Study on Winter Road Surface Friction Characteristics and Their - - PowerPoint PPT Presentation

Study on Winter Road Surface Friction Characteristics and Their Reproducibility

Roberto TOKUNAGA1, Akihiro FUJIMOTO1, Kenji SATO1, Naoto TAKAHASHI1, Tateki ISHIDA1 and Makoto KIRIISHI2

1Civil Engineering Research Institute for Cold Region, P.W.R.I., Japan 2Hokkaido Regional Development Bureau, M.L.I.T., Japan

1

2016 International Conference & Workshop on Winter Maintenance and Surface Transportation Weather

Contents

Background & purpose Friction monitoring on winter roads Variations in friction on a highway in winter

Under different weather conditions, and reproducibility of those variations

Summary and future issues

2

 The patterns of variation in friction values, and their reproducibility, are examined, based on monitoring data  A technique to identify critical management sections more accurately and without constant monitoring can be established

Background & Purpose

 Winter road surface conditions are determined and assessed visually

 But… it is difficult to achieve accountability for such public works

 A technique for monitoring roadway friction values has been developed…

 However, constant monitoring of the road surface is difficult

Aim & Purpose

Monitoring of friction on winter roads

Monitoring

On a 44-km section of NH230 between downtown Sapporo and Nakayama Pass Monitored since 2007/2008 winter

Continuous Friction Tester (CFT)

The CFT can be attached to a SUV This device calculates friction value by measuring the axial force created by installing a test tire 1-2 degrees

• ff axis from the direction of travel

4

Downtown （KP1.0 - 8.0） Suburban area （KP8.0 - 18.0） Rural area （KP18.0 - 26.0） Mountainous area （KP26.0 - 37.0） Mountain pass (KP38.5 - 45.0)

Monitored section of Natl. Hwy. 230

Continuous Friction Tester (CFT)

Angle: 1°~2° Tire

F

Towing Vehicle Direction

Spatial distribution of HFN values

5

Spatial distribution of HFN values on NH230 (January)

Monthly cumulative snow (cm) Hours of sow removal(h)

• Dec. Jan. Feb.

2008 2009 2010 2011 2012 2013 2014 Year

Monthly average temperature (ºC)

• Dec. Jan. Feb.

Monthly average temperature in Sapporo area Monthly Cumulative Snowfall in Sapporo area Snow removal activities on NH230 (Jan.)

Elev.: 25 m Kita 1-jo Ave Under-pass (road heating section) Snow Shed & Tunnel Sections Elev.: 835 m Hot Springs Nakayama Pass

Spatial Distribution of HFN Values (%)

January 2013 January 2012 January 2011 January 2010 January 2009

• Jan. 2014
• Jan. 2013
• Jan. 2012
• Jan. 2011
• Jan. 2010
• Jan. 2009
• Jan. 2008

Variation of friction on winter road surface under different weather conditions

Classification of weather conditions

We used HFN values from KP0.9 to KP20.0, measured from predawn (3 a.m.) on weekdays in January 2014. Classification criteria

Daily minimum temperature and snowfall during 12 night-time hours Temperatures thresholds at 0 ºC & -8 ºC1 Snowfall thresholds at 0 cm & 5 cm2

Data from Local Meteorological Observatory were used

6

Classification of weather conditions

1 Winter Road Surface Management Manual (Draft) 2 Handbook for Snow Removal and Snow Control

Temperature-based weather classifications Snowfall-based weather classifications

Non-winter

0°C < daily min. temperature

No snow

Snowfall during 12 night-time hrs.: 0 cm

Normal

• 8°C < daily min. temperature ≤ 0°C

Snow Scant

Snowfall during 12 night-time hrs.: 0 cm ≤ 5cm

Severe

daily min. temperature ≤ -8°C

Heavy

Snowfall during 12 night-time hrs.: >5cm

0.0 20.0 40.0 60.0 80.0 100.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

HF HFN KP

• Jan. 11
• Jan. 22
• Jan. 24
• Jan. 31

HFN values for a specific weather condition (1)

The patterns of variation for HFN values resemble each other for these four days HFN values dropped below 40 at some spots for some days

 Did road surface conditions of the previous day affect those of the next day? 7

Patterns of variation of HFN Values (a) “No snowfall - normal winter day" (daily min. temperature: -8 to 0°C, snowfall during 12 night-time hrs.: 0 cm) HFN

0.0 20.0 40.0 60.0 80.0 100.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

HF HFN KP

• Jan. 21
• Jan. 23
• Jan. 28

HFN values for a specific weather condition (2)

As for Jan 21 & 28, the patterns of variation for HFN values resemble each other Overall, HFN values are lower (around 40) on these days than on non-snowfall normal winter days As for January 23, unlike other days, HFN values exceed 60 from KP14.0 onward

Do weather conditions vary within the section?

8

Patterns of variation of HFN Values (b) "Scant snowfall - normal winter day"

(daily min. temperature: -8 to 0°C, snowfall during 12 night-time hrs. : 0 – 5 cm)

HFN

0.0 20.0 40.0 60.0 80.0 100.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

HF HFN KP

• Jan. 25
• Jan. 30

HFN values for a specific meteorological condition (3)

Only two days fell into the sub-class of “heavy snowfall - normal winter day” The patterns of variation in HFN values are mutually similar Unlike the sub-class of “scant snowfall - normal winter day”, no obvious difference is found between the two days

9

Patterns of variation of HFN Values (c) "Heavy snowfall normal winter day"

(daily min. temperature: -8 to 0°C, snowfall during 12 night-time hrs.: 5 cm or more)

HFN

• 40
• 30
• 20
• 10

10 20 30 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

ΔHFN KP

• Jan. 11
• Jan. 22
• Jan. 24
• Jan. 31

Reproducibility of the pattern of variation of HFN values

10

HFN KP HFN HFNave ΔHFN Patterns of variation of ΔHFN and ΔHFN for four days under “non snowfall - normal winter conditions” at a section of NH230

Based on Shao’s method of making thermal maps of road surface temperature…  First, determine the HFNave, the spatial mean value. Then, calculate the ΔHFN by subtracting the HFNave from each HFN value  The ΔHFN , the mean value of ΔHFN, is determined from each ΔHFN

• f each day that fell into the same weather condition

Reproducibility of patterns of variation in HFN values

Friction maps have distinguishing features. On the red curve in this friction map, values between KP7.6 and KP8.1 (with road heating) are markedly higher than values at other sections On the blue curve in this friction map, values fall significantly at certain points (intersections, bridges, etc.)

11 Heavy snowfall normal winter day No snowfall normal winter day

Friction maps of a “heavy snowfall - normal winter day” and a “no snowfall - normal winter day” on a section of NH230

Downtown Suburban area

Kawazoe Underpass (road heating section) Minami ku-jo (South 9) intersection Kitanosawa Bridge Ishiyama-ohashi Bridge

Avg. Max. Min. ±6 ±12

• Jan. 25

0.0 10.5

• 13.1

80.7 99.5

• Jan. 30

0.0 13.1

• 10.5

80.7 99.5 Mean value 80.7 99.5 E Reproducibility (%) Avg. Max. Min. ±6 ±12

• Jan. 21
• 0.1

16.2

• 17.3

69.3 90.6

• Jan. 23

0.1 24.3

• 18.6

41.1 71.9

• Jan. 28

0.1 14.7

• 18.1

56.3 94.3 Mean value 55.6 85.6 E Reproducibility (%) Avg. Max. Min. ±6 ±12

• Jan. 11

0.1 17.9

• 19.2

67.7 93.8

• Jan. 22

0.1 16.0

• 25.5

70.8 89.1

• Jan. 24
• 0.3

12.3

• 40.8

81.3 97.4

• Jan. 31

0.1 11.7

• 20.5

81.8 96.4 Mean value 75.4 94.1 E Reproducibility (%)

• 40
• 30
• 20
• 10

10 20 30 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

ΔHFN KP

• Jan. 11
• Jan. 22
• Jan. 24
• Jan. 31

Reproducibility of the pattern of variation in HFN values (assessment)

E: the difference between ΔHFN and ΔHFN Scant snow - normal winter day Non-snowfall - normal winter day Heavy snow - normal winter day

Overall accuracy Differences within E±6 (µ0.05) E±12 (µ0.10) ±6 ±12 70.6% 91.8%

Results of the assessment of their reproducibility

Summary and Future Issues

Friction values

Road surface conditions in the winter of 2014 were generally favorable in terms of HFN values, despite severe weather

Due to intensive maintenance work (snow removal activities)?

Patterns of variation in HFN values

The pattern of variation in HFN values varied by weather condition The patterns for different days resembled each other when the weather conditions were same However, no obvious differences of patterns were found between different weather conditions

 Revision of the classification of weather conditions

Reproducibility

To some extent, we were able to quantitatively demonstrate the reproducibility of patterns of variation in HFN values under different weather conditions

More data accumulation and examining the accuracy of reproducibility

13

Thank you for your interest & attention!

14

roberto-1097ga@ceri.go.jp Civil Engineering Research Institute for Cold Region, PWRI, Japan

2016 International Conference & Workshop on Winter Maintenance and Surface Transportation Weather