Policy making and scenario evaluation tool
for road transport emissions
COPERT - SIBYL workshop
21-22 October 2015
Brussels, Belgium
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Presentation Outline
PART I
Introduction
Development history
SIBYL model
Model structure Methodology Baseline
Indicative results Future work
2
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Presentation Outline
PART II
Software demonstration Graphic User Interface Scenario building options
• Stock
• Activity
Hands-on Examples Scenario building approach Test cases
3
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Presentation Outline
PART III
Software demonstration Scenario building options
• Fuels,
• Energy
• Pollutants
Hands-on Examples Scenario building approach Test cases
4
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
GENERAL FEATURES
5
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL development history
SIBYL 0.9b (Jan ‘12) Beta version containing the GUI and the basic model structure
SIBYL 1.0 (June ‘12) Bottom-up modelling framework
Fuel consumption / emissions based on COPERT 4
Calibrated baseline results based on EC4MACS, PRIMES 2010
Full stock dataset for 26 EU Countries
Built-in, state-of-the-art technologies
Intuitive and flexible custom technology introduction
Statistics-based annual mileage distribution modelling
Interactive chart features
Standard import-export options in various formats
6
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL development history
SIBYL 2.0 (Aug ‘13) Introduction of type-approval CO2 related scenarios
Generation of input file to COPERT 4
Baseline extended to 2050 for all EU-27 MS
Activity consistent with latest PRIMES 2012 REF
New vehicle classes (G < 0.8l and D < 1.4l)
Mileage, speed and share detailed customization: • All three features can now be fully customized with respect to vehicle class, year and age
• Mileage degradation (age dependency)
• Updated mileage, speed and share baseline values
Automatic mileage calibration per vehicle class (user-defined calibration option)
Euro standards implementation
Efficiency trend following imposed type-approval regulations
7
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL development history
SIBYL 3.0 (June ‘14) Almost 10x faster scenario execution
Baseline projection consistent with the most updated estimations in the EU-28 region for the 2010-2050 period, in line with historical data up to 2013.
Baseline consistency with the output of the TRACCS project.
More intuitive graphical user interaction and pre-defined trends
Scenario type options • Full scenario execution
• Stock estimation only – with the option to export data to COPERT
Scenario calibration options with respect to the baseline • Activity balancing per sector
• Fleet balancing per sector
• No balancing
Custom fuel consumption factors introduction capability for all vehicle classes in addition to real-world and type-approval factors.
8
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL development history
SIBYL 4.0 (July ‘15) Baseline projection consistent with the most updated estimations in the EU-28 region plus Switzerland, Norway, Turkey, Iceland and FYROM for the 2010 - 2050 period, in line with historical data up to 2014.
Additional group entities (other than EU28): EU27, EU15, EU12, Non-EU
Updated air pollutant emission factors
Well-to-tank CO2 and Energy emissions are now included; baseline WtT factors are included
Custom baseline input option
Detailed heavy duty trucks and buses classification, in line with the latest COPERT model
GUI enhancements (e.g. Undo/Redo option)
9
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL model analysis
Model structure Modular base Basic module functions
Methodology highlights
Stock estimation Energy & CO2 emissions Advanced technologies Trip modelling
Integrated baseline Historical data sources Projections Calibration
10
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
MODEL STRUCTURE
11
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Vehicle technology coverage
12
Conventional vehicle technologies (COPERT)
Electricity-based vehicle technologies:
• hybrid electric (various)
• range extender
• battery electric
• plug-in hybrid electric
• Fuel cell electric
Biofuels
• Fuel blends (EtOH, Biodiesel)
• Flexi-fuel (E10-E85)
Dual-fuel vehicles
• CNG (various)
• LPG (various) User-specified vehicle
technologies
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL modular structure
13
ScrappageStock
Equilibration
Total Stock
Implementation Matrix
Mileage Distribution Modelling
Activity
Emission Modelling
Constraints
Target
reached?
Energy Consumption
Modelling
Registration
User input:· Stock
· Driving Patterns
· Efficiency
· TtW and WtT factors
· Fuel usage
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL modules
14
ScrappageStock
Equilibration
Total Stock
Implementation Matrix
Registration
Scrappage module: calculates the total scrapped stock based on the base year stock and the survival rates.
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL modules
15
Stock Equilibration module: attempts to balance the stock change.
ScrappageStock
Equilibration
Total Stock
Implementation Matrix
Registration
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL modules
16
Registration module: distributes the new and second-hand registrations.
ScrappageStock
Equilibration
Total Stock
Implementation Matrix
Registration
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL modules
17
Total Stock module: combines all the previously calculated data, to yield the detailed stock.
ScrappageStock
Equilibration
Total Stock
Implementation Matrix
Registration
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL modules
18
Implementation matrix module: produces the Euro standard attribute for the stock.
ScrappageStock
Equilibration
Total Stock
Implementation Matrix
Registration
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL modules
19
Mileage Distribution Modelling Module: processes the mileage-related data and produces the exact mileage per vehicle and age.
Mileage Distribution Modelling
Activity
Stock
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL modules
20
Activity Module: combines stock and mileage data to produce the detailed activity. The activity/stock can be set to match baseline values on a sector basis
Mileage Distribution Modelling
Activity
Stock
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL modules
21
Energy Consumption Modelling Module: combines activity data with consumption functions to produce the detailed energy consumption.
Stock and activity
Emission Modelling
Energy Consumption
Modelling
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL modules
22
Emission Modelling Module: similar to the previous module but uses the emission factors instead.
Stock and activity
Emission Modelling
Energy Consumption
Modelling
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL model sequence
23
Vehicles (class, age)
Module sequence
Results (class, age)
Year Year +1
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
SIBYL model
24
Stock Activity Energy/
Emissions
Time line
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
METHODOLOGY HIGHLIGHTS
25
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Methodology highlights
Stock Estimation
Energy consumption
Activity modelling
26
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Stock estimation
Year Year+1
27
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Stock equilibrium examples
if the stock difference is higher than the baseline, extra second-hand registrations will be added to cover for the required extra stock.
If the stock difference is lower than the baseline, existing second-hand registrations will be zeroed and deregistrations will increase.
28
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Energy consumption in SIBYL
29
TtW
WtT
WtW
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Energy calculation - TtW
30
Base energy consumption
function
Efficiency development
factor
Final energy consumption
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Energy consumption factors - TtW
31
Consumed Energy
Type-Approval factors
Real-world factors
Custom factors
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Energy consumption factors - TtW
Real-world factors
COPERT methodology • All conventional vehicles use COPERT methodology to calculate fuel consumption and
CO2 emissions
Advanced technologies modelling methodology • Advanced (mainly electrified) technologies use a custom consumption/CO2 calculation
method
Type-approval/custom factors
Instead of real-world emission factors, type-approval or even custom factors can be used
32
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Upstream modelling
33
TtW
• find required fuel
WtT
• CO2
• Energy
WtW
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Energy/CO2 factors - WtT
The energy/emissions modules also consides the Well-to-Tank (WtT) energy use and CO2 emissions for each fuel source.
JEC WtT CO2 factors per unit of consumed fuel (energy) are used to calculate the corresponding WtT CO2 emissions and WtT energy [JEC WtT Report 4a, 2014].
These WtT factors (gCO2/MJ or Mj/MJ) are provided per fuel: An array of pathways are available
Custom pathways may be inserted
Time-dependency can be used
34
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Advanced technologies modelling
Target to develop representative emission and consumption factors for complex vehicle technologies
Modelling approach use specific vehicle features to design powertrain system level simulations
calibrate vehicle model performance
extract consumption functions
35
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Simulation process
Vehicle Specifications (physical characteristics, architecture and control)
Model Building
Startup Simulations
(NEDC, UDC, EUDC, acceleration tests)
Model Modifications (calibrate consumption by
tuning of subsystems performance)
Revised Simulations
(Artemis cycles, etc.
Validation with chassis models (e.g. hybrid electric vehicles)
36
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Example: Range-extender EV simulation
AVL Cruise model
37
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Simulation process
38
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Basic Assumptions: Activity
39
Base mileage
Age modifier
Final Annual Mileage
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Activity modelling
40
Activity Split
COPERT Mileage
share method
Trip length modelling
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Trip length distribution
COPERT methodology
Urban/rural/highway share split
Custom methodology: trip length distribution modelling
Trips are characterised by length to model dual-propulsion vehicles with range limitations
Binned mileage allows for customization of specifications for different trips (e.g. EV, flexi-fuel, dual-fuel)
Mileage and trip distribution are inter-connected to simulate mobility scenarios
Baseline trip distribution based on statistics
41
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Trip length distribution
42
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.1 1 10 100 1000
Pro
bab
ilit
y
Travelled distance (km)
0%
5%
10%
15%
20%
25%
30%
35%
40%P
erc
en
tage
of
Trip
s
Trip length class (km)
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Advanced technologies modelling
Battery consumption Fuel Consumption
43
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Trip length distribution modelling
Consumption functions (v, l, iSoC)
Trip length distribution
Integration for each bin Total
consumption
44
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
BASELINE COMPILATION
45
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Baseline building procedure
46
Baseline
Constraints
Projections
Historical data
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Baseline data sources
Baseline building in SIBYL is a process in which a set of historical and projection sources are combined with expert judgement:
Statistical data to setup the detailed starting year Secondary statistical data used to establish the basics for following historical years Assumptions and constraints defined by projection data sources Detailed baseline is corrected and calibrated.
47
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Baseline data requirements
48
Stock
• New regs
• Stock
• Survival rates
• 2- hand regs
Activity
• Annual mileage
• Age dependency
• Trip patterns
Energy/emissions
• Efficiency improvement
• Energy/Emission factors
• EURO standards
• WtT factors
Additional details
• Conversion coefficients
• Other
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Basic Assumptions: Stock
Statistical data sources
Stock structure: TRACCS project • TRACCS includes detailed fleet structure data for the 2005-2010 period for the EU-28
region plus Iceland, Norway, Switzerland, FYROM and Turkey.
• Detailed allocation of fleet into specific vehicles classes per age. Also provides vehicle deregistrations, second-hand registrations and annual mileage data
• Base year: 2010
New registrations: ACEA and ACEM data used to obtain reported new registrations for the 2011-2014 period.
Stock estimation for 2011: EUROSTAT data were used to set total stock trends for conventional passenger cars.
49
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Basic Assumptions: Stock
Data processing
TRACCS • To be inline with COPERT methodology, capacity-based vehicle classification conversion
from market segments is required.
ACEA/ACEM • Disaggregation of new registrations must be performed: figures refer to broad
classification only, e.g. total passenger cars.
EUROSTAT • Expert judgement is required to explain/correct unrealistic trends.
50
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Basic Assumptions: Stock
Stock projections PRIMES 2012 Baseline scenario with adopted measures was used to determine total stock trends up to 2050.
New registrations After 2014, they were set to follow linear trends
Detailed stock structure and development Stock lifetime projections (survival rates) are based on an S-function approximation obtained from the 2010-2011 transition;
Modifications are applied to compensate for stock trend changes afterwards.
51
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Basic Assumptions: Stock
Data processing
Regarding electrified vehicles the projected market vehicle penetration rates of are based on the slightly decarbonised scenario presented by Pasaoglou, Honselaar and Thiel,
PRIMES data include aggregated vehicle classification based on fuels. Further disaggregation required for SIBYL is based on expert judgement.
PRIMES estimations were deemed very optimistic for the 2010-2013 period; reasonable stock trend deviations estimations were allowed
52
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Basic Assumptions: Activity
Base mileage and age effect figures based on TRACCS data
53
Base mileage
Age modifier
Final Annual Mileage
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Basic Assumptions: Activity
Data processing
Activity development follows PRIMES 2012 relative activity/energy development per fuel and sector
Baseline calibration requires that base mileage values are modified to satisfy activity match with PRIMES.
Up to 2013, the baseline mileage for gasoline/diesel vehicles was modified to meet the statistical consumed fuel.
54
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Basic Assumptions: Energy & Efficiency
Efficiency development CO2 type-approval emission figures were used for passenger car vehicle technologies to set 2010 efficiency base
Imposed 2015 and 2020 CO2 targets for PCs used to extract efficiency trends up to 2020. No efficiency development was assumed post-2020.
LCVs use the corresponding targets to set their own efficiency pattern
Mopeds, Motorcycles, HDVs and HDBs use an efficiency development rate defined by expert judgement.
55
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
INDICATIVE RESULTS
56
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Accelerated scrappage scheme
• Accelerated scrappage scheme for PCs in Austria in the period 2017-2019 while activity (vkm) remains constant
• Total energy consumption by passenger cars:
57
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Impact of super-credits
• Impact of super-credits on conventional car CO2,TA emissions in the case of Germany – Two penetration scenarios for EVs
58
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Average refinery fuel barrel
Indicative result
59
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
FUTURE WORK
60
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Development plan for SIBYL 5.0 (2015)
Model fully compatible with the upcoming COPERT 5
Scenario building wizard (step-by-step walkthrough and examples e.g. technology replacement)
Update based on the latest PRIMES scenario
Customized interfacing to accommodate each user’s needs e.g. simplified edit options
61
COPERT – SIBYL Workshop, 21-22 October 2015, Brussels, Belgium
Thank you for your attention…
Any questions…?
62
Dr. Petros Katsis [email protected]