver, and Ambient.
§ The Driver module is a closed-loop controller
US and Canada HDV fuel consumption certification
6
Engine dyno mapping
Coastdown test
GEM - Engine cycle generation
GEM
Engine testing:Cycle averaged map
Declared CO2 and FC value
Standard vehicle specifications
Powertrain testing option
Tires test
Transmission and axle test
Wind average correction
Off cycle technologies
Transient correction
Powertrain correction
Input data Dyno testing Correction SimulationOutput data
7
VECTO’s model architecture
30.032.0
38.0
40.0
41.0
44.0
44.044.5
44.8
Torq
ue/Nm
0
500
1000
1500
2000
2500
Engine speed / rpm600 1000 1400 1800 2200
Brake thermal efficiency / %
7700
70
8800
80
9900
90
6600
60
5500
50
4400
30
Inputto
rque/Nm
-6000
-4000
-2000
0
2000
4000
6000
Input speed / 1/min0
i=n Torque loss / Nm
7700
70
8800
80
9900
90
6600
60
5500
50
4400
30
Inputto
rque/Nm
-6000
-4000
-2000
0
2000
4000
6000
Input speed / 1/min0
i=10 Torque loss / Nm
7700
70
8800
80
9900
90
6600
60
5500
50
4400
30
Inputto
rque/Nm
-6000
-4000
-2000
0
2000
4000
6000
Input speed / 1/min0
i=11 Torque loss / Nm
7700
70
8800
80
9900
90
6600
60
5500
50
4400
30
Inputto
rque/Nm
-6000
-4000
-2000
0
2000
4000
6000
Input speed / 1/min0
i=12 Torque loss / Nm
100
200
300
400500
600700800
9001000
11001200
1300
1400
1500
Inputto
rque/Nm
0
1x104
2X104
3X104
4x104
5x104
Input speed / 1/min0 500 1000 1500 2000
Torque loss / Nm
Driver
Vehicle
Wheels
Axle
Transmission
EngineAuxiliaries Success?
Shiftingstrategy
Forcerequest
Accelerationrequest
Torquerequest
Torquerequest
Torquerequest
Fuelconsumption
Torque
request
Torque
loss
Torque
loss
Rolling
resistancecoeffi
cients
Vehicle
weig
htAirdrag
area
Driving
cycle
Write simulationresults and advance
to next point
Yes
No
Reduce acceleration requestor operate the brakes
FRR
§ VECTO was developed in C# as a backward-looking model: the simulation flow occurs in the opposite direction to the way it takes place in the actual vehicle.
§ The Driver Model converts the drive cycle information into an acceleration request, to ultimately locate an appropriate operating point in the engine fuel map
§ Once a valid engine operating point is found, the simulation moves to the next point in the driving cycle.
8
Europe HDV fuel consumption certification
Engine dyno mapping
Constant speed air drag test
WHTC test
VECTO
Transient correction
Declared CO2 and FC value
Tires test
Transmission and axle test
Auxiliaries
Cross wind and speed profile
correction
Vertical load correction
Input data Dyno testing Correction SimulationOutput data
9
GEM-VECTO comparison: Definition of base vehicle
Component Parameter 4x2 tractor-trailer 6x2 tractor-trailer
Engine
Displacement 11.0 liters 15.0 liters
Idle speed 650 rpm 600 rpm
Power 262 kW @ 1715 rpm 340 kW @ 1726 rpm
Transient correction None None
Transmission and axleTransmission type 10 speed, AMT 10 speed, AMT
Axle ratio Between 3:1 and 4:1 Between 3:1 and 4:1
Aerodynamic dragCdA Between 4 and 6 m2 Between 4 and 6 m2
Cross-wind correction None None
Tires
Crr steering axle Between 4 and 7 N/kN Between 4 and 7 N/kN
Crr drive/tandem axles Between 4 and 7 N/kN Between 4 and 7 N/kN
Crr trailer axles 6 N/kN 6 N/kN
Tire dynamic radius 512 mm 512 mm
AccessoriesAccessory power 3500 W (constant) 3500 W (constant)
Acc. load reduction None None
Vehicle
Base vehicle mass 9570 kg 14741 kgBase payload 11340 kg 17237 kgVehicle weight reduction Up to 2000 kg Up to 3000 kgTotal number of axles 4 5
10
GEM-VECTO comparison: Powertrains
600
900
1200
1500
1800 210
0240060
0900
1200
1500
1800 210
02400
5 10
1520
25
30
35
40
4550
Torq
ue/Nm
-300
0
300
600
900
1200
1500
1800
Engine speed / rpm
11 L, 262 kW engine.Fuel consumption in kg/h
10
20
30
35
40
45
5060
55
55
5
15
1525
25
Torq
ue/Nm
-500
0
500
1000
1500
2000
2500
Engine speed / rpm
15 L, 340 kW engine.Fuel consumption in kg/h
50
5500
100
110000
150
115500
200
220000Inpu
tto
rque
/10
00
Nm
-6
-4
-2
0
2
4
6
Input speed / rpm
0500
1000 150
02000
2500
3000
Gears 1, 2, 3, 4, and 5Torque loss / Nm
2200
40
4400
60
6600
80
8800
100
110000
120
112200
Inpu
tto
rque
/10
00
Nm
-6
-4
-2
0
2
4
6
Input speed / rpm
0500
1000 150
02000
2500
3000
Gears 6, 7, 8, and 10Torque loss / Nm
2244
21
1188
15
1122
9
Inpu
tto
rque
/10
00
Nm
-6
-4
-2
0
2
4
6
Input speed / rpm
0500
1000 150
02000
2500
3000
Gear 9 (direct drive)Torque loss / Nm
1000
900
800
700
600
500
400
300
100
200Inpu
tto
rque
/10
00
Nm
0
10
20
30
40
50
Input speed / rpm
0500
1000 15
002000
2500
Axle ratio = 4Torque loss / Nm
700
600
500
400
300
100
200
Inpu
tto
rque
/10
00
Nm
0
10
20
30
40
50
Input speed / rpm
0500
1000 15
002000
Axle ratio = 3Torque loss / Nm
Engine: ~ 46% peak efficiency
Axle: ~ 98% mechanical efficiency
Transmission: ~95-99% mech. efficiency
11
GEM-VECTO comparison: Randomized vehicle generation
4.0 4.5 5.0 5.5 6.0
Freq
uenc
y/%
0
5
10
15
20
Axle ratio / -3.0
03.253.503.7
54.0
0
CdA / m2 Crr / N/kN4.0
04.7
55.5
06.2
57.0
0
SteerDrive/Tandem
Weight reduction / kg
0750150
022503000
A total of 500 unique vehicle configurations were randomly generated for each one of the two vehicle types. As an example, the distribution of the values for the 500 different 6x2 tractor-trailers simulated are below
12
The US GHG Phase 2 cycles were used for the comparison
55 mph and 65 mph cycles
Time / s
0100
200
300
400
500
600
700
ARB Transient cycle
Spee
d/km
/h
0
30
60
90
Grade
/%
-7-5-3-11357
Distance / km
0 5 10 15 20
Spee
d/km
/h60
90
120
§ The regulatory cycles for the US GHG Phase 2 standard were used, since they cannot be changed in GEM.
§ The 55 mph and 65 mph cycles are distance-based cycles with grade.
§ The ARB Transient cycle is a time-based cycle without grade
13
Comparison results: Constant speed cycles with grade
22 24 26 28 30 32 34VECTO fuel consumption / gfuel / tonne-km
22
24
26
28
30
32
34
GEM
fuel
cons
umption
/g
fuel/to
nne-
km
6x2, sleeper cab, tractor-trailerGEM = 1.009 * VECTO, R2=0.920
4x2, day cab, tractor-trailerGEM = 1.030 * VECTO, R2=0.964
6 7 8 9 10 11VECTO engine work / kWh
6
7
8
9
10
11
GEM
enginewor
k/kW
h
4x2, day cab, tractor-trailerGEM = 1.008 * VECTO, R2=0.860
6x2, sleeper cab, tractor-trailerGEM = 1.000 * VECTO, R2=0.952
ARB Transient cycle
55mph and 65 mph cycles with grade
10 12 14 16 18 20 22 24VECTO fuel consumption / gfuel / tonne-km
10
12
14
16
18
20
22
24
GEM
fuel
cons
umption
/g
fuel/to
nne-
km
6x2, sleeper cab, tractor-trailerGEM = 1.007 * VECTO, R2=1.000
4x2, day cab, tractor-trailerGEM = 0.998 * VECTO, R2=0.996
17 19 21 23 25 27 29 31 33 35VECTO engine work / kWh
17
19
21
23
25
27
29
31
33
35
GEM
enginewor
k/kW
h
6x2, sleeper cab, tractor-trailerGEM = 1.010 * VECTO, R2=1.000
4x2, day cab, tractor-trailerGEM = 1.008 * VECTO, R2=1.000
§ The engine work is useful to gauge the agreement in energy flows observed by the engine.
§ The fuel consumption is useful to assess the impact of the shifting strategies. For a given engine work, the shifting strategy determines the regions of the engine map.
Absolute error: 0.64%
14
Comparison results: Transient cycle
22 24 26 28 30 32 34VECTO fuel consumption / gfuel / tonne-km
22
24
26
28
30
32
34
GEM
fuel
cons
umption
/g
fuel/to
nne-
km
6x2, sleeper cab, tractor-trailerGEM = 1.009 * VECTO, R2=0.920
4x2, day cab, tractor-trailerGEM = 1.030 * VECTO, R2=0.964
6 7 8 9 10 11VECTO engine work / kWh
6
7
8
9
10
11
GEM
enginewor
k/kW
h
4x2, day cab, tractor-trailerGEM = 1.008 * VECTO, R2=0.860
6x2, sleeper cab, tractor-trailerGEM = 1.000 * VECTO, R2=0.952
ARB Transient cycle
55mph and 65 mph cycles with grade
10 12 14 16 18 20 22 24VECTO fuel consumption / gfuel / tonne-km
10
12
14
16
18
20
22
24
GEM
fuel
cons
umption
/g
fuel/to
nne-
km
6x2, sleeper cab, tractor-trailerGEM = 1.007 * VECTO, R2=1.000
4x2, day cab, tractor-trailerGEM = 0.998 * VECTO, R2=0.996
17 19 21 23 25 27 29 31 33 35VECTO engine work / kWh
17
19
21
23
25
27
29
31
33
35
GEM
enginewor
k/kW
h
6x2, sleeper cab, tractor-trailerGEM = 1.010 * VECTO, R2=1.000
4x2, day cab, tractor-trailerGEM = 1.008 * VECTO, R2=1.000
§ Despite the differences in model architecture (forward vs backward-looking), driver model, and shifting strategy; both VECTO and GEM produce similar results in terms of engine work and fuel consumption.
Absolute error: 2.03%
15
§ VECTO and GEM show very good agreement when simulated over a large set of identical vehicles
§ The accurate simulation of CO2 emissions of HDVs is more dependent on the component input data than on the selected model (VECTO vs GEM). Harmonization of component certification benefits the implementation of future regulatory measures.
§ Both GEM and VECTO can be adapted to account for the differences across regions. VECTO’s engineering mode provides a user friendly interface to modify drive cycles, payloads, and vehicle details. GEM can also be modified accessing the source code, however, this implies more effort.
Takeaway messages
Top Related