- Email: [email protected]
Effects of vehicle speeds on the hydrodynamic pressure of pavement surface: Measurement with a designed device
Measurement 98 (2017) 1–9
Contents lists available at ScienceDirect
Measurement journal homepage: www.elsevier.com/locate/measurement
Effects of vehicle speeds on the hydrodynamic pressure of pavement surface: Measurement with a designed device Yong Lei a, Xiaodi Hu b, Hainian Wang a,⇑, Zhanping You c, Yonglian Zhou d, Xu Yang c a
School of Highway, Chang’an University, South Erhuan Road Middle Section, Xi’an, Shanxi 710064, China School of Resource and Civil Engineering, Wuhan Institute of Technology, 693 Xiongchu Avenue, Wuhan, Hubei 430073, China c Department of Civil and Environment Engineering, Michigan Technological University, 1400 Townsend Drive, Houghton, MI 46631, USA d Hubei Jianke International Project Co., Ltd, Hubei Construction Science and Technology Group, East Lake New Technology Development Zone, Wuhan, Hubei 430070, China b
a r t i c l e
i n f o
Article history: Received 2 July 2016 Received in revised form 5 October 2016 Accepted 15 November 2016 Available online 16 November 2016 Keywords: Hydrodynamic pressure Vehicle speeds FBG sensor Pavement surface Directional anisotropy
a b s t r a c t The hydraulic characteristics of pavement, such as pore water pressure, hydrodynamic pressure, and permeability, have a great influence on the functional performance and durability of pavement. The main objective of this study is to design a device for measuring the hydrodynamic pressure of pavement surface and analyze its characteristics with different vehicle speeds. In order to investigate the influence of vehicle speeds on the hydrodynamic pressure of pavement surface, a device equipped with five fiber Bragg grating (FBG) sensors was designed and utilized to measure the surface hydraulic characteristics at four vehicle speeds in the field. The calibration of FBG sensor was conducted by the interrogator (SmartScan Aero) under given pressures from 0 MPa to 1 MPa with an interval of 0.1 MPa. Then, the device was embedded in the pavement and compacted by an auto wheel, which traveled over the device at different speeds. Furthermore, the experimental data was collected by the interrogator during the process of a moving tire touching the surface of the device. The experimental results showed that the correlation coefficient between the given pressure and the center wavelength change of the FBG sensor was 0.99, which means the FBG sensor was accurate and reliable. The hydrodynamic pressure of the pavement surface increased with the increase in vehicle speed from 40 km/h to 100 km/h. Moreover, the directional anisotropy of the hydraulic pressure was found to be dependent on vehicle speed. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Pavement bears the wheel loading subjected from vehicles and provides friction for tires to ensure vehicle movement. In the rainy days, the rain will fall directly on the pavement surface. A certain thickness of water film or layer would be formed due to longterm rain, which has adverse influence on the safety and durability of the pavement. The water film decreases the friction between the tire surface and the pavement surface [1]. It is a great threat to driving safety as braking efficiency and steering precision at high speeds would be weakened by the decrease in friction. The hydrodynamic behavior is produced when the moving tire exerts pressure on the water film. The hydrodynamic pressure leads to splashing and spraying, which has an effect on the visibility of driving and induces traffic accidents [2,3]. Moreover, the splash and spray caused by hydrodynamic pressure would produce noise, and the quality of life of the residents living around the road would ⇑ Corresponding author. E-mail address: [email protected] (H. Wang). http://dx.doi.org/10.1016/j.measurement.2016.11.029 0263-2241/Ó 2016 Elsevier Ltd. All rights reserved.
be affected by the noise. In terms of asphalt pavement, the pore structure of pavement is filled with the invading water and the surface of pavement is covered with a layer of water on a rainy day. The pore hydraulic pressure would be produced when the surface hydraulic pressure infiltrates the pores under the effect of loading on the pavement surface. Gao et al. studied the pore hydraulic pressure in asphalt pavement and its effects on permeability, and found that the pore hydraulic pressure increased with increasing vehicle speed [4]. The decline in the adhesive ability between aggregates and asphalt binder result in stripping and raveling of the pavement surface when asphalt pavement was in saturated conditions for a long time [5,6]. One of the stripping mechanisms is crack propagation in thin asphalt film, and the morphological characteristic of aggregates affects the homogeneity of asphalt film [7]. Hydrodynamic pressure induces stripping. Additionally, the alternation between positive pressure and negative pressure and the anisotropy of hydrodynamic pressure at different orientations promoted the progress of stripping and raveling of asphalt mixture particles, and the durability of the pavement performance was also decreased [8]. Asphalt pavement would be deteriorated further to
2
Y. Lei et al. / Measurement 98 (2017) 1–9
form distresses such as cracks, pits and subsidence under the effect of hydrodynamic pressure, especially in conditions of high vehicle speeds and heavy traffic loads. At present, two methods were utilized to research hydrodynamic pressure. The one is the numerical simulation technology, such as finite element simulation. The permeability coefficient and the hydraulic characteristic of different types of pavement structures were analyzed by building a numerical model. The other one used to investigate the hydrodynamic pressure at the surface and the pores with different vehicle speeds was the experimental field test. Fwa et al. studied the effect of rib tires and pavement grooves on hydroplaning by means of three-dimensional finite element modeling [9–11]. Benedetto et al. built a model to simulate unsteady water flow in the pores of the mixture based on theoretical equations [12]. Kutay et al. investigated the dynamic fluid flow in asphalt pavement by means of building a model [13]. Xue et al. studied the effects of permeability, pavement modulus, and thickness on the pore water pressure of asphalt pavement [14]. However, most present studies on hydrodynamic pressure concentrated on the analysis of the vertical hydrodynamic pressure at different vehicle speeds or indifferent pavement structures. There was little attention paid to the anisotropy characteristic of the hydrodynamic pressure at different orientations. In fact, the anisotropy of surface hydrodynamic pressure has an effect on the orientation of moving water film [15]. Generally, the drifting phenomenon or drifting trend would be produced whilst steering a vehicle at high speeds. Vehicle handling would also become more challenging as the reacting force from the surface hydrodynamic pressure acts on the tire as the vehicle is steered. The speed was one of key factors affecting centrifugation and hydrodynamic behavior. Consequently, it is of great significance to study the effect of anisotropy involved in surface hydrodynamics at different speeds on the safety of driving on rainy days. Generally, the types of Fiber Optic Sensors (FOS) include FBG sensors, interferometric FOS, fiber optic micro bend sensors, distributed sensors, polarimetric sensors and hybrid sensors [16]. According to the different periods of the refractive index modulation, fiber Bragg gratings (FBG) and Long Period Gratings (LPG) were the two considered representative gratings [17,18]. Due to its high efficiency, the FBG sensor has been the most widely used device among them to measure axial strain. In regards of this, the FBG sensor was utilized in this study to measure the hydrodynamic pressure of the pavement surface. Previous studies have stated that temperature has an influence on the characteristic of water and the sensitivity of the FBG sensor [19,20]. It is necessary to measure the hydrodynamic pressure of the pavement surface at a relatively constant temperature. In this study, five specific dimensional FBG sensors were fabricated, and a device that can measure the pressure with five orientations was designed. The experimental results include the hydrodynamic pressure, the change trend and the change rate of pressure, which can be used to analyze the characteristics of the hydrodynamic pressure of the pavement surface. The measured data of the surface hydrodynamic pressure provided accurate parameters for the analysis of numerical simulation, and further improved the precision of numerical simulation. This study laid out a certain foundation for researching the characteristics of hydrodynamic pressure of pavement surface further in the future.
which can measure a slight change in hydrodynamic pressure. The hydrodynamic pressure between the pavement surface and the moving tire has significantly complicated characteristics, negatively impacting the safety and durability of pavement. This research introduced the principle and calibration of the FBG sensor. A device equipped with five FBG sensors was selected to measure the hydrodynamic pressure of pavement surface in different orientations. The vehicle speeds ranged from 40 km/h to 100 km/h in line accordance with the pavement condition on rainy days and relevant legal provisions for motorways in China. The technological process of this research is shown in Fig. 1. 3. Mechanism and methodology 3.1. The Mechanism and calibration of the FBG sensor As shown in Fig. 2, FBG is an optical device that uses a strong ultraviolet laser to burn grating on the center of the fiber and has the function to select a specific wavelength. The light with a specific wavelength would be reflected strongly when it transfers through FBG, while the remaining lights with other wavelengths could transfer through FBG without energy loss. This wavelength can be regarded as the characteristic wavelength of an FBG. The main concept of the FBG sensor is that the characteristic wavelength of an FBG shifts when the stress within the FBG changes [21]. The stress can be back-calculated by detecting the shift in the central wavelength of the FBG. According to the couple-mode theory, the equation involved in the reflection wavelength (kB) of uniform FBG is as follows [22]:
kB ¼ 2ne K
ð1Þ
where ne is the effective refractive index and K is the period of grating. From Eq. (1), it can be seen that the reflection wavelength of FBG is directly proportional to the effective refractive index and the period of FBG. In terms of non-uniform fiber, the period of FBG would be influenced by the shrinkage of the fiber due to temperature. The shift in wavelength of the FBG is determined by the following equation:
DkB ¼ kB ð1 qaÞDn þ kB ða þ nÞDT
where DkB stands for the wavelength shift of FBG, qa is the elasticoptic coefficient, a is the thermal expansion coefficient, n is the calorescence coefficient, Dn is the strain shift and DT is the temperature change.
Literature search Mechanism of FBG sensor
Design of mould
Design of FBG sensor
Fabrication of devicel
Calibration of FBG sensor
Installation of device with sensors
Measurement and analysis of hydrodynamic pressure
2. Research objectives and scope The main objective of this study is to design a device to measure the hydrodynamic pressure of the pavement surface and analyze its characteristics with different vehicle speeds. The FBG sensor employed in this study is an accurate and reliable instrument,
ð2Þ
Characteristic with different orientations
Characteristic with different speeds
Fig. 1. The technological process.
Characteristic with different stages
3
Y. Lei et al. / Measurement 98 (2017) 1–9
Fig. 2. The structure of FBG.
When the temperature is unchanged, Eq. (2) can be simplified as Eq. (3):
DkB ¼ kB ð1 qaÞDn
ð3Þ
Owing to the relatively low sensitivity of bare fiber, this study selected a circular diaphragm to enhance its sensitivity by means of fixing the diaphragm to the surface of the FBG. The slight strain of the diaphragm would be produced when it is pressed by a uniform pressure. The radical strain can be calculated using Eq. (4)
er ¼
3P 2
8h E
ð1 l2 ÞðR2 3r 2 Þ
ð4Þ
where P is the uniform pressure, h is the thickness of the diaphragm, r is the radius of the diaphragm, R is the fixed radius of the diaphragm, l is the Poisson’s ratio of the diaphragm material and E is the elastic modulus of the diaphragm. Fig. 3 shows the distribution of radial strain of circular diaphragm. The maximum radial strain occurs in the center of the circular diaphragm [23]. Consequently, FBG is pasted in the center of the diaphragm to enhance the sensitivity to a greater extent. Then, the radial strain can be calculated via the following Eq. (5):
ec ¼
3P 2
8h E
ð1 l2 ÞR2
ð5Þ
The radial strain can be replaced by strain shift in this study. According to Eqs. (3) and (5), the formula for change in wavelength of FBG is as follows:
DkB ¼ K p P Kp ¼
3kB 2
8h E
ð6Þ 1 l2 R2 ð1 qaÞ
ð7Þ
where Kp is the sensitivity coefficient of the FBG sensor. Eq. (6) indicates that the wavelength of the FBG would increase with increasing pressure. As shown in Fig. 4, the FBG sensor is designed as a cylindrical shape. The sensor’s shell is made of stainless steel, the length of
Fig. 3. The distribution of radial strain.
Fig. 4. The appearance of FBG sensors.
the sensor is 85 mm and the diameter of the sensor is 22 mm. One small hole with a 2.5 mm diameter and four big holes with a 4.5 mm diameter are fabricated on a pressure sensing surface providing an aisle to measure the pressure. The calibration test was conducted in laboratory at the temperature of 25 °C and the moisture content of 50%. The central wavelength of five FBG sensors is recorded under a given pressure from 0 MPa to 1 MPa with an interval of 0.1 MPa. Finally, the results obtained from the interrogator were utilized to verify the reliability and accuracy of the FBG sensors. 3.2. The design of device Aimed at measuring the hydrodynamic pressure of the pavement surface in different orientations, a device equipped with five FBG sensors was designed, as shown in Fig. 5. First of all, a mold composed of boards and pipe was made. Five pipes were set in the mold with different positions and orientations, and a suspended cube made of boards was set in the center of the top surface. The measurement of the suspended cube was 70 mm by 70 mm by 70 mm, and that of the mold was 240 mm by 240 mm by 210 mm. The internal diameter of the pipes was 22 mm. Four pipes were set horizontally and one pipe vertically. The distance between the center point of the pipes and the top surface of the suspended cube was 25 mm. The distance between the pressure sensing face of the horizontal sensors was 25 mm. The device was made of cement concrete. 3.3. The measurement of hydrodynamic pressure After designing the device and calibrating the FBG sensors, the following task was to measure the hydrodynamic pressure of the pavement surface in the field. As shown in Fig. 6, the five FBG sensors equipped in the device were marked as the horizontal front, the horizontal rear, the horizontal left, the horizontal right and the vertical down, respectively. The device was installed in a hole drilled in the surface of the road and adjusted to make sure the top surface of the device was at the same height as the road surface. Then, the interrogator was connected to five FBG sensors after the device was fixed. Subsequently, the suspended cube in the device was filled with water to simulate the saturated condition of the pavement surface on a rainy day. The suspended cube was suspended in water for 30 min to ensure the sealing quality in the interface between the FBG and the pipe, and to allow the water to reach a static state. The interrogator was turned on to measure the hydrodynamic pressure in different orientations when a car wheel passes over the device at speeds of 40 km/h, 60 km/h, 80 km/h and 100 km/h.
4
Y. Lei et al. / Measurement 98 (2017) 1–9
(a) mold
(b) appearance
Fig. 5. The mold and appearance of the device.
(a) schematic diagram
(b) field test
Fig. 6. Schematic diagram for measurement.
In this study, the interrogator (SmartScan Aero) captured the shift in the central wavelength of the sensor at a 400 ls interval. The car used to provide dynamic pressure was a sport utility vehicle, and weighed 1.59 tons. The tire pressure was 0.25 MPa, and the tire specification was 225/65/R17. At a certain speed, the number of controlled clinical trials was three, and the average value of tests was taken as reference data. All of the experimental data was collected by the interrogator, and the hydrodynamic pressures at different vehicle speeds were calculated using Eqs. (6) and (7).
4. Results and analysis 4.1. The calibration of FBG sensors The calibration of the fiber Bragg grating sensor was conducted by the interrogator (SmartScan Aero) under given pressures from 0 MPa to 1 MPa with an interval of 0.1 MPa. The applied pressure range in the calibration is sufficient for the real test. The results from the calibration test are as shown in Table 1. It can be seen that
the correlation coefficient of the five sensors were as high as 0.99, which indicates that the linear correlation between the central wavelength and the pressure was obvious and excellent [24,25]. According to a nonlinear scale and quoted error, the FBG sensor can satisfy the requirements for measuring the pressure accurately. 4.2. The hydrodynamic pressure of pavement surface The hydrodynamic pressure of the pavement surface was obtained from the field test at four vehicle speeds. A pressure characteristic was captured for five orientations based on the designed device. The temperature of water filled in the suspended cube was 25 °C. The hydrodynamic pressure behavior was formed while the film of water on the pavement surface was pressured by the moving tire. Along the direction of vehicle travel, the positive pressure was produced in a partial area in front of a contact point between the tire and pavement, and the negative pressure was produced in a partial area behind the contact point. As shown in Figs. 7–10, the hydrodynamic pressure of the pavement surface presented a
Table 1 The calibration of five FBG sensors. Number
Sensor-1 Sensor-2 Sensor-3 Sensor-4 Sensor-5
Curve-fitting equation
y = 1.677x + 1547.8 y = 1.536x + 1553.4 y = 1.638x + 1536.8 y = 1.714x + 1553.3 y = 1.593x + 1542.2
Kp
1.677 1.536 1.638 1.714 1.593
R2
0.99 0.99 0.99 0.99 0.99
Nonlinear scale (%)
Quoted error (%)
Measure
Require
Measure
Require
0.88 0.90 0.16 0.32 0.65
61 61 61 61 61
1.03 0.99 0.25 0.43 0.87
61.5 61.5 61.5 61.5 61.5
5
120
120
80
80
40
40
Pressure (Kpa)
Pressure (Kpa)
Y. Lei et al. / Measurement 98 (2017) 1–9
0
-40
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
-80
0
-40
-80
-120
-120 456
458
460
462
464
466
468
470
472
474
476
Time (ms)
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down 1164 1166 1168 1170 1172 1174 1176 1178 1180 1182 1184
Time (ms)
Fig. 7. The pressure characteristic at 40 km/h.
Fig. 10. The pressure characteristic at 100 km/h.
120
80
Pressure (Kpa)
40
0
-40
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
-80
-120 790
792
794
796
798
800
802
804
806
808
810
Time (ms) Fig. 8. The pressure characteristic at 60 km/h.
120
80
Pressure (Kpa)
40
0
-40
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
-80
-120 748
750
752
754
756
758
760
762
764
766
768
Time (ms) Fig. 9. The pressure characteristic at 80 km/h.
progress that the positive pressure and the negative pressure occurred cyclically with the effect of vehicle load. The higher the
vehicle speed, the more obvious the hydrodynamic pressure. The phenomenon of hydrodynamic pressure became complicated when vehicle speed increased from 60 km/h to 100 km/h. As shown in Figs. 8–10, positive pressure and the negative pressure occurred cyclically; a similar pressure circulation would be found subsequently. The phenomenon may be attributed to the oscillating effect from the ripple produced by a moving load, and the phenomenon revealed the dissipation of energy [26]. The forming and dissipation of ‘‘decay” positive pressure progressed while negative pressure was unchanged when the vehicle speed was 60 km/h, after hydrodynamic pressure circulation. There was ‘‘decay” hydrodynamic pressure circulation and an increase in the forming and dissipation of ‘‘decay” positive pressure when the vehicle speed was 80 km/h after hydrodynamic pressure circulation. A similar oscillation phenomenon can be seen when the vehicle speed was in 100 km/h. However, the degree of oscillation at 100 km/h was greater than that at 80 km/h, which means that a higher vehicle speed produced a more obvious hydrodynamic pressure. The hydrodynamic pressure dissipated gradually with the oscillation. Specifically, with vehicle loading, the hydrodynamic pressure ended up with a recognizable forming and dissipation of ‘‘decay” positive pressure. The reason hydrodynamic pressure only ends up with positive pressure might be attributed to the effect of gravitational force. Water spread into its surroundings as positive pressure was exerted, while the water converged and gathered inwardly with negative pressure. The water located on the pavement’s surface generated a vertical force due to weight. Water confined in a certain space would accumulate and spread, and positive pressure was formed simultaneously. Consequently, the hydrodynamic pressure of the pavement surface was balanced between the ‘‘decay” positive pressure and the weight. The maximum pressure in different orientations of the pavement surface in the positive stage and in the negative stage can be seen in Figs. 11 and 12, respectively. The maximum pressure at different vehicle speeds can reflect the strength of hydrodynamic pressure [27]. The hydrodynamic pressure in the horizontal left and that in the horizontal right were nearly the same, which reflected the symmetry of the hydrodynamic pressure in certain conditions as well as proved the sensibility and reliability of FBG sensors. In the stage of positive pressure, the maximum hydrodynamic pressure in different orientations of the pavement surface increased with increasing vehicle speeds. Among them, the growth rate of lateral pressure (including pressure in the horizontal left
6
Y. Lei et al. / Measurement 98 (2017) 1–9
Maximum positive pressure (Kpa)
120
90
60
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
30
0 40
60
80
100
Vehicle speed (km/h) Fig. 11. The maximum positive pressure at different speeds.
Maximum negative pressure (Kpa)
0
is traveling straight on saturated pavement surface, the effect of hydrodynamic pressure on driving safety is relatively poor due to the symmetry of lateral pressures. However, the anisotropy of hydrodynamic pressure had an adverse effect on security when steering on the saturated pavement. The energy of hydrodynamic pressure generated by moving a tire at the same speed was constant. But, the occupation proportions of pressures in different orientations were varied. Figs. 13 and 14 show the occupied portions of the maximum pressures in five orientations, in the positive stage and negative stage, respectively. In terms of the positive pressure stage, the maximum pressure in the horizontal rear occupied more than 22% of the total pressure at four vehicle speeds, while that in the horizontal front occupied less than 17% of the total pressure at four vehicle speeds. Taking the symmetry of lateral pressures into consideration, the synthesized orientation of hydrodynamic pressure was substantially below the rear in the positive stage. In the negative stage, all of the absolute values of the maximum pressure in the horizontal rear were less than that in the horizontal front, which indicated that the synthesized orientation of hydrodynamic pressure was above the rear in the negative stage. 4.3. The rate of change of the hydrodynamic pressure In order to analyze the characteristics of hydrodynamic pressure conveniently, the progression of pressure was divided into three stages, the positive pressure stage, the semi-circulation pressure stage and the negative pressure stage, as shown in Fig. 15. In the positive pressure stage, the rate of change of pressure at different orientations is defined as follows:
-30
-60
CRp ¼ the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
-90
Pmp Ps 100% T mp T i
In the semi-circulation pressure stage, the rate of change of pressure at different orientations is defined as follows:
Pmp Pmn 100% T mn T mp
CRci ¼
-120 60
80
100
Vehicle speed (km/h) Fig. 12. The maximum negative pressure at different speeds.
and horizontal right) with increasing vehicle speeds was the fastest in this study. On the basis of hydrodynamic pressure at 40 km/h, the maximum lateral pressure increased 52.8%, 57.1% and 27.1% when the vehicle speeds increased with a gradient of 20 km/h, respectively. The sequence of the maximum positive pressure was the horizontal rear direction, followed by the horizontal lateral, vertical down, and the horizontal front. In the stage of negative pressure, the sequence of the absolute value of the maximum negative pressure at various vehicle speeds, except for the 100 km/h, in a lateral orientation was as follows: the horizontal rear < the horizontal lateral < the vertical down < the horizontal front. This trend was opposite the sequence of maximum positive pressure. The absolute value of the maximum hydrodynamic pressure in different orientations of the pavement surface except for the horizontal front at 100 km/h increased with increasing vehicle speeds. Similar to the positive pressure stage, the growth rate of the lateral pressure with the increasing vehicle speeds was the fastest in this study. On the basis of hydrodynamic pressure at 40 km/h, the absolute value of the maximum lateral pressure increased 196.7%, 15.9% and 25.2% when the vehicle speeds increased with the gradient of 20 km/h, respectively. Taking Figs. 11 and 12 into consideration, the growth rate of the lateral pressure was the fastest in this study. If the vehicle traveling
ð9Þ
In the negative pressure stage, the rate of change of pressure at different orientations is defined as follows:
CRn ¼
Ps Pmn 100% T e T mn
ð10Þ
30
24
the proportion (%)
40
ð8Þ
18
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
12
6 40
60
80
Vehicle speed (km/h) Fig. 13. The proportion of maximum positive pressures.
100
7
Y. Lei et al. / Measurement 98 (2017) 1–9
36 the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
the proportion (%)
30
24
18
12 40
60
80
100
Vehicle speed (km/h) Fig. 14. The proportion of maximum negative pressures.
Fig. 15. The topical process of hydrodynamic pressure.
As shown in Figs. 16 and 18, the rate of change of the maximum hydrodynamic pressure in five orientations increased with increasing vehicle speeds in the positive pressure stage and the semicirculation stage, respectively. In the positive pressure stage, the growth rate of hydrodynamic pressure in the horizontal front was the lowest compared with that of the other orientations, and the subsequent one was the vertical down. In the semicirculation pressure stage, the rate of change of hydrodynamic pressure in five orientations was comparatively equal. As shown in Fig. 17, the variability in change rate of the maximum hydrodynamic pressures at a low vehicle speed of 40 km/h was relatively greater in the negative pressure stage. A decreasing but disordered trend in the rate of change of the maximum hydrodynamic pressure in five orientations can be seen when the vehicle speeds were increased gradually. Subsequently, the oscillation occurred, which indicated the complexity of energy dissipation during the progression of hydrodynamic pressure. As shown in Fig. 19, the coefficient of variation of hydrodynamic pressure at different speeds was utilized to quantitatively analyze the characteristic of pressures in three stages. At a 40 km/h vehicle speed, the coefficients of variation of the rate of change of hydrodynamic pressure in the positive pressure stage and the negative pressure stage were 30.4% and 61.5%, respectively. The relatively high coefficient of variation indicated that the forming and dissipation of hydrodynamic pressure was disorderly. Owing to the decrease in dissipation time of hydrodynamic pressure with increasing vehicle speed [28], the balance between hydrodynamic pressure and gravity needed more time, and the rates of change of hydrodynamic pressure in different orientations had more discrepancy at low vehicle speeds. When the vehicle speeds rose from 60 km/h to 100 km/h, the coefficient of variability of the rate of change of hydrodynamic pressure in three stages progressed with increasing vehicle speeds, which means that the anisotropy characteristic of hydrodynamic pressure of the pavement surface would be more obvious with increasing vehicle speed. At the vehicle speeds of 60 km/h, 80 km/h and100 km/h, the sequence of the coefficient of variability of pressure in the positive pressure stages was the highest, followed by the negative pressure stage and the circulation pressure stage. It indicated that the anisotropy characteristic of the positive pressure stage was more obvious than that of the negative pressure stage.
20
60
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
Change rate of pressure (Kpa/ms)
Change rate of pressure (Kpa/ms)
75
45
30
15
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
16
12
8
4
0
0 40
60
80
Vehicle speed (km/h) Fig. 16. The change rate in positive pressure stage.
100
40
60
80
Vehicle speed (km/h) Fig. 17. The change rate in negative pressure stage.
100
8
Y. Lei et al. / Measurement 98 (2017) 1–9
(4) The vehicle speed had a visible effect on the directional anisotropy of hydrodynamic pressure. When the vehicle speed was high (more than 60 km/h), the directional anisotropy increased with an increase in vehicle speed. However, when the vehicle was low (40 km/h), the directional anisotropy did not follow this trend. More field experiments may be needed in the future to further confirm this finding. The anisotropy of hydrodynamic pressure had an adverse effect on driving safety when the vehicle is steering on saturated pavement.
Change rate of pressure (Kpa/ms)
15
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
12
9
6
6. Future research 3
0 40
60
80
100
Vehicle speed (km/h) Fig. 18. The change rate in semi-circulation pressure stage.
75 the positive pressure stage the negative pressure stage the circulation pressure stage
Coefficient of variation (%)
61.5 60
These results support the use of the FBG sensor to measure the hydrodynamic pressure of the pavement surface due to its reliable accuracy. Study on the measurement of the hydrodynamic pressure on the pavement surface in different orientations enhances the understanding of the characteristics of hydrodynamic behavior and provides a portion of experimental data for numerical simulation. However, the shape of the designed device and the condition of the pavement surface as well as the effect of temperature on hydrodynamic pressure was taken into consideration in this study. The potential interaction among the FBG in different directions may also need to be investigated since they were very close to each other in this study. Therefore, the challenge in measuring the hydrodynamic pressure needs to be researched further. Acknowledgement
45
30.4 30
23.7 15
11.4
14.1
12.8
12 7.5 3.9
12.1
8.8 7
This research is supported by the National Natural Science Foundation of China (NSFC) (No. 51378074, No. 51578075), the Fundamental and Applied Research Project of the Chinese National Transportation Department (2014 319 812 180), and the Special Fund for Basic Scientific Research of Central Colleges, Chang’an University (CHD310821153503). Appendix A. Supplementary material
0 40
60
80
100
Vehicle speed(km/h) Fig. 19. The coefficient of variation at different speeds.
5. Conclusion This study focused on the measurement and analysis of hydrodynamic pressure within the pavement surface under vehicle loads at different speeds. A device equipped with five FBG sensors in different orientations was designed and utilized to measure hydrodynamic pressure. The tests were carried out in field, and the data was analyzed to obtain the characteristics of hydrodynamic pressure by means of an interrogator. The main conclusions of this study can be summarized as follows: (1) The FBG sensor is reliable and accurate for measuring the hydrodynamic pressure of pavement surface in the field. The device, equipped with five FBG sensors, can be used to measure the hydrodynamic pressure in five different orientations of the pavement surface. (2) The hydrodynamic pressure of the pavement surface increased with the increase in vehicle speeds, especially for positive pressures. (3) The maximum hydrodynamic pressures of different orientations are different. The pressure in the horizontal rear orientation was found to be the highest, followed by the horizontal lateral, vertical down, and the horizontal front.
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.measurement. 2016.11.029. References [1] V. Cerezo, M.T. Do, D. Prevost, Friction/water depth relationship – in-situ observations and its integration in tire/road friction models, J. Eng. Tribol. 228 (11) (2014) 1285–1297. [2] Pichaya Rungruangvirojn, Kunnawee Kanitpong, Measurement of visibility loss due to splash and spray: porous, SMA and conventional asphalt pavements, Int. J. Pavement Eng. 11 (6) (2010) 499–510. [3] T.D. Sherard, Suppression of Vehicle Splash and Spray, Training, 730718, 1973. [4] J. Gao, C. Guo, Y. Liu, Measurement of pore water pressure in asphalt pavement and its effects on permeability, Measurement 62 (4) (2015) 81–87. [5] Óscar Javier Reyes-Ortiz, S. Millán-Montejo, Influence of temperature, granulation, and water in the cohesion of asphalt mixtures, J. Surg. Oncol. 13 (2) (2009) 309–324. [6] Y.Q. Tan, M. Guo, Using surface free energy method to study the cohesion and adhesion of asphalt mastic, Constr. Build. Mater. 47 (5) (2013) 254–260. [7] H.N. Wang, Y. Bu, Y.Z. Wang, The effect of morphological characteristic of coarse aggregates measured with fractal dimension on asphalt mixture’s hightemperature performance, Adv. Mater. Sci. Eng. 2016 (1787) (2016) 1–9. [8] X.Z. Cui, X.Z. Cui, Dynamic numerical analysis of antimoisture-damage mechanism of permeable pavement base, Int. J. Geomech. 10 (6) (2010) 230–235. [9] T.F. Fwa, S.S. Kumar, G.P. Ong, Analytical modeling of effects of rib tires on hydroplaning, Transport. Res. Rec. J. Transport. Res. Board 2068 (2068) (2008) 109–118. [10] G.P. Ong, T.F. Fwa, Modeling and analysis of truck hydroplaning on highways, Transport. Res. Rec. J. Transport. Res. Board 2068 (2068) (2008) 99–108. [11] T.F. Fwa, G.P. Ong, Transverse pavement grooving against hydroplaning. II: Design, J. Transport. Eng. 132 (6) (2006) 449–457.
Y. Lei et al. / Measurement 98 (2017) 1–9 [12] A. Benedetto, A. Umiliaco, Evaluation of hydraulic permeability of open-graded asphalt mixes using a full numerical simulation, J. Mater. Civ. Eng. 26 (4) (2014) 599–606. [13] M.E. Kutay, A.H. Aydilek, T. Harman, Dynamic hydraulic conductivity (permeability) of asphalt pavements, in: GeoCongress 2006@sGeotechnical Engineering in the Information Technology Age, ASCE, 2014, pp. 1–6. [14] Q. Xue, L. Liu, Hydraulic-stress coupling effects on dynamic behavior of asphalt pavement structure material, Constr. Build. Mater. 43 (3) (2013) 31–36. [15] H.C. Dan, L.H. He, L.H. Zhao, Coupled hydro-mechanical response of saturated asphalt pavement under moving traffic load, Int. J. Pavement Eng. 16 (2) (2014) 125–143. [16] M. Ramakrishnan, G. Rajan, Y. Semenova, Overview of fiber optic sensor technologies for strain/temperature sensing applications in composite materials, Sensors 16 (1) (2016). [17] L. Alwis, T. Sun, K.V. Grattan, Analysis of polyimide-coated optical fiber longperiod grating-based relative humidity sensor, IEEE Sens. J. 13 (2) (2013) 767– 771. [18] J. Huang, Z. Zhou, D. Zhang, A fiber Bragg grating pressure sensor and its application to pipeline leakage detection, Adv. Mech. Eng. 2013 (2) (2013) 1310–1313. [19] J.H. Yi, Laboratory tests on local damage detection for jacket-type offshore structures using optical FBG sensors based on statistical approaches, Ocean Eng. 124 (2016) 94–103. [20] X. Fan, Q. Jiang, G. Li, Engine cylinder head temperature measurement based on optical fiber probe, J. Test. Eval. 43 (4) (2015) 758–764.
9
[21] Y.L. Zhou, Research on Hydrodynamic Pressure in Asphalt Pavement Based on a Five-Dimensional Optical Fiber Sensors, Wuhan Institute of Technology, 2015 (in Chinese). [22] F. Surre, R.H. Scott, P. Banerji, Study of reliability of fibre Bragg grating fibre optic strain sensors for field-test applications, Sens. Actuators, A Phys. 185 (5) (2012) 8–16. [23] Hao Chen, Simulation and Measurement of the Dynamic Hydraulic Pressure in Pavement Using fiber Bragg Grating Sensor, Nanjing University of Aeronautics and Astronautics, 2009 (in Chinese). [24] J.L.T. Jian-Neng Wang, Feasibility of fiber Bragg grating and long-period fiber grating sensors under different environmental conditions, Sensors 10 (11) (2010) 10105–10127. [25] D.Y. Chen, D.L. Mo, Study of fiber Bragg grating pressure sensor, J. Beijing Univ. Chem. Technol. 34 (3) (2007) 318–320 (in Chinese). [26] X.Z. Cui, Q. Jin, Q.S. Shang, Numerical simulation of dynamic pore pressure in asphalt pavement, J. Southeast Univ. 25 (1) (2009) 79–82 (in Chinese). [27] Markus Oeser, Paul Hovagimian, Ute Kabitzke, Hydraulic and mechanical properties of porous cement-stabilised materials for base courses of PICPs, Int. J. Pavement Eng. 13 (1) (2012) 68–79. [28] H.B. Li, Y.P. Sheng, Study on vehicle speed in pore water pressure of rigid pavement base using poro-elasticity, Appl. Mech. Mater. 178–181 (4) (2012) 2615–2618.
Contents lists available at ScienceDirect
Measurement journal homepage: www.elsevier.com/locate/measurement
Effects of vehicle speeds on the hydrodynamic pressure of pavement surface: Measurement with a designed device Yong Lei a, Xiaodi Hu b, Hainian Wang a,⇑, Zhanping You c, Yonglian Zhou d, Xu Yang c a
School of Highway, Chang’an University, South Erhuan Road Middle Section, Xi’an, Shanxi 710064, China School of Resource and Civil Engineering, Wuhan Institute of Technology, 693 Xiongchu Avenue, Wuhan, Hubei 430073, China c Department of Civil and Environment Engineering, Michigan Technological University, 1400 Townsend Drive, Houghton, MI 46631, USA d Hubei Jianke International Project Co., Ltd, Hubei Construction Science and Technology Group, East Lake New Technology Development Zone, Wuhan, Hubei 430070, China b
a r t i c l e
i n f o
Article history: Received 2 July 2016 Received in revised form 5 October 2016 Accepted 15 November 2016 Available online 16 November 2016 Keywords: Hydrodynamic pressure Vehicle speeds FBG sensor Pavement surface Directional anisotropy
a b s t r a c t The hydraulic characteristics of pavement, such as pore water pressure, hydrodynamic pressure, and permeability, have a great influence on the functional performance and durability of pavement. The main objective of this study is to design a device for measuring the hydrodynamic pressure of pavement surface and analyze its characteristics with different vehicle speeds. In order to investigate the influence of vehicle speeds on the hydrodynamic pressure of pavement surface, a device equipped with five fiber Bragg grating (FBG) sensors was designed and utilized to measure the surface hydraulic characteristics at four vehicle speeds in the field. The calibration of FBG sensor was conducted by the interrogator (SmartScan Aero) under given pressures from 0 MPa to 1 MPa with an interval of 0.1 MPa. Then, the device was embedded in the pavement and compacted by an auto wheel, which traveled over the device at different speeds. Furthermore, the experimental data was collected by the interrogator during the process of a moving tire touching the surface of the device. The experimental results showed that the correlation coefficient between the given pressure and the center wavelength change of the FBG sensor was 0.99, which means the FBG sensor was accurate and reliable. The hydrodynamic pressure of the pavement surface increased with the increase in vehicle speed from 40 km/h to 100 km/h. Moreover, the directional anisotropy of the hydraulic pressure was found to be dependent on vehicle speed. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Pavement bears the wheel loading subjected from vehicles and provides friction for tires to ensure vehicle movement. In the rainy days, the rain will fall directly on the pavement surface. A certain thickness of water film or layer would be formed due to longterm rain, which has adverse influence on the safety and durability of the pavement. The water film decreases the friction between the tire surface and the pavement surface [1]. It is a great threat to driving safety as braking efficiency and steering precision at high speeds would be weakened by the decrease in friction. The hydrodynamic behavior is produced when the moving tire exerts pressure on the water film. The hydrodynamic pressure leads to splashing and spraying, which has an effect on the visibility of driving and induces traffic accidents [2,3]. Moreover, the splash and spray caused by hydrodynamic pressure would produce noise, and the quality of life of the residents living around the road would ⇑ Corresponding author. E-mail address: [email protected] (H. Wang). http://dx.doi.org/10.1016/j.measurement.2016.11.029 0263-2241/Ó 2016 Elsevier Ltd. All rights reserved.
be affected by the noise. In terms of asphalt pavement, the pore structure of pavement is filled with the invading water and the surface of pavement is covered with a layer of water on a rainy day. The pore hydraulic pressure would be produced when the surface hydraulic pressure infiltrates the pores under the effect of loading on the pavement surface. Gao et al. studied the pore hydraulic pressure in asphalt pavement and its effects on permeability, and found that the pore hydraulic pressure increased with increasing vehicle speed [4]. The decline in the adhesive ability between aggregates and asphalt binder result in stripping and raveling of the pavement surface when asphalt pavement was in saturated conditions for a long time [5,6]. One of the stripping mechanisms is crack propagation in thin asphalt film, and the morphological characteristic of aggregates affects the homogeneity of asphalt film [7]. Hydrodynamic pressure induces stripping. Additionally, the alternation between positive pressure and negative pressure and the anisotropy of hydrodynamic pressure at different orientations promoted the progress of stripping and raveling of asphalt mixture particles, and the durability of the pavement performance was also decreased [8]. Asphalt pavement would be deteriorated further to
2
Y. Lei et al. / Measurement 98 (2017) 1–9
form distresses such as cracks, pits and subsidence under the effect of hydrodynamic pressure, especially in conditions of high vehicle speeds and heavy traffic loads. At present, two methods were utilized to research hydrodynamic pressure. The one is the numerical simulation technology, such as finite element simulation. The permeability coefficient and the hydraulic characteristic of different types of pavement structures were analyzed by building a numerical model. The other one used to investigate the hydrodynamic pressure at the surface and the pores with different vehicle speeds was the experimental field test. Fwa et al. studied the effect of rib tires and pavement grooves on hydroplaning by means of three-dimensional finite element modeling [9–11]. Benedetto et al. built a model to simulate unsteady water flow in the pores of the mixture based on theoretical equations [12]. Kutay et al. investigated the dynamic fluid flow in asphalt pavement by means of building a model [13]. Xue et al. studied the effects of permeability, pavement modulus, and thickness on the pore water pressure of asphalt pavement [14]. However, most present studies on hydrodynamic pressure concentrated on the analysis of the vertical hydrodynamic pressure at different vehicle speeds or indifferent pavement structures. There was little attention paid to the anisotropy characteristic of the hydrodynamic pressure at different orientations. In fact, the anisotropy of surface hydrodynamic pressure has an effect on the orientation of moving water film [15]. Generally, the drifting phenomenon or drifting trend would be produced whilst steering a vehicle at high speeds. Vehicle handling would also become more challenging as the reacting force from the surface hydrodynamic pressure acts on the tire as the vehicle is steered. The speed was one of key factors affecting centrifugation and hydrodynamic behavior. Consequently, it is of great significance to study the effect of anisotropy involved in surface hydrodynamics at different speeds on the safety of driving on rainy days. Generally, the types of Fiber Optic Sensors (FOS) include FBG sensors, interferometric FOS, fiber optic micro bend sensors, distributed sensors, polarimetric sensors and hybrid sensors [16]. According to the different periods of the refractive index modulation, fiber Bragg gratings (FBG) and Long Period Gratings (LPG) were the two considered representative gratings [17,18]. Due to its high efficiency, the FBG sensor has been the most widely used device among them to measure axial strain. In regards of this, the FBG sensor was utilized in this study to measure the hydrodynamic pressure of the pavement surface. Previous studies have stated that temperature has an influence on the characteristic of water and the sensitivity of the FBG sensor [19,20]. It is necessary to measure the hydrodynamic pressure of the pavement surface at a relatively constant temperature. In this study, five specific dimensional FBG sensors were fabricated, and a device that can measure the pressure with five orientations was designed. The experimental results include the hydrodynamic pressure, the change trend and the change rate of pressure, which can be used to analyze the characteristics of the hydrodynamic pressure of the pavement surface. The measured data of the surface hydrodynamic pressure provided accurate parameters for the analysis of numerical simulation, and further improved the precision of numerical simulation. This study laid out a certain foundation for researching the characteristics of hydrodynamic pressure of pavement surface further in the future.
which can measure a slight change in hydrodynamic pressure. The hydrodynamic pressure between the pavement surface and the moving tire has significantly complicated characteristics, negatively impacting the safety and durability of pavement. This research introduced the principle and calibration of the FBG sensor. A device equipped with five FBG sensors was selected to measure the hydrodynamic pressure of pavement surface in different orientations. The vehicle speeds ranged from 40 km/h to 100 km/h in line accordance with the pavement condition on rainy days and relevant legal provisions for motorways in China. The technological process of this research is shown in Fig. 1. 3. Mechanism and methodology 3.1. The Mechanism and calibration of the FBG sensor As shown in Fig. 2, FBG is an optical device that uses a strong ultraviolet laser to burn grating on the center of the fiber and has the function to select a specific wavelength. The light with a specific wavelength would be reflected strongly when it transfers through FBG, while the remaining lights with other wavelengths could transfer through FBG without energy loss. This wavelength can be regarded as the characteristic wavelength of an FBG. The main concept of the FBG sensor is that the characteristic wavelength of an FBG shifts when the stress within the FBG changes [21]. The stress can be back-calculated by detecting the shift in the central wavelength of the FBG. According to the couple-mode theory, the equation involved in the reflection wavelength (kB) of uniform FBG is as follows [22]:
kB ¼ 2ne K
ð1Þ
where ne is the effective refractive index and K is the period of grating. From Eq. (1), it can be seen that the reflection wavelength of FBG is directly proportional to the effective refractive index and the period of FBG. In terms of non-uniform fiber, the period of FBG would be influenced by the shrinkage of the fiber due to temperature. The shift in wavelength of the FBG is determined by the following equation:
DkB ¼ kB ð1 qaÞDn þ kB ða þ nÞDT
where DkB stands for the wavelength shift of FBG, qa is the elasticoptic coefficient, a is the thermal expansion coefficient, n is the calorescence coefficient, Dn is the strain shift and DT is the temperature change.
Literature search Mechanism of FBG sensor
Design of mould
Design of FBG sensor
Fabrication of devicel
Calibration of FBG sensor
Installation of device with sensors
Measurement and analysis of hydrodynamic pressure
2. Research objectives and scope The main objective of this study is to design a device to measure the hydrodynamic pressure of the pavement surface and analyze its characteristics with different vehicle speeds. The FBG sensor employed in this study is an accurate and reliable instrument,
ð2Þ
Characteristic with different orientations
Characteristic with different speeds
Fig. 1. The technological process.
Characteristic with different stages
3
Y. Lei et al. / Measurement 98 (2017) 1–9
Fig. 2. The structure of FBG.
When the temperature is unchanged, Eq. (2) can be simplified as Eq. (3):
DkB ¼ kB ð1 qaÞDn
ð3Þ
Owing to the relatively low sensitivity of bare fiber, this study selected a circular diaphragm to enhance its sensitivity by means of fixing the diaphragm to the surface of the FBG. The slight strain of the diaphragm would be produced when it is pressed by a uniform pressure. The radical strain can be calculated using Eq. (4)
er ¼
3P 2
8h E
ð1 l2 ÞðR2 3r 2 Þ
ð4Þ
where P is the uniform pressure, h is the thickness of the diaphragm, r is the radius of the diaphragm, R is the fixed radius of the diaphragm, l is the Poisson’s ratio of the diaphragm material and E is the elastic modulus of the diaphragm. Fig. 3 shows the distribution of radial strain of circular diaphragm. The maximum radial strain occurs in the center of the circular diaphragm [23]. Consequently, FBG is pasted in the center of the diaphragm to enhance the sensitivity to a greater extent. Then, the radial strain can be calculated via the following Eq. (5):
ec ¼
3P 2
8h E
ð1 l2 ÞR2
ð5Þ
The radial strain can be replaced by strain shift in this study. According to Eqs. (3) and (5), the formula for change in wavelength of FBG is as follows:
DkB ¼ K p P Kp ¼
3kB 2
8h E
ð6Þ 1 l2 R2 ð1 qaÞ
ð7Þ
where Kp is the sensitivity coefficient of the FBG sensor. Eq. (6) indicates that the wavelength of the FBG would increase with increasing pressure. As shown in Fig. 4, the FBG sensor is designed as a cylindrical shape. The sensor’s shell is made of stainless steel, the length of
Fig. 3. The distribution of radial strain.
Fig. 4. The appearance of FBG sensors.
the sensor is 85 mm and the diameter of the sensor is 22 mm. One small hole with a 2.5 mm diameter and four big holes with a 4.5 mm diameter are fabricated on a pressure sensing surface providing an aisle to measure the pressure. The calibration test was conducted in laboratory at the temperature of 25 °C and the moisture content of 50%. The central wavelength of five FBG sensors is recorded under a given pressure from 0 MPa to 1 MPa with an interval of 0.1 MPa. Finally, the results obtained from the interrogator were utilized to verify the reliability and accuracy of the FBG sensors. 3.2. The design of device Aimed at measuring the hydrodynamic pressure of the pavement surface in different orientations, a device equipped with five FBG sensors was designed, as shown in Fig. 5. First of all, a mold composed of boards and pipe was made. Five pipes were set in the mold with different positions and orientations, and a suspended cube made of boards was set in the center of the top surface. The measurement of the suspended cube was 70 mm by 70 mm by 70 mm, and that of the mold was 240 mm by 240 mm by 210 mm. The internal diameter of the pipes was 22 mm. Four pipes were set horizontally and one pipe vertically. The distance between the center point of the pipes and the top surface of the suspended cube was 25 mm. The distance between the pressure sensing face of the horizontal sensors was 25 mm. The device was made of cement concrete. 3.3. The measurement of hydrodynamic pressure After designing the device and calibrating the FBG sensors, the following task was to measure the hydrodynamic pressure of the pavement surface in the field. As shown in Fig. 6, the five FBG sensors equipped in the device were marked as the horizontal front, the horizontal rear, the horizontal left, the horizontal right and the vertical down, respectively. The device was installed in a hole drilled in the surface of the road and adjusted to make sure the top surface of the device was at the same height as the road surface. Then, the interrogator was connected to five FBG sensors after the device was fixed. Subsequently, the suspended cube in the device was filled with water to simulate the saturated condition of the pavement surface on a rainy day. The suspended cube was suspended in water for 30 min to ensure the sealing quality in the interface between the FBG and the pipe, and to allow the water to reach a static state. The interrogator was turned on to measure the hydrodynamic pressure in different orientations when a car wheel passes over the device at speeds of 40 km/h, 60 km/h, 80 km/h and 100 km/h.
4
Y. Lei et al. / Measurement 98 (2017) 1–9
(a) mold
(b) appearance
Fig. 5. The mold and appearance of the device.
(a) schematic diagram
(b) field test
Fig. 6. Schematic diagram for measurement.
In this study, the interrogator (SmartScan Aero) captured the shift in the central wavelength of the sensor at a 400 ls interval. The car used to provide dynamic pressure was a sport utility vehicle, and weighed 1.59 tons. The tire pressure was 0.25 MPa, and the tire specification was 225/65/R17. At a certain speed, the number of controlled clinical trials was three, and the average value of tests was taken as reference data. All of the experimental data was collected by the interrogator, and the hydrodynamic pressures at different vehicle speeds were calculated using Eqs. (6) and (7).
4. Results and analysis 4.1. The calibration of FBG sensors The calibration of the fiber Bragg grating sensor was conducted by the interrogator (SmartScan Aero) under given pressures from 0 MPa to 1 MPa with an interval of 0.1 MPa. The applied pressure range in the calibration is sufficient for the real test. The results from the calibration test are as shown in Table 1. It can be seen that
the correlation coefficient of the five sensors were as high as 0.99, which indicates that the linear correlation between the central wavelength and the pressure was obvious and excellent [24,25]. According to a nonlinear scale and quoted error, the FBG sensor can satisfy the requirements for measuring the pressure accurately. 4.2. The hydrodynamic pressure of pavement surface The hydrodynamic pressure of the pavement surface was obtained from the field test at four vehicle speeds. A pressure characteristic was captured for five orientations based on the designed device. The temperature of water filled in the suspended cube was 25 °C. The hydrodynamic pressure behavior was formed while the film of water on the pavement surface was pressured by the moving tire. Along the direction of vehicle travel, the positive pressure was produced in a partial area in front of a contact point between the tire and pavement, and the negative pressure was produced in a partial area behind the contact point. As shown in Figs. 7–10, the hydrodynamic pressure of the pavement surface presented a
Table 1 The calibration of five FBG sensors. Number
Sensor-1 Sensor-2 Sensor-3 Sensor-4 Sensor-5
Curve-fitting equation
y = 1.677x + 1547.8 y = 1.536x + 1553.4 y = 1.638x + 1536.8 y = 1.714x + 1553.3 y = 1.593x + 1542.2
Kp
1.677 1.536 1.638 1.714 1.593
R2
0.99 0.99 0.99 0.99 0.99
Nonlinear scale (%)
Quoted error (%)
Measure
Require
Measure
Require
0.88 0.90 0.16 0.32 0.65
61 61 61 61 61
1.03 0.99 0.25 0.43 0.87
61.5 61.5 61.5 61.5 61.5
5
120
120
80
80
40
40
Pressure (Kpa)
Pressure (Kpa)
Y. Lei et al. / Measurement 98 (2017) 1–9
0
-40
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
-80
0
-40
-80
-120
-120 456
458
460
462
464
466
468
470
472
474
476
Time (ms)
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down 1164 1166 1168 1170 1172 1174 1176 1178 1180 1182 1184
Time (ms)
Fig. 7. The pressure characteristic at 40 km/h.
Fig. 10. The pressure characteristic at 100 km/h.
120
80
Pressure (Kpa)
40
0
-40
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
-80
-120 790
792
794
796
798
800
802
804
806
808
810
Time (ms) Fig. 8. The pressure characteristic at 60 km/h.
120
80
Pressure (Kpa)
40
0
-40
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
-80
-120 748
750
752
754
756
758
760
762
764
766
768
Time (ms) Fig. 9. The pressure characteristic at 80 km/h.
progress that the positive pressure and the negative pressure occurred cyclically with the effect of vehicle load. The higher the
vehicle speed, the more obvious the hydrodynamic pressure. The phenomenon of hydrodynamic pressure became complicated when vehicle speed increased from 60 km/h to 100 km/h. As shown in Figs. 8–10, positive pressure and the negative pressure occurred cyclically; a similar pressure circulation would be found subsequently. The phenomenon may be attributed to the oscillating effect from the ripple produced by a moving load, and the phenomenon revealed the dissipation of energy [26]. The forming and dissipation of ‘‘decay” positive pressure progressed while negative pressure was unchanged when the vehicle speed was 60 km/h, after hydrodynamic pressure circulation. There was ‘‘decay” hydrodynamic pressure circulation and an increase in the forming and dissipation of ‘‘decay” positive pressure when the vehicle speed was 80 km/h after hydrodynamic pressure circulation. A similar oscillation phenomenon can be seen when the vehicle speed was in 100 km/h. However, the degree of oscillation at 100 km/h was greater than that at 80 km/h, which means that a higher vehicle speed produced a more obvious hydrodynamic pressure. The hydrodynamic pressure dissipated gradually with the oscillation. Specifically, with vehicle loading, the hydrodynamic pressure ended up with a recognizable forming and dissipation of ‘‘decay” positive pressure. The reason hydrodynamic pressure only ends up with positive pressure might be attributed to the effect of gravitational force. Water spread into its surroundings as positive pressure was exerted, while the water converged and gathered inwardly with negative pressure. The water located on the pavement’s surface generated a vertical force due to weight. Water confined in a certain space would accumulate and spread, and positive pressure was formed simultaneously. Consequently, the hydrodynamic pressure of the pavement surface was balanced between the ‘‘decay” positive pressure and the weight. The maximum pressure in different orientations of the pavement surface in the positive stage and in the negative stage can be seen in Figs. 11 and 12, respectively. The maximum pressure at different vehicle speeds can reflect the strength of hydrodynamic pressure [27]. The hydrodynamic pressure in the horizontal left and that in the horizontal right were nearly the same, which reflected the symmetry of the hydrodynamic pressure in certain conditions as well as proved the sensibility and reliability of FBG sensors. In the stage of positive pressure, the maximum hydrodynamic pressure in different orientations of the pavement surface increased with increasing vehicle speeds. Among them, the growth rate of lateral pressure (including pressure in the horizontal left
6
Y. Lei et al. / Measurement 98 (2017) 1–9
Maximum positive pressure (Kpa)
120
90
60
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
30
0 40
60
80
100
Vehicle speed (km/h) Fig. 11. The maximum positive pressure at different speeds.
Maximum negative pressure (Kpa)
0
is traveling straight on saturated pavement surface, the effect of hydrodynamic pressure on driving safety is relatively poor due to the symmetry of lateral pressures. However, the anisotropy of hydrodynamic pressure had an adverse effect on security when steering on the saturated pavement. The energy of hydrodynamic pressure generated by moving a tire at the same speed was constant. But, the occupation proportions of pressures in different orientations were varied. Figs. 13 and 14 show the occupied portions of the maximum pressures in five orientations, in the positive stage and negative stage, respectively. In terms of the positive pressure stage, the maximum pressure in the horizontal rear occupied more than 22% of the total pressure at four vehicle speeds, while that in the horizontal front occupied less than 17% of the total pressure at four vehicle speeds. Taking the symmetry of lateral pressures into consideration, the synthesized orientation of hydrodynamic pressure was substantially below the rear in the positive stage. In the negative stage, all of the absolute values of the maximum pressure in the horizontal rear were less than that in the horizontal front, which indicated that the synthesized orientation of hydrodynamic pressure was above the rear in the negative stage. 4.3. The rate of change of the hydrodynamic pressure In order to analyze the characteristics of hydrodynamic pressure conveniently, the progression of pressure was divided into three stages, the positive pressure stage, the semi-circulation pressure stage and the negative pressure stage, as shown in Fig. 15. In the positive pressure stage, the rate of change of pressure at different orientations is defined as follows:
-30
-60
CRp ¼ the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
-90
Pmp Ps 100% T mp T i
In the semi-circulation pressure stage, the rate of change of pressure at different orientations is defined as follows:
Pmp Pmn 100% T mn T mp
CRci ¼
-120 60
80
100
Vehicle speed (km/h) Fig. 12. The maximum negative pressure at different speeds.
and horizontal right) with increasing vehicle speeds was the fastest in this study. On the basis of hydrodynamic pressure at 40 km/h, the maximum lateral pressure increased 52.8%, 57.1% and 27.1% when the vehicle speeds increased with a gradient of 20 km/h, respectively. The sequence of the maximum positive pressure was the horizontal rear direction, followed by the horizontal lateral, vertical down, and the horizontal front. In the stage of negative pressure, the sequence of the absolute value of the maximum negative pressure at various vehicle speeds, except for the 100 km/h, in a lateral orientation was as follows: the horizontal rear < the horizontal lateral < the vertical down < the horizontal front. This trend was opposite the sequence of maximum positive pressure. The absolute value of the maximum hydrodynamic pressure in different orientations of the pavement surface except for the horizontal front at 100 km/h increased with increasing vehicle speeds. Similar to the positive pressure stage, the growth rate of the lateral pressure with the increasing vehicle speeds was the fastest in this study. On the basis of hydrodynamic pressure at 40 km/h, the absolute value of the maximum lateral pressure increased 196.7%, 15.9% and 25.2% when the vehicle speeds increased with the gradient of 20 km/h, respectively. Taking Figs. 11 and 12 into consideration, the growth rate of the lateral pressure was the fastest in this study. If the vehicle traveling
ð9Þ
In the negative pressure stage, the rate of change of pressure at different orientations is defined as follows:
CRn ¼
Ps Pmn 100% T e T mn
ð10Þ
30
24
the proportion (%)
40
ð8Þ
18
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
12
6 40
60
80
Vehicle speed (km/h) Fig. 13. The proportion of maximum positive pressures.
100
7
Y. Lei et al. / Measurement 98 (2017) 1–9
36 the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
the proportion (%)
30
24
18
12 40
60
80
100
Vehicle speed (km/h) Fig. 14. The proportion of maximum negative pressures.
Fig. 15. The topical process of hydrodynamic pressure.
As shown in Figs. 16 and 18, the rate of change of the maximum hydrodynamic pressure in five orientations increased with increasing vehicle speeds in the positive pressure stage and the semicirculation stage, respectively. In the positive pressure stage, the growth rate of hydrodynamic pressure in the horizontal front was the lowest compared with that of the other orientations, and the subsequent one was the vertical down. In the semicirculation pressure stage, the rate of change of hydrodynamic pressure in five orientations was comparatively equal. As shown in Fig. 17, the variability in change rate of the maximum hydrodynamic pressures at a low vehicle speed of 40 km/h was relatively greater in the negative pressure stage. A decreasing but disordered trend in the rate of change of the maximum hydrodynamic pressure in five orientations can be seen when the vehicle speeds were increased gradually. Subsequently, the oscillation occurred, which indicated the complexity of energy dissipation during the progression of hydrodynamic pressure. As shown in Fig. 19, the coefficient of variation of hydrodynamic pressure at different speeds was utilized to quantitatively analyze the characteristic of pressures in three stages. At a 40 km/h vehicle speed, the coefficients of variation of the rate of change of hydrodynamic pressure in the positive pressure stage and the negative pressure stage were 30.4% and 61.5%, respectively. The relatively high coefficient of variation indicated that the forming and dissipation of hydrodynamic pressure was disorderly. Owing to the decrease in dissipation time of hydrodynamic pressure with increasing vehicle speed [28], the balance between hydrodynamic pressure and gravity needed more time, and the rates of change of hydrodynamic pressure in different orientations had more discrepancy at low vehicle speeds. When the vehicle speeds rose from 60 km/h to 100 km/h, the coefficient of variability of the rate of change of hydrodynamic pressure in three stages progressed with increasing vehicle speeds, which means that the anisotropy characteristic of hydrodynamic pressure of the pavement surface would be more obvious with increasing vehicle speed. At the vehicle speeds of 60 km/h, 80 km/h and100 km/h, the sequence of the coefficient of variability of pressure in the positive pressure stages was the highest, followed by the negative pressure stage and the circulation pressure stage. It indicated that the anisotropy characteristic of the positive pressure stage was more obvious than that of the negative pressure stage.
20
60
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
Change rate of pressure (Kpa/ms)
Change rate of pressure (Kpa/ms)
75
45
30
15
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
16
12
8
4
0
0 40
60
80
Vehicle speed (km/h) Fig. 16. The change rate in positive pressure stage.
100
40
60
80
Vehicle speed (km/h) Fig. 17. The change rate in negative pressure stage.
100
8
Y. Lei et al. / Measurement 98 (2017) 1–9
(4) The vehicle speed had a visible effect on the directional anisotropy of hydrodynamic pressure. When the vehicle speed was high (more than 60 km/h), the directional anisotropy increased with an increase in vehicle speed. However, when the vehicle was low (40 km/h), the directional anisotropy did not follow this trend. More field experiments may be needed in the future to further confirm this finding. The anisotropy of hydrodynamic pressure had an adverse effect on driving safety when the vehicle is steering on saturated pavement.
Change rate of pressure (Kpa/ms)
15
the horizontal front the horizontal rear the horizontal left the horizontal right the vertical down
12
9
6
6. Future research 3
0 40
60
80
100
Vehicle speed (km/h) Fig. 18. The change rate in semi-circulation pressure stage.
75 the positive pressure stage the negative pressure stage the circulation pressure stage
Coefficient of variation (%)
61.5 60
These results support the use of the FBG sensor to measure the hydrodynamic pressure of the pavement surface due to its reliable accuracy. Study on the measurement of the hydrodynamic pressure on the pavement surface in different orientations enhances the understanding of the characteristics of hydrodynamic behavior and provides a portion of experimental data for numerical simulation. However, the shape of the designed device and the condition of the pavement surface as well as the effect of temperature on hydrodynamic pressure was taken into consideration in this study. The potential interaction among the FBG in different directions may also need to be investigated since they were very close to each other in this study. Therefore, the challenge in measuring the hydrodynamic pressure needs to be researched further. Acknowledgement
45
30.4 30
23.7 15
11.4
14.1
12.8
12 7.5 3.9
12.1
8.8 7
This research is supported by the National Natural Science Foundation of China (NSFC) (No. 51378074, No. 51578075), the Fundamental and Applied Research Project of the Chinese National Transportation Department (2014 319 812 180), and the Special Fund for Basic Scientific Research of Central Colleges, Chang’an University (CHD310821153503). Appendix A. Supplementary material
0 40
60
80
100
Vehicle speed(km/h) Fig. 19. The coefficient of variation at different speeds.
5. Conclusion This study focused on the measurement and analysis of hydrodynamic pressure within the pavement surface under vehicle loads at different speeds. A device equipped with five FBG sensors in different orientations was designed and utilized to measure hydrodynamic pressure. The tests were carried out in field, and the data was analyzed to obtain the characteristics of hydrodynamic pressure by means of an interrogator. The main conclusions of this study can be summarized as follows: (1) The FBG sensor is reliable and accurate for measuring the hydrodynamic pressure of pavement surface in the field. The device, equipped with five FBG sensors, can be used to measure the hydrodynamic pressure in five different orientations of the pavement surface. (2) The hydrodynamic pressure of the pavement surface increased with the increase in vehicle speeds, especially for positive pressures. (3) The maximum hydrodynamic pressures of different orientations are different. The pressure in the horizontal rear orientation was found to be the highest, followed by the horizontal lateral, vertical down, and the horizontal front.
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.measurement. 2016.11.029. References [1] V. Cerezo, M.T. Do, D. Prevost, Friction/water depth relationship – in-situ observations and its integration in tire/road friction models, J. Eng. Tribol. 228 (11) (2014) 1285–1297. [2] Pichaya Rungruangvirojn, Kunnawee Kanitpong, Measurement of visibility loss due to splash and spray: porous, SMA and conventional asphalt pavements, Int. J. Pavement Eng. 11 (6) (2010) 499–510. [3] T.D. Sherard, Suppression of Vehicle Splash and Spray, Training, 730718, 1973. [4] J. Gao, C. Guo, Y. Liu, Measurement of pore water pressure in asphalt pavement and its effects on permeability, Measurement 62 (4) (2015) 81–87. [5] Óscar Javier Reyes-Ortiz, S. Millán-Montejo, Influence of temperature, granulation, and water in the cohesion of asphalt mixtures, J. Surg. Oncol. 13 (2) (2009) 309–324. [6] Y.Q. Tan, M. Guo, Using surface free energy method to study the cohesion and adhesion of asphalt mastic, Constr. Build. Mater. 47 (5) (2013) 254–260. [7] H.N. Wang, Y. Bu, Y.Z. Wang, The effect of morphological characteristic of coarse aggregates measured with fractal dimension on asphalt mixture’s hightemperature performance, Adv. Mater. Sci. Eng. 2016 (1787) (2016) 1–9. [8] X.Z. Cui, X.Z. Cui, Dynamic numerical analysis of antimoisture-damage mechanism of permeable pavement base, Int. J. Geomech. 10 (6) (2010) 230–235. [9] T.F. Fwa, S.S. Kumar, G.P. Ong, Analytical modeling of effects of rib tires on hydroplaning, Transport. Res. Rec. J. Transport. Res. Board 2068 (2068) (2008) 109–118. [10] G.P. Ong, T.F. Fwa, Modeling and analysis of truck hydroplaning on highways, Transport. Res. Rec. J. Transport. Res. Board 2068 (2068) (2008) 99–108. [11] T.F. Fwa, G.P. Ong, Transverse pavement grooving against hydroplaning. II: Design, J. Transport. Eng. 132 (6) (2006) 449–457.
Y. Lei et al. / Measurement 98 (2017) 1–9 [12] A. Benedetto, A. Umiliaco, Evaluation of hydraulic permeability of open-graded asphalt mixes using a full numerical simulation, J. Mater. Civ. Eng. 26 (4) (2014) 599–606. [13] M.E. Kutay, A.H. Aydilek, T. Harman, Dynamic hydraulic conductivity (permeability) of asphalt pavements, in: GeoCongress 2006@sGeotechnical Engineering in the Information Technology Age, ASCE, 2014, pp. 1–6. [14] Q. Xue, L. Liu, Hydraulic-stress coupling effects on dynamic behavior of asphalt pavement structure material, Constr. Build. Mater. 43 (3) (2013) 31–36. [15] H.C. Dan, L.H. He, L.H. Zhao, Coupled hydro-mechanical response of saturated asphalt pavement under moving traffic load, Int. J. Pavement Eng. 16 (2) (2014) 125–143. [16] M. Ramakrishnan, G. Rajan, Y. Semenova, Overview of fiber optic sensor technologies for strain/temperature sensing applications in composite materials, Sensors 16 (1) (2016). [17] L. Alwis, T. Sun, K.V. Grattan, Analysis of polyimide-coated optical fiber longperiod grating-based relative humidity sensor, IEEE Sens. J. 13 (2) (2013) 767– 771. [18] J. Huang, Z. Zhou, D. Zhang, A fiber Bragg grating pressure sensor and its application to pipeline leakage detection, Adv. Mech. Eng. 2013 (2) (2013) 1310–1313. [19] J.H. Yi, Laboratory tests on local damage detection for jacket-type offshore structures using optical FBG sensors based on statistical approaches, Ocean Eng. 124 (2016) 94–103. [20] X. Fan, Q. Jiang, G. Li, Engine cylinder head temperature measurement based on optical fiber probe, J. Test. Eval. 43 (4) (2015) 758–764.
9
[21] Y.L. Zhou, Research on Hydrodynamic Pressure in Asphalt Pavement Based on a Five-Dimensional Optical Fiber Sensors, Wuhan Institute of Technology, 2015 (in Chinese). [22] F. Surre, R.H. Scott, P. Banerji, Study of reliability of fibre Bragg grating fibre optic strain sensors for field-test applications, Sens. Actuators, A Phys. 185 (5) (2012) 8–16. [23] Hao Chen, Simulation and Measurement of the Dynamic Hydraulic Pressure in Pavement Using fiber Bragg Grating Sensor, Nanjing University of Aeronautics and Astronautics, 2009 (in Chinese). [24] J.L.T. Jian-Neng Wang, Feasibility of fiber Bragg grating and long-period fiber grating sensors under different environmental conditions, Sensors 10 (11) (2010) 10105–10127. [25] D.Y. Chen, D.L. Mo, Study of fiber Bragg grating pressure sensor, J. Beijing Univ. Chem. Technol. 34 (3) (2007) 318–320 (in Chinese). [26] X.Z. Cui, Q. Jin, Q.S. Shang, Numerical simulation of dynamic pore pressure in asphalt pavement, J. Southeast Univ. 25 (1) (2009) 79–82 (in Chinese). [27] Markus Oeser, Paul Hovagimian, Ute Kabitzke, Hydraulic and mechanical properties of porous cement-stabilised materials for base courses of PICPs, Int. J. Pavement Eng. 13 (1) (2012) 68–79. [28] H.B. Li, Y.P. Sheng, Study on vehicle speed in pore water pressure of rigid pavement base using poro-elasticity, Appl. Mech. Mater. 178–181 (4) (2012) 2615–2618.