Final??? Review of WLTP Definitions Bill Coleman – Tokyo Sept./Oct. 2015 WLTP-12-20e.
WLTP-DHC-05-04 1 WLTC* Methodology ~ Initial Analysis Results & Modification on proposed methodology...
-
Upload
elisabeth-wade -
Category
Documents
-
view
215 -
download
0
Transcript of 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
2.1. 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
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
・ Speed limit ≤ 60km/h
・ exclude mountain areas
Motorways
(within City and between Cities)
・ Speed limit ≤ 100km/h
・ exclude mountain areas
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 1.17%Off peak 3.26%Weekend 0.62%
Peak 6.64% 2.98% 1.98%Off peak 14.05% 8.93% 14.68%Weekend 6.63% 4.76%
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%
Peak 0.18% 2.98% 1.98%Off peak 0.39% 13.69% 9.92%Weekend 0.12% 4.76%
Peak 0.32%Off peak 0.89%Weekend 0.52%
Peak 0.10%Off peak 0.24%Weekend 0.05%
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
Urban Rural Motorway
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
Urban Rural Motorway
Ave
rage
spe
ed (
km/h
)
Japan EU KoreaUS India China
2.3.1. Parameter Comparison-2
◆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
Urban Rural Motorway
Idling
Cruise
Deceletation
Acceleration
0
200
400
600
800
1000
1200
PC LDCV ALL PC LDCV ALL PC LDCV ALL
Urban Rural Motorway
Ave
rage
sho
rt tr
ip d
urat
ion
(s)
Japan EU KoreaUS India China
0
5
10
15
20
25
30
35
40
45
50
PC LDCV ALL PC LDCV ALL PC LDCV ALL
Urban Rural Motorway
Ave
rage
idlin
g du
ratio
n (s
)
Japan EU KoreaUS India China
2.3.1. Parameter Comparison-3
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 %
Average: 25.6 km/h Idling ratio: 27.1 % Average: 52.9 km/h Idling ratio: 6.8 % Average: 99.4 km/h Idling ratio: 2.7 %
Average: 17.7 km/h Idling ratio: 38.5 % Average: 40.6 km/h Idling ratio: 14.5 % Average: 49.4 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
Average: 63 km/h Average: 87 km/h Average: 124 km/h
Average: 47 km/h Average: 72 km/h Average: 85 km/h
Urban Rural Motorway
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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
Urban Rural Motorway
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
Short trip duration (s)
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
Short trip duration (s)
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
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. 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
Urban Rural Motorway
◆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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
Average: 12.1 km/h Idling ratio: 41.0 % Average: 32.2 km/h Idling ratio: 21.6 % Average: 69.0 km/h Idling ratio: 3.1 %
Average: 13.8 km/h Idling ratio: 35.8 % Average: 37.9 km/h Idling ratio: 10.1 % Average: 75.7 km/h Idling ratio: 2.6 %
Average: 11.7 km/h Idling ratio: 43.6 % Average: 31.1 km/h Idling ratio: 21.3 % Average: 59.1 km/h Idling ratio: 4.0 %
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
Urban Rural Motorway
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
Short trip duration (s)
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
Short trip duration (s)
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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
Maximum speed (km/h)
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
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 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
Short trip duration (s)
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・・・