1. Introduction limit reductions: energy, environmental and safety - - PDF document

▶
1 introduction
SMART_READER_LITE
LIVE PREVIEW

1. Introduction limit reductions: energy, environmental and safety - - PDF document

Aug 19, 2022 250 likes •342 views

Using naturalistic driving data to evaluate speed 1. Introduction limit reductions: energy, environmental and safety Motivation assessment Significant impacts of transport sector: o Energy consumption ( security of supply) o GHG

slide-1
SLIDE 1

31st ICTCT conference 25 - 26 October, 2018 Porto, Portugal

Using naturalistic driving data to evaluate speed limit reductions: energy, environmental and safety assessment

Patrícia Baptista, Marta Faria, Gonçalo Duarte

  • 1. Introduction

Motivation

2

  • Significant impacts of transport sector:
  • Energy consumption (→ security of supply)
  • GHG emissions (→ global warming)
  • Local pollutants (→ air quality, health)
  • Accidents (→ injuries, fatalities)
  • Etc.

CO2 emissions

  • 1. Introduction

Motivation

3

  • Significant impacts of transport sector:
  • Energy consumption (→ security of supply)
  • GHG emissions (→ global warming)
  • Local pollutants (→ air quality, health)
  • Accidents (→ injuries, fatalities)
  • Etc.
  • Alternative options on urban mobility:
  • Cross-modal electrification
  • Transport system integration coupled with Mobility-as-a-Service (MaaS) to

promote modal shift

  • Redesign of transport infrastructure (urban plazas, reduction of speed limits, etc.)
  • Etc.
  • 1. Introduction

Motivation

4

  • Alternative options on urban mobility:
  • Redesign of infrastructure → reduction of speed limits
slide-2
SLIDE 2
  • 1. Introduction

Motivation

5

  • Alternative options on urban mobility:
  • Redesign of infrastructure → reduction of speed limits

Reduce accident risk Improve the urban environment for pedestrians and biker Promotion of MaaS products (bike, scooter-sharing, etc.)

  • Alternative options on urban mobility:
  • Redesign of infrastructure → reduction of speed limits

Reduce accident risk Improve the urban environment for pedestrians and bikers Promotion of MaaS products (bike, scooter-sharing, etc.)

  • 1. Introduction

Motivation

6

  • 1. Introduction

Objective

7

  • Assess the impacts of reducing speed limits on a one-vehicle/driver

perspective:

  • Using real world driving data (naturalistic driving data)
  • Considering the type of road where the vehicle drives
  • Case study: Lisbon Metropolitan area
  • 2. Data and Methods

2.1. Data collection

8

  • 12 drivers under real world driving conditions
  • Data collection in Lisbon metropolitan area
  • 1 month of data (September 2016), corresponding to 16412 km and 441 hours of

driving

  • Vehicles’ characteristics:

Driver number GENRE AGE_GROUP EXPERIENCE FUEL_TYPE DISPLACEMENT CAR_DATE 1 Female 35-49 years 10-24 years Diesel 1598 2011 2 Male 25-34 years 10-24 years Diesel 1598 2013 3 Male 35-49 years 25-49 years Diesel 1461 2008 4 Female 25-34 years 10-24 years Diesel 1461 2008 5 Female 35-49 years 25-49 years Diesel 1560 2010 6 Female 50-64 years 25-49 years Diesel 1995 2013 7 Female 35-49 years 10-24 years Gasoline 1242 2006 8 Female 35-49 years 10-24 years Gasoline 1198 2006 9 Female 35-49 years 25-49 years Diesel 1493 2005 10 Male 35-49 years 25-49 years Diesel 1598 2012 11 Male 25-34 years 10-24 years Diesel 1968 2008 12 Male 35-49 years 10-24 years Diesel 1991 2007

slide-3
SLIDE 3
  • 2. Data and Methods

2.1. Data collection

9

  • 2. Data and Methods

2.2. Quantification of energy consumption

10

Analysis of 1 Hz driving data to assess:

  • Vehicle Specific Power – methodology to correlate vehicle dynamics

with fuel use, pollutant emissions

VSP = d dt EKinetic + EPotential + FRolling ∙ v + FAerod ynamic ∙ v m =

= ∙ ∙ 1 + !" + # ∙ $%&' + ()*++ , + (-%.* ∙ 3

VSP accounts for driver aggressiveness through speed and acceleration

  • 2. Data and Methods

2.2. Quantification of energy consumption

11 Measured drive cycle Per vehicle

  • 2. Data and Methods

2.2. Quantification of energy consumption

12 VSP time distribution Fuel consumption per VSP mode Measured drive cycle

7.6 l/100km

Per vehicle

slide-4
SLIDE 4
  • 2. Data and Methods

2.2. Quantification of energy consumption

13 VSP time distribution Fuel consumption per VSP mode Measured drive cycle Another drive cycle

7.6 l/100km

VSP time distribution

6.9 l/100km

Per vehicle

  • 2. Data and Methods

2.3. Model development – drive cycle adjustment

14 Reverse geocoding for each second of driving:

  • Level 1 – arterial streets
  • Level 2 – minor arterial streets
  • Level 3 – distributor and collector streets
  • Level 4 – local streets

From GPS → type pf road Level 4 Level 4 Level 3 Level 2 Original drive cycle Per driver

  • 2. Data and Methods

2.3. Model development – drive cycle adjustment

15 Second-by-second power requirement (VSP) Original 1 month driving data (speed, altitude) Verification of speed criteria:

  • Maximum speed (km/h) according to road level

Adjustment of speed and acceleration to fulfill criteria by maintaining:

  • Total distance
  • Stops along road infrastructure
  • Power requirements

From GPS → type pf road Level 4 Level 4 Level 3 Level 2 Original drive cycle Per driver

  • 2. Data and Methods

2.3. Model development – drive cycle adjustment

16 Per driver From GPS → type pf road Level 4 Level 4 Level 3 Level 2 Original drive cycle New drive cycle with modified speed limits

slide-5
SLIDE 5
  • 2. Data and Methods

2.4. Definition of scenarios

17

4 scenarios were considered and compared to the real-world driving cycle (BAU):

BAU Sc 1. Sc 2. Sc 3. Sc 4. L1

  • 120

90 90 90 L2

  • 50

50 50 50 L3

  • 50

50 50 30 L4

  • 50

50 30 30 Speed limit (km/h) Street level

  • 2. Data and Methods

2.4. Application of scenarios

18

Application of scenarios to assess:

  • total driving time (for same

distance)

  • average speed (km/h)
  • acceleration
  • energy consumption (g/s and

l/100km)

  • energy consumption reduction

potential (%)

  • 3. Results

Average of all drivers

19

  • Up to 11 km/h reduction in Scenario 4, with corresponding increase

in travel time (44%)

  • Sc. 1
  • Sc. 2
  • Sc. 3
  • Sc. 4

Distance (km) Difference to BAU (%) 0% 0% 0% 0% Difference to BAU (%) 1% 17% 41% 44% Difference in hours 0.30 7.57 18.13 18.90 Difference to BAU (%)

  • 58% -1340% -2641% -2787%

Difference in km/h

  • 0.29
  • 5.58
  • 10.51
  • 10.92

Maximum speed (km/h) Difference to BAU (%)

  • 17%
  • 38%
  • 38%
  • 38%

acc slope >0 Difference to BAU (%)

  • 8%
  • 52%
  • 125%
  • 104%

acc slope <0 Difference to BAU (%)

  • 2%
  • 103%
  • 176%
  • 185%

Average positive VSP (W/kg) Difference to BAU (%)

  • 4%
  • 35%
  • 50%
  • 53%

Energy consumption (MJ/km) Difference to BAU (%)

  • 13%
  • 2%

5% 7% Scenarios Time (h) Average Speed (km/h)

  • 3. Results

Average of all drivers

20

  • Very significant reductions in acceleration (connected with driver

aggressiveness), mostly noticeable in deceleration events

  • Sc. 1
  • Sc. 2
  • Sc. 3
  • Sc. 4

Distance (km) Difference to BAU (%) 0% 0% 0% 0% Difference to BAU (%) 1% 17% 41% 44% Difference in hours 0.30 7.57 18.13 18.90 Difference to BAU (%)

  • 58% -1340% -2641% -2787%

Difference in km/h

  • 0.29
  • 5.58
  • 10.51
  • 10.92

Maximum speed (km/h) Difference to BAU (%)

  • 17%
  • 38%
  • 38%
  • 38%

acc slope >0 Difference to BAU (%)

  • 8%
  • 52%
  • 125%
  • 104%

acc slope <0 Difference to BAU (%)

  • 2%
  • 103%
  • 176%
  • 185%

Average positive VSP (W/kg) Difference to BAU (%)

  • 4%
  • 35%
  • 50%
  • 53%

Energy consumption (MJ/km) Difference to BAU (%)

  • 13%
  • 2%

5% 7% Scenarios Time (h) Average Speed (km/h)

slide-6
SLIDE 6
  • 3. Results

Average of all drivers

21

  • Scenario 1 presents slight impacts → understandable at urban scale;

Scenario 3 and 4 have very similar impacts

  • Sc. 1
  • Sc. 2
  • Sc. 3
  • Sc. 4

Distance (km) Difference to BAU (%) 0% 0% 0% 0% Difference to BAU (%) 1% 17% 41% 44% Difference in hours 0.30 7.57 18.13 18.90 Difference to BAU (%)

  • 58% -1340% -2641% -2787%

Difference in km/h

  • 0.29
  • 5.58
  • 10.51
  • 10.92

Maximum speed (km/h) Difference to BAU (%)

  • 17%
  • 38%
  • 38%
  • 38%

acc slope >0 Difference to BAU (%)

  • 8%
  • 52%
  • 125%
  • 104%

acc slope <0 Difference to BAU (%)

  • 2%
  • 103%
  • 176%
  • 185%

Average positive VSP (W/kg) Difference to BAU (%)

  • 4%
  • 35%
  • 50%
  • 53%

Energy consumption (MJ/km) Difference to BAU (%)

  • 13%
  • 2%

5% 7% Scenarios Time (h) Average Speed (km/h)

  • 3. Results

Results per driver

22

  • D5, 6, 10, 11 and 12

with higher reductions in Avg. Speed

  • Association with the

context and conditions of driving

  • 3. Results

Energy impacts per driver

23

  • This change in drive

cycle is reflected in modifications in energy consumption

  • Sc. 1 results in

reduction in energy consumption, but speed limitation in Sc 2, 3 and 4 result in up to 15% increases

  • 3. Results

Energy impacts per driver

24

Fuel consumption Local pollutants (HC, CO, NOx, PM) Opportunity for electric mobility for mitigating impacts

slide-7
SLIDE 7
  • 3. Results

Connection with risk of injury accidents

25

  • Estimation of risk of injury accidents reduction based on speed1

Risk of injury accidents

1 Nilsson, 2004; Taylor, 2002; Taylor, 2000.

Drivers Sc . 1

  • Sc. 2
  • Sc. 3
  • Sc. 4

D10

  • D12

=

  • D6

=

  • D11

=

  • D5

=

  • D4

=

  • D8

=

  • D1

=

  • D2

=

  • D7

=

  • D9

=

  • D3

=

  • Caption:

=

~0%

  • <0% to -30%
  • 60% to -30%
  • <-60%
  • 3. Results

Connection with fatal accidents risk

26

  • Estimation of risk of fatal accidents reduction based on speed1

Risk of fatal accidents

1 Nilsson, 2004; Taylor, 2002; Taylor, 2000. Caption:

=

~0%

  • <0% to -30%
  • 60% to -30%
  • <-60%

Drivers

  • Sc. 1
  • Sc. 2
  • Sc. 3
  • Sc. 4

D10

  • D12
  • D6
  • D11
  • D5
  • D4

=

  • D8

=

  • D1

=

  • D2

=

  • D7

=

  • D9

=

  • D3

=

  • 4. Conclusions

and future work

27

  • Assessment of impacts of reducing speed limits on a driver perspective

→ Combination of naturalistic driving data + drive cycle modelling approach → Reductions of up to 11 km/h resulting in:

  • increase of up to 15% in energy

consumption

  • decrease in driver aggressiveness

(up to 180%)

  • significant reductions in the risk
  • f injury and fatal accidents
  • 4. Conclusions

and future work

28

  • Several improvements can be performed in:

→ Limiting aggressive behavior (acceleration) and not only speed → Increase sample of drivers but focus on high accident rate zones and characterize driving behavior → Improve speed and accident risk relation → Perform traffic simulation of areas of interest for quantification

  • f impacts at fleet scale including vehicle interactions
slide-8
SLIDE 8

Acknowledgements

This work was supported by Fundação para a Ciência e Tecnologia, through IN+, Strategic Project UID/EEA/50009/2013.

Using naturalistic driving data to evaluate speed limit reductions: energy, environmental and safety assessment

29

Thank you!

Contact information: patricia.baptista@tecnico.ulisboa.pt http://patriciacbaptista.weebly.com/

30

Using naturalistic driving data to evaluate speed limit reductions: energy, environmental and safety assessment

  • 3. Results

Results per driver

31

  • D5, 6, 10, 11 and 12

with higher reductions in Avg. Speed

  • Performance of

drivers varies across levels: decreasing speed trend with levels

L1 L2 L3 L4 D3 72.4 28.8 18.9 19.9 D12 70.8 28.2 23.7 29.0 D4 66.5 22.7 15.9 14.2 D6 66.4 24.2 15.4 11.9 D11 65.9 33.0 26.6 32.2 D2 60.9 29.5 22.5 16.8 D5 58.2 29.4 20.5 19.0 D9 54.6 30.3 21.1 31.7 D10 54.4 23.6 18.0 20.1 D8 51.3 20.1 20.6 14.7 D1 51.0 26.8 18.8 17.6 D7 48.9 22.6 17.9 15.5 Average speed (km/h) Drivers

  • 3. Results

Connection with accident risk

32

  • Quantification of risk reduction based on Lund, 2005: correlations between

reduction in speed and risk of accident