WLTP-DHC-05-04 1 WLTC* Methodology ~ Initial Analysis Results & Modification on proposed methodology...

1

WLTP-DHC-05-04

WLTC* Methodology~ Initial Analysis Results & Modification on proposed methodology ~

Presented by Japan

GRPE/WLTP-IG/ DHC subgroup

14 October 2010

*) WLTC: Worldwide harmonized Light duty driving Test Cycle

2

WLTP-DHC-05-04Table of Contents

1. Purpose2. Initial analysis results

2.1. Overview of collected in-use data2.2. Data analysis method2.3. Result of data analysis2.4. Conclusion

3. Validity of L/M/H method3.1. Overview of L/M/H method3.2. Consideration of threshold speed3.3. Analyzed example of L/M/H method3.4. Conclusion

4. Modification on proposed methodology4.1. Modification list4.2. Modification on proposed methodology

3

WLTP-DHC-05-041. Purpose

1. Early studies Japan and UK proposed the methodology to develop WLTC

(WLTP-DHC-02-04/05/06 as U/R/M method) Japan proposed the alternative technique to convert the in-use

data to another categorization (WLTP-DHC-03-02 as L/M/H method) The alternative technique was well recognized but need more clear

explanation

2. Initial analysis to make clear whether there is discrepancy in same road

category or not to derive the potential problem in the current methodologies

3. Validity of L/M/H method to confirm that the alternative technique (L/M/H method) is valid

to obtain similarity to make a decision which methodology is more appropriate

4. Modification on proposed methodology to improve the methodology based on the initial analysis

4

WLTP-DHC-05-04

1. Purpose2. Initial Analysis Results




2. Initial Analysis Results

5

WLTP-DHC-05-042.1.1. Definition of Road Type

Urban Rural Motorway

IndiaPaved roads in urban areas with a speed limit ≤40 km/hour (exclude mountain areas)

Paved non-motorways outside and inside urban areas with a speed limit between 40 and 60 km/hour

Paved motorways (multi-lane roads specifically constructed and controlled for fast traffic) with a speed 60 to 80 km/hour

Korea

Arterial, collector and local road inside and/or near central business district (CBD). Speed limit is from 40 to 80 km/h, depends on road type

Arterial, collector and local road inside non-urban area. Speed limit is from 50 to 80 km/h, depends on road type

Motorway which is designed, constructed and controlled for faster traffic in urban and rural area. Speed limit is from 100 to 120 km/h, depends on area

Japan

Densely Inhabited District (DID)

･ Speed limit ≤ 60km/h

･ exclude mountain areas

･ Non-Densely Inhabited District

･ Non motorways



Motorways

(within City and between Cities)



EU The definition depends on EU countries

The definition depends on EU countries

The definition depends on EU countries

USA ? ? ?

China ? ? ?

It was agreed that each CP has its own unique definition of each road type (urban/rural/motorway) during 1st DHC meeting. This may lead the discrepancy on speed-acceleration distribution among CPs in same road category.

6

WLTP-DHC-05-042.1.2. Overview of Collected In-use Data

Country / Region

Area

No. of vehiclesTotal driving

duration(*)

(hr)

Total driving distance(*)

(km)Passenger car

Light duty commercial

vehicle

EU

(*)ongoing

Germany, Italy, Slovenia, France, Belgium, Switzerland

67 6Approx. 2,800

Approx.

145,000

Japan

12 prefecture:

Tokyo, Osaka , Kyoto, Nagoya, Shizuoka, etc 11 13

Approx. 1,800

Approx. 55,000

Korea(**) Seoul 6 2Approx. 1,100

Approx. 35,000

(*) Exclude NG data

(**) Korea: Not exist Weekend data in Urban and Rural

7

WLTP-DHC-05-042.1.3. Traffic Volume Weighting in Driving Conditions

Japan EU Korea US India ChinaPeak 4.36% 2.98% 1.98%

Off peak 12.80% 8.93% 14.68%Weekend 6.24% 4.76%

Peak 1.04% 2.98% 1.98%Off peak 3.13% 8.93% 14.68%Weekend 0.56% 4.76%

Peak 5.67%Off peak 13.50%Weekend 6.32%



Peak 1.81% 2.98% 1.98%Off peak 4.42% 13.69% 14.68%Weekend 0.83%

Peak 0.68% 2.98% 1.98%Off peak 1.72% 8.93% 9.92%Weekend 1.73% 4.76% 4.76%




Sum 100.00% 100.00% 100.00% 0.00% 0.00% 0.00%

◆ EU and Korea･Urban : Rural : Motorway = 1 : 1 : 1, PC : LDCV = 1 : 1

6 hours4 hours

Off peak 18 / 20Weekend 2 days 24 hours

PC

LDCV

Urban-2

PC

LDCV

Driving condition weighting factorDrivingperiod

Motorway-2

PC

LDCV

Rural

PC

LDCV

Motorway-1

PC

5 daysWeekday

Road type Vehicle type

Urban

Rural

Motorway

LDCV

Urban-1

PeakEU: 7:00~10:00, 16:00~19:00KOR: 7:00~9:00, 18:00~20:00

Other time

Temporary weighting for initial analysis

8

WLTP-DHC-05-042.2.1. Preparation for Data Analysis

Common points Remove the noise by filtering process. （ T4253H filter ） Find the acceleration by applying differential equation 　 αi = (Vi+1 - Vi-1) / (Ti+1 - Ti-1) Generate the Speed-acceleration distribution, V ： 5km/h intervals, α ： 1km/h/s interv

als Eliminate the ST with max speed 250km/h or higher

Japanese data Include incomplete ST that transfer to High phase (e.g. Vmax 80, 90km/h)≧ (See the next slide for more details on incomplete ST ）

EU data （ Pre-check methodology ） not_ok is set “true”, when failures or inconsistencies were found during data preproces

sing and when the following indicators are set “true”, not_ok_dv is set “true”, when the difference between the original vehicle speed v and t

he smoothed speed vg is < -7 km/h or > 7 km/h. This is typically the case, when there are gaps or jumps in the speed pattern.

not_ok_v_start_end is set to “true”, when v_start and/or v_end of a short trip is higher than 12.6 km/h.

not_ok_a_min_max is set to “true”, when the acceleration exceeds 4 m/s² or is less then -4.5 m/s².

Datasets and short trips, indicated with those not_ok indicators should be excluded from the further analysis.

Korean data Apply same method as Japanese data

9

WLTP-DHC-05-042.2.2. Incomplete Short Trip

Both complete and incomplete in-use data that transfer to High phase are analyzed.

Because of long driving duration at high speed, including the STs with no start acceleration phase or braking phase does not impact.

If the data is excluded, the few remaining data might not represent actual driving conditions.

Time

Ve

h.

spe

ed

Time

Ve

h.

spe

ed

Time Time

①Complete ST

②Incomplete ST

10

WLTP-DHC-05-04

0

10

20

30

40

50

60

Pea

k

Off

peak

Wee

kend

Pea

k

Off

peak

Wee

kend

Tot

al

Pea

k

Off

peak

Wee

kend

Pea

k

Off

peak

Wee

kend

Tot

al

Pea

k

Off

peak

Wee

kend

Pea

k

Off

peak

Wee

kend

Tot

al

PC LDCV All PC LDCV All PC LDCV All


Driv

ing

dist

ance

( ×

103 k

m)

Japan EU KoreaUS India China

2.3.1. Parameter Comparison-1

◆Driving distance

11

WLTP-DHC-05-04

0

20

40

60

80

100

120

Pea

k

Off

peak

Wee

kend

Pea

k

Off

peak

Wee

kend

Wei

ghte

d

Pea

k

Off

peak

Wee

kend

Pea

k

Off

peak

Wee

kend

Wei

ghte

d

Pea

k

Off

peak

Wee

kend

Pea

k

Off

peak

Wee

kend

Wei

ghte

d

PC LDCV All PC LDCV All PC LDCV All


Ave

rage

spe

ed (

km/h

)



◆Average speed

12

WLTP-DHC-05-04

25

.3

26

.7

25

.0

0.0

0.0

0.0

24

.8

29

.3

31

.2

0.0

0.0

0.0

21

.6

17

.6 27

.4

0.0

0.0

0.0

23

.7

25

.1

22

.1

0.0

0.0

0.0

20

.6 26

.6

24

.8

0.0

0.0

0.0

20

.6

16

.2

24

.1

0.0

0.0

0.0

19

.2 21

.1

14

.5

0.0

0.0

0.0

39

.1

37

.3

29

.5

0.0

0.0

0.0

49

.5

63

.6 40

.1

0.0

0.0

0.0

31

.8

27

.1

38

.5

0.0

0.0

0.0

15

.5 6.8

14

.5

0.0

0.0

0.0

8.3 2

.7

8.3

0.0

0.0

0.00%

20%

40%

60%

80%

100%

Japa

n

EU

Kor

ea US

Indi

a

Chi

na

Japa

n

EU

Kor

ea US

Indi

a

Chi

na

Japa

n

EU

Kor

ea US

Indi

a

Chi

na


Idling

Cruise

Deceletation

Acceleration

0

200

400

600

800

1000

1200

PC LDCV ALL PC LDCV ALL PC LDCV ALL


Ave

rage

sho

rt tr

ip d

urat

ion

(s)


0

5

10

15

20

25

30

35

40

45

50

PC LDCV ALL PC LDCV ALL PC LDCV ALL


Ave

rage

idlin

g du

ratio

n (s

)



Average ST duration Average IDLE duration

Ratio of 4 driving mode

13

WLTP-DHC-05-04

Average: 22.3 km/h Idling ratio: 31.8 % Average: 37.8 km/h Idling ratio: 15.5 % Average: 62.5 km/h Idling ratio: 8.3 %



Rural

Japan

Korea

EU

Urban Motorway

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

EU Urban

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

EU Rural

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

EU Motorway

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Japan Urban

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Japan Rural

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Japan Motorway

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Korea Urban

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Korea Rural

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Korea Motorway

2.3.2. Speed Acceleration Distribution

Discrepancy in speed-acceleration distribution was observed in each road category.

14

WLTP-DHC-05-042.3.3. Frequency of ST Maximum Speed （ Y axis： Frequency of driving duration）

Discrepancy in ST maximum speed distribution was observed.

Average: 49 km/h Average: 63 km/h Average: 89 km/h




3 area

Japan

Korea

EU

0

2

4

6

8

10

12

14

16

18

2 8 14 20 26 32 38 44 50 56 62 68 74 80 86 92 98 104

110

116

122

128

134

140

146

152

158

Maximum speed (km/h)

Fre

quen

cy (

%)

EU Urban

0

2

4

6

8

10

12

14

16

18

2 8 14 20 26 32 38 44 50 56 62 68 74 80 86 92 98 104

110

116

122

128

134

140

146

152

158


Fre

quen

cy (

%)

EU Rural

0

2

4

6

8

10

12

14

16

18

2 8 14 20 26 32 38 44 50 56 62 68 74 80 86 92 98 104

110

116

122

128

134

140

146

152

158


Fre

que

ncy

(%)

EU Motorway

0

2

4

6

8

10

12

14

16

18

2 8 14 20 26 32 38 44 50 56 62 68 74 80 86 92 98 104

110

116

122

128

134

140

146

152

158


Fre

quen

cy (

%)

Japan Urban

0

2

4

6

8

10

12

14

16

18

2 8 14 20 26 32 38 44 50 56 62 68 74 80 86 92 98 104

110

116

122

128

134

140

146

152

158


Fre

quen

cy (

%)

Japan Rural

0

2

4

6

8

10

12

14

16

18

2 8 14 20 26 32 38 44 50 56 62 68 74 80 86 92 98 104

110

116

122

128

134

140

146

152

158


Fre

quen

cy (

%)

JapanMotorway

0

2

4

6

8

10

12

14

16

18

2 8 14 20 26 32 38 44 50 56 62 68 74 80 86 92 98 104

110

116

122

128

134

140

146

152

158


Fre

quen

cy (

%)

Korea Urban

0

2

4

6

8

10

12

14

16

18

2 8 14 20 26 32 38 44 50 56 62 68 74 80 86 92 98 104

110

116

122

128

134

140

146

152

158


Fre

quen

cy (

%)

Korea Rural

0

2

4

6

8

10

12

14

16

18

2 8 14 20 26 32 38 44 50 56 62 68 74 80 86 92 98 104

110

116

122

128

134

140

146

152

158


Fre

quen

cy (

%)

KoreaMotorway

0

102030

4050

6070

8090

100

2 8 14 20 26 32 38 44 50 56 62 68 74 80 86 92 98 104

110

116

122

128

134

140

146

152

158


Cum

. fre

quen

cy (

%)

EU UrbanJapan UrbanKorea Urban

0

102030

4050

6070

8090

100

2 8 14 20 26 32 38 44 50 56 62 68 74 80 86 92 98 104

110

116

122

128

134

140

146

152

158


Cum

. fre

quen

cy (

%)

EU RuralJapan RuralKorea Rural

0

102030

4050

6070

8090

100

2 8 14 20 26 32 38 44 50 56 62 68 74 80 86 92 98 104

110

116

122

128

134

140

146

152

158

Maximum speed (km/h)C

um. f

requ

ency

(%

)

EU MotorwayJapan MotorwayKorea Motorway

15

WLTP-DHC-05-042.3.4. Other Parameters

Discrepancy in ST/IDLE duration distribution was observed.

Idlingduration

Short tripduration


0

10

20

30

40

50

60

70

80

90

100

11

12

13

14

15

16

17

18

19

11

01

11

11

21

13

11

41

15

11

61

17

11

81

19

12

01

Idling duration (s)

Cu

m.

fre

qu

en

cy (

%)

EU Urban Cum. frequency

Japan Urban Cum. frequency

Korea Urban Cum. frequency

0

10

20

30

40

50

60

70

80

90

100

11

12

13

14

15

16

17

18

19

11

01

11

11

21

13

11

41

15

11

61

17

11

81

19

12

01

Idling duration (s)

Cu

m.

fre

qu

en

cy (

%)

EU Rural Cum. frequency

Japan Rural Cum. frequency

Korea Rural Cum. frequency

0

10

20

30

40

50

60

70

80

90

100

11

12

13

14

15

16

17

18

19

11

01

11

11

21

13

11

41

15

11

61

17

11

81

19

12

01

Idling duration (s)

Fre

qu

en

cy (

%)

EU Motorway Cum. frequency

Japan Motorway Cum. frequency

Korea Motorway Cum. frequency

0

10

20

30

40

50

60

70

80

90

100

1

31

61

91

12

1

15

1

18

1

21

1

24

1

27

1

30

1

Short trip duration (s)

Cu

m.

fre

qu

en

cy (

%)

EU Urban Cum. frequency

Japan Urban Cum. frequency

Korea Urban Cum. frequency

0

10

20

30

40

50

60

70

80

90

100

1

10

1

20

1

30

1

40

1

50

1

60

1

70

1

80

1

90

1


Cu

m.

fre

qu

en

cy (

%)

EU Rural Cum. frequency

Japan Rural Cum. frequency

Korea Rural Cum. frequency0

10

20

30

40

50

60

70

80

90

100

1

30

1

60

1

90

1

12

01

15

01

18

01

21

01

24

01

27

01

30

01

33

01

36

01


Fre

qu

en

cy (

%)

EU Motorway Cum. frequency

Japan Motorway Cum. frequency

Korea Motorway Cum. frequency

16

WLTP-DHC-05-042.4. Conclusion of Initial Analysis

It was observed that speed-acceleration distribution/driving characteristics in each region have discrepancy in same road categories.

This discrepancy may mislead the representativeness when developing the harmonized driving cycle.

Therefore, it is necessary to convert to another category where driving characteristics are similar in each region. L/M/H methods is one of countermeasures.

17

WLTP-DHC-05-04





3. Validity of L/M/H Method

18

WLTP-DHC-05-043.1. Overview of WLTP-DHC-03-02

WLTP-DHC-03-02 described the technique to convert the in-use data into vehicle speed oriented categories.

Urban/Rural/Motorway -> Low/Middle/High


◆Image

Low Middle High

The segments that composed of ST and IDLE move into L/M/H categories with the compensated WF.

19

WLTP-DHC-05-043.2.1. Detail of L/M/H Method

Considering the threshold vehicle speed Consider threshold based on each countries’ traffic

condition and driving characteristic Find the threshold that shows similar VA distribution

of each countries Calculate the compensated WF

Using Drive condition WF and total driving duration Convert in-use data（ U/R/M => L/M/H） Analyze VA distribution and driving

characteristics （ L/M/H） Generate the driving cycle in each phase

20

WLTP-DHC-05-04

χ 2 : 0.39 χ 2ave

: 2.37E-03 χ 2 : 0.41 χ 2ave

: 1.75E-03 χ 2 : 0.69 χ 2ave

: χ 2 : 0.95 χ 2ave

: 2.58E-03

χ 2 : 0.56 χ 2ave

: 3.39E-03 χ 2 : 1.55 χ 2ave

: 6.59E-03 χ 2 : 2.48 χ 2ave

: χ 2 : 3.09 χ 2ave

: 8.41E-03

χ 2 : 0.96 χ 2ave

: 5.81E-03 χ 2 : 1.00 χ 2ave

: 4.23E-03 χ 2 : 1.62 χ 2ave

: χ 2 : 1.69 χ 2ave

: 4.59E-03

Σ χ 2: 1.91 Sum: 1.16E-02 Σ χ 2: 2.96 Sum: 1.26E-02 Σ χ 2: 4.79 Sum: Σ χ 2: 5.73 Sum: 1.56E-02

Average

Low (Vmax≤40) Low (Vmax≤70）Low (Vmax≤50） Low (Vmax≤60）

J apan

Korea

EU

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

EU ≤ 40km/h

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

EU ≤ 50km/h

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

EU ≤ 70km/h

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Japan ≤ 40km/h

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Japan ≤ 50km/h

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Japan ≤ 70km/h

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Average ≤ 50km/h

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Korea ≤ 40km/h

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Korea ≤ 50km/h

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Korea ≤ 70km/h

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Average ≤ 70km/h

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Average ≤ 40km/h

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

EU ≤ 60km/h

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Japan ≤ 60km/h

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Average ≤ 60km/h

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Korea ≤ 60km/h

3.2.2. Consideration of Threshold Speed - 1

<Method1> Based on similarity of VA distributionThe threshold of Low/Middle

Vmax ≤ 40 ?, Vmax ≤ 50 ?, Vmax ≤ 60 ?, Vmax ≤ 70 ?

The threshold of Middle/High

VL/M < Vmax ≤70 ?, VL/M < Vmax ≤80 ?, VL/M < Vmax ≤90 ?, VL/M < Vmax ≤100 ?

21

WLTP-DHC-05-04

JPN EU KOR Max-Min

JPN EU KOR Max-Min

JPN EU KOR Max-Min

JPN EU KOR Max-Min

JPN EU KOR Max-Min

Urban 22 26 18 8 0.18 0.19 0.19 0.01 32 27 38 11 67 77 55 22 28 24 30 7

≦ 40 8 8 7 1 0.21 0.21 0.20 0.01 47 46 50 5 33 30 28 5 25 18 22 7

≦ 50 12 14 12 2 0.20 0.21 0.20 0.01 41 36 44 8 43 43 39 4 26 18 25 8

≦ 60 17 19 16 3 0.19 0.21 0.19 0.02 36 29 39 10 54 56 50 7 27 18 28 9

≦ 70 20 22 18 4 0.18 0.20 0.19 0.02 34 26 36 10 62 67 59 8 28 18 29 10

Rural 38 53 41 15 0.12 0.14 0.15 0.03 15 7 14 9 132 397 184 264 21 17 29 12

40 - 80 28 33 26 7 0.18 0.19 0.18 0.02 24 14 26 12 102 149 149 47 31 19 38 19

40 - 90 29 34 27 7 0.17 0.19 0.18 0.02 25 12 24 12 104 151 127 47 31 19 38 20

50 - 90 32 38 31 7 0.17 0.19 0.18 0.02 22 10 21 11 120 185 158 65 32 19 41 22

60 - 90 37 42 37 5 0.16 0.18 0.17 0.02 19 8 16 11 144 238 221 94 31 18 41 23

70 - 110 45 52 46 6 0.15 0.16 0.15 0.01 15 5 10 10 207 359 351 153 30 18 39 21

Motorway 63 99 49 50 0.09 0.07 0.11 0.04 8 3 8 6 295 828 312 533 19 20 19 1

80< 59 70 53 17 0.12 0.13 0.15 0.03 6 3 7 4 421 658 518 236 25 19 36 17

90< 69 76 59 17 0.11 0.12 0.15 0.04 3 3 4 1 696 782 785 89 20 19 31 12

110< 85 87 64 23 0.09 0.11 0.17 0.08 2 2 2 0 1086 1117 1213 127 18 19 24 6

Road typeSpeed rage

UrbanLow

RuralMiddle

MotorwayHigh

Average short tripduration

s

Average idling duration

s

Idling ratio

%

Average speed

km/h

Relative positiveacceleration

m/s2

3.2.2. Consideration of threshold speed - 2

<Method2> Based on parameter value

Select the candidate threshold speed based on least discrepancy in each characteristic.

22

WLTP-DHC-05-043.2.2. Consideration of threshold speed - 3

<Method3> Based on maximum speed distribution

0

2

4

6

8

10

12

14

16

18

2 814 20 26 32 38 44 50 56 62 68 74 80 86 92 98

104

110

116

122

128

134

140

146

152

158


Freq

uenc

y (%

)

Japan Total

0

2

4

6

8

10

12

14

16

18

2 814 20 26 32 38 44 50 56 62 68 74 80 86 92 98

104

110

116

122

128

134

140

146

152

158


Freq

uenc

y (%

)

EU Total

0

2

4

6

8

10

12

14

16

18

2 814 20 26 32 38 44 50 56 62 68 74 80 86 92 98

104

110

116

122

128

134

140

146

152

158


Freq

uenc

y (%

)

Korea Total

50 / 90?

0

2

4

6

8

10

12

2 814 20 26 32 38 44 50 56 62 68 74 80 86 92 98

104

110

116

122

128

134

140

146

152

158


Freq

uenc

y (%

)Japan Urban

Japan Rural

Japan Motorw ay

0

2

4

6

8

10

12

2 814 20 26 32 38 44 50 56 62 68 74 80 86 92 98

104

110

116

122

128

134

140

146

152

158


Freq

uenc

y (%

)

EU Urban

EU RuralEU Motorw ay

0

2

4

6

8

10

12

2 814 20 26 32 38 44 50 56 62 68 74 80 86 92 98

104

110

116

122

128

134

140

146

152

158


Freq

uenc

y (%

)

Korea Urban

Korea Rural

Korea Motorw ay

65 / 110?

60 / 90?

After completion of all data acquisition, final threshold speed will be determined by taking into account of three methods.

50 / 80?

23

WLTP-DHC-05-04




Low (Vmax≦ 50) High (90＜Vmax)Middle (Vmax≦ 90）

Japan

Korea

EU

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

EU Low

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

EU Middle

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

EU High

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Japan Low

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Japan Middle

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Japan High

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Korea Low

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Korea Middle

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Korea High

3.3.1. Analysis Sample - VA distribution

It was confirmed that VA distribution is similar in L/M category but slight discrepancy in High category still exist.

24

WLTP-DHC-05-043.3.2. Analysis Sample - Average speed

U/R/M L/M/H (threshold: 50/90)

As an example, convert into L/M/H at 50/90km/h

22.3

25.6

17.9

0.0

0.0

0.0

37.8

52.9

40.6

0.0

0.0

0.0

62.5

99.4

49.4

0.0

0.0

0.0

0

20

40

60

80

100

120

Japa

n

EU

Kor

ea US

Indi

a

Chi

na

Japa

n

EU

Kor

ea US

Indi

a

Chi

na

Japa

n

EU

Kor

ea US

Indi

a

Chi

na


Ave

rage

spe

ed (

km/h

)

Average speed km/h

12.1

13.8

11.7

32.2 37

.9

31.1

69.0 75

.7

59.1

0

20

40

60

80

100

120

Japa

n

EU

Kor

ea US

Indi

a

Chi

na

Japa

n

EU

Kor

ea US

Indi

a

Chi

na

Japa

n

EU

Kor

ea US

Indi

a

Chi

na

Low≦ 50

Middle50－90

High90<

Ave

rage

spe

ed (

km/h

)

Average speed km/h

25

WLTP-DHC-05-043.3.3. Analysis Sample - Parameters

Max.speed

Idlingduration

Short tripduration

Low (Vmax≦ 50) Middle (Vmax≦ 90） High (90＜Vmax)

0

10

20

30

40

50

60

70

80

90

100

1 11 21 31 41 51 61 71 81 91 101

111

121

131

141

151

161

171

181

191

201

Idling duration (s)

Cum

. fre

quen

cy (

%)

EU Low

Japan Low

Korea Low

0

10

20

30

40

50

60

70

80

90

100

1 11 21 31 41 51 61 71 81 91 101

111

121

131

141

151

161

171

181

191

201

Idling duration (s)

Cum

. fre

quen

cy (

%)

EU Middle

Japan Middle

Korea Middle

0

10

20

30

40

50

60

70

80

90

100

1 11 21 31 41 51 61 71 81 91 101

111

121

131

141

151

161

171

181

191

201

Idling duration (s)

Cum

. fre

quen

cy (

%)

EU High

Japan High

Korea High

0

10

20

30

40

50

60

70

80

90

100

1 21 41 61 81 101

121

141

161

181

201

221

241

261

281

301


Cum

. fre

quen

cy (

%)

EU Low

Japan Low

Korea Low

0

10

20

30

40

50

60

70

80

90

100

1

101

201

301

401

501

601

701

801

901

Short trip duration (s)C

um. f

requ

ency

(%

)

EU Middle

Japan Middle

Korea Middle

0

10

20

30

40

50

60

70

80

90

100

1

301

601

901

1201

1501

1801

2101

2401

2701

3001

3301

3601


Cum

. fre

quen

cy (

%)

EU High

Japan High

Korea High

0

10

20

30

40

50

60

70

80

90

100

2 8 14 20 26 32 38 44 50 56 62 68 74 80 86 92 98 104

110

116

122

128

134

140

146

152

158


Cum

. fre

quen

cy (

%)

EU Low

Japan Low

Korea Low

0

10

20

30

40

50

60

70

80

90

100

2 8 14 20 26 32 38 44 50 56 62 68 74 80 86 92 98 104

110

116

122

128

134

140

146

152

158


Cum

. fre

quen

cy (

%)

EU Middle

Japan Middle

Korea Middle

0

10

20

30

40

50

60

70

80

90

100

2 8 14 20 26 32 38 44 50 56 62 68 74 80 86 92 98 104

110

116

122

128

134

140

146

152

158


Cum

. fre

quen

cy (

%)

EU High

Japan High

Korea High

26

WLTP-DHC-05-043.4. Summary of L/M/H Method

It was observed that each region driving characteristics and VA distributions comparatively show similarity by converting into Low/Middle/High speed based on ST max speed.

The appropriate threshold will be decided based on each countries’ traffic condition, driving characteristic and VA distribution after completion of all in-use data collection.

27

WLTP-DHC-05-044. Modification on Proposed Methodology




4. Modification on Proposed Methodology4.1. Modification Lists4.2. Modification on proposed methodology

28

WLTP-DHC-05-044.1. Modification Lists

Based on the initial in-use data analysis, it was observed that the following items need to be revised for improvement of driving cycle development and for efficient work.

Method for ST/IDLE selection Method for cycle development of high speed mode Method for ST selection combination in each mode

29

WLTP-DHC-05-044.2. Method for ST/IDLE Selection

Original (Boundary)Divide cumulative frequency dist

ribution into “Ni, ST +1”

Select the durations on each boundary (dividing) line

Revised (Average)Divide cumulative frequency

distribution into “Ni, ST”

Select the average duration in each class

ST/ID duration

Cum

ulat

ive

freq

uenc

y

STD1STD2 STDN・・・・・

ST/ID durationC

umul

ativ

e fr

eque

ncy

STD1 STD2 STDN・・・・・

100%(N+1)

100%N

It is possible to develop more representative cycle by applying revised method

30

WLTP-DHC-05-044.3.1. Concerns on High-speed Driving Cycle Development

Since most of ST duration （ e.g. 696sec, Japan ） in high-speed mode

are longer than cycle duration, the number of potential STs will be dramati

cally reduced. This may lead the critical discrepancy in its representativen

ess.

Example) Japan High phase（ 90km/h < ）

0.0

0.5

1.0

1.5

2.0

2.5

3.0

1

251

501

751

1001

1251

1501

1751

2001

2251

2501

2751

3001

3251

3501

3751

4001

4251

4501

4751

5001

5251

5501

5751

6001

6251

6501

6751

7001

ST duration (s)

Fre

quen

cy (

%)

0

10

20

30

40

50

60

70

80

90

100

Cum

. fre

quen

cy (

%)

Frequency %

Cum. frequency %

Average ST duration: 696 sec

Average Idling duration: 20 sec

Number of ST: (600 - 20) / (696 + 20) = 0

Idling ratio: 3.1 %

Driving ratio: 96.9 %

31

WLTP-DHC-05-044.3.2. Process of high-speed cycle development

1. Determine the high-speed cycle duration (e.g. 600 sec.)

2. Determine the ST duration based on average ST duration and idling duration ratio of in-use data, the divide the ST into 5 parts (take-off1, take-off2, cruise, slow-down1, slow-down2).

3. Extract the driving data which meet each part configuration from the complete in-use ST.

4. Select the least chi-squared extracted driving data in each part, then combine the selected data to develop the complete cycle.

(note) if the complete in-use data is less chi-squared value than

combined cycle, this specific data can be used for high-speed cycle.

32

WLTP-DHC-05-044.3.3. Divide into 5 Parts of High-Speed Cycle

-16

-12

-8

-4

0

4

8

12

16

-20 0 20 40 60 80 100 120 140 160Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Japan High Speed rangeFrequency

(%)Target

durationDivide

part

Idling 3.1 18 9

0~75, 75~0 36.1 217 108

75~90, 90~75 30.0 179 89

90~ 30.9 186

Total 100 600

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0 ~ 75 km/h, 108 sec 75 ~ 90 km/h, 89 sec 90 ~ 90 km/h, 188 sec 90 ~ 75 km/h, 89 sec 75 ~ 0 km/h, 108 sec

30% 31%36%

Adjust 2 sec.

100 ％ = 600 sec.

3 %

Generate speed range and duration Divide into 3 speed range based on dividing frequency distribution

Example: 0~75 km/h, 75~90 km/h, 90 km/h～ Decide target duration in each range, then divide into two portions (take-off and slow-down)

Example of 0~75km/h: 36% => 217 sec => take-off part 108 sec， slow-down 108 sec

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

(take-off1) (Cruise)

(slow-down2)

(take-off2) (slow-down1)

33

WLTP-DHC-05-044.3.4. Extraction of Driving Data

Extract the driving data which meet each part definition* from complete ST. Sample definition <Take-off 1 part>

speed range : 0 ~ 75 km/h (with±0.5km/h), duration : 108sec.

Time

Ve

h.

spe

ed

Time

Ve

h.

spe

ed

Time

Ve

h.

spe

ed

0~75 75~90 90~90 ・・・ 75~0・・・

・・・

・・・

・・・

・・・

0～ 75 km/h

108 s

0～ 75 km/h

108 s

34

WLTP-DHC-05-044.3.5. Method to Develop High-Speed Driving Cycle

-10

-8

-6

-4

-2

0

2

4

6

8

10

-20 0 20 40 60 80 100 120 140Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Reference database High 90<

0

20

40

60

80

100

120

140

0 100 200 300 400 500 600

Time (s)

Spe

ed (

km/h

)(take-off1) (Cruise) (slow-down2)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

(take-off2) (slow-down1)

Candidate driving data

combined

Least chi-squared distribution

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

0 50 100 150 200

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

140

0 100 200 300 400 500 600

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

140

0 100 200 300 400 500 600

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

140

0 100 200 300 400 500 600

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

140

0 100 200 300 400 500 600

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

140

0 100 200 300 400 500 600

Time (s)

Spe

ed (

km/h

)

0

20

40

60

80

100

120

140

0 100 200 300 400 500 600

Time (s)

Spe

ed (

km/h

)

Seek the combination cycle with least chi-square value compared with unified cycle

35

WLTP-DHC-05-04

0

20

40

60

80

100

120

0 100 200 300 400 500 600Time (s)

Spe

ed (

km/h

)

Actual trip-1

Actual trip-2

4.3.6. Selection of Complete Short Trip

◆complete short trip -1 (X2=1.12) ◆complete short trip -2 (X2=1.15)

Few possibility to find the complete in-use data to represent the unified cycle

-10

-8

-6

-4

-2

0

2

4

6

8

10

-20 0 20 40 60 80 100 120 140Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Reference database High 90<

◆United VA distribution (90<)

◆There is only a few complete STs with target duration in Japan database

◆Comparison of VA distribution

-10

-8

-6

-4

-2

0

2

4

6

8

10

-20 0 20 40 60 80 100 120 140Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Actual short trip ST1

-10

-8

-6

-4

-2

0

2

4

6

8

10

-20 0 20 40 60 80 100 120 140Speed (km/h)

Acc

ele

ratio

n (

km/h

/s)

Actual short trip ST2

36

WLTP-DHC-05-044.4.1. Method for the ST Combinations

◆Japan reference database

Average ST duration: 42.7 sec

Average Idling duration: 26.5 dec

TST 8 15 22 30 39 51 67 111

NST 960 764 685 689 556 419 250 75

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

11

12

13

14

15

16

17

18

19

11

01

11

11

21

13

11

41

15

11

61

17

11

81

19

12

01


Fre

qu

en

cy (

%)

0

10

20

30

40

50

60

70

80

90

100

Cu

m. f

req

ue

ncy

(%

)

Japan Low Frequency

Japan Low Cum. frequency

Combination number = 1.5 × 1021 (Only Japan data)

When it takes 0.1 sec for chi-square calculation per one combination, it takes 4.8×1012 Year （ 1.5×1020 Sec ） for all possible combinations (only Japanese data). Therefore, it is necessary to reduce number of combinations for practical work.

◆The candidates for each short trip duration for the total combination numbers and short trip numbers

◆ Short trip duration in Low phase (threshold: 50/90)

◆Test cycle

Desired cycle duration: 600 sec

NST = (600 - 26.5) / (42.7 + 26.5) = 8.3

Number of short trips NST: 8

TST: 8, 15, 22, 30, 39, 51, 67, 111 = 343 sec

Average ST duration: 42.9 sec

37

WLTP-DHC-05-044.4.2. Selection of each ST duration

ST selection criteria ST within average ±1σ in each ST duration

Average vehicle speed Acceleration duration ratio Deceleration duration ratio

Smaller chi-squared value is higher priority The number of potential STs in each ST duration

Longer ST has more potential STs since it has bigger influence on chi-square value.

Total number of combinations is less than 107 ～ 8 . Approximately 3 days on Xeon X5492 (Quad core, 3.4GHz)

-10

0

10

20

30

40

50

60

70

0 50 100 150 200 250 300 350

X1 X2 X3 X4 X5 sec. Y sec.

N unitsN * X5/Y unitsN * X4/Y・・・

WLTP-DHC-05-04 1 WLTC* Methodology ~ Initial Analysis Results & Modification on proposed methodology...

Documents

Transcript of WLTP-DHC-05-04 1 WLTC* Methodology ~ Initial Analysis Results & Modification on proposed methodology...