交通运输工程学报

Evaluation of lateral flow deformation of asphalt mixture based on FBG

GAO Jun-qi, WEI Lu-nan, HOU Yan-ming

Abstract: In order to analyze the lateral flow deformation of asphalt mixture, a rutting test of asphalt mixture was conducted, and the law of lateral strain on the surface of asphalt mixture was researched by using the fiber Bragg grating (FBG) sensor installed on the slab surface of asphalt mixture. With the maximum strain and the absolute value of lateral strain rate in creep stability stage as evaluation indexes, the lateral flow deformation of asphalt mixture was analyzed. Analysis result indicates that the lateral flow deformation decreases with the decreases of both the maximum strain of asphalt mixture and the absolute value of lateral strain rate. The lateral flow deformation develops continuously under the action of cyclic loading. The nearer the test point to the wheel, the heavier its flow deformation is. When the rubber powder contents are 0, 15% and 18%, the lateral strain rates at test point with a distance of 63 mm to the wheel are 6.8×10^-6, 4.0×10^-7 and 6.4×10^-6 min^-1, respectively. Therefore, the asphalt mixture with 15% rubber powder content has larger capacity to resist lateral flow deformation in high temperature. For the asphalt mixtures with 15% rubber powder content, when their aggregate gradations are selected as coarse gradation of AC-13 and fine gradation of AC-13, the lateral strain rates at test point with a distance of 28 mm to the wheel are 6.0×10^-7 and 7.7×10^-6 min^-1, respectively, so the anti-lateral flow deformation capability of AC-13 coarse-graded mixture in high temperature is better than that of AC-13 fine-graded mixture. The maximum strain of rubber powder modified asphalt mixture is 1.96×10^-4, but for the asphalt mixture modified by rubber powder and rutting resistance additive, the value is only 1.22×10^-4, which shows that under the condition of high temperature, the asphalt mixture modified by rubber powder and rutting resistance additive has higher overall structural strength and can bear the direct effect from the wheel load without lateral movement to both sides, which could cause larger lateral flow deformation. The evaluation of lateral flow deformation to asphalt mixture based on the FBG lateral strain can well illustrate the effect of different material and gradation characteristics on the lateral flow deformation of asphalt pavement.More>

2019, 19(4): 1-11. doi: 10.19818/j.cnki.1671-1637.2019.04.001

FullText HTML (110) PDF (1433KB) (1361)

Mechanical property of new type of prefabricated inverted T-shape voided slab bridge

WU Qing-xiong, HUANG Wan-kun, WANG Qu, CHEN Kang-ming, CHEN Bao-chun

Abstract: To solve the disease of hinged joint in the existed prefabricated voided slab bridges, a new type of prefabricated inverted T-shape voided slab bridge was presented. The full-scale model experiment and the nonlinear finite element analysis on the inverted T-shape voided slab bridge with 8 m span were conducted. The stress, deflection and crack distributions of each component of inverted T-shape voided slab bridge subjected to the vehicle load were analyzed. The force mechanism and failure mode of inverted T-shape voided slab bridge were obtained. The mechanical behaviors of inverted T-shape voided slab bridge and voided slab bridge with gate-type steel bars were compared. The effectiveness of inverted T-shape voided slab solving the problem of cracking in hinged joint was verified. Research result shows that the failure process of inverted T-shape voided slab bridge can be divided into the elastic stage, cracking stage of voided slab, cracking stage of concrete in the field-cast structure layer, and yielding stage of tensile steel bars and steel plates. The inverted T-shape voided slab bridge have a good overall mechanical property, its ultimate capacity is 1.4 times of that of voided slab bridge with gate-type steel bars. The concrete in the tensile area above the Ω-shape steel plate is a weak part, for its tensile stress firstly reaching the limit 3.17 MPa. Due to the existences of Ω-shape and L-shape steel plates, when the concrete in the field-cast structure layer cracks, the normal and tangential bonding stresses of each junction surface as high as the field-cast structure layer will not exceed their limit 2.30 and 0.29 MPa, respectively, avoiding the bonding failure on the junction surface. Comparing with the mechanical property of voided slab bridge with gate-type steel bars, the inverted T-shape voided slab structure does not reduce the cracking load of voided slab, and the junction surface between the old and new concretes cracks after the cracking of voided slab. It can fundamentally solve the problem that the hinged joint cracks before the cracking of voided slab under the vehicle load.More>

2019, 19(4): 12-23. doi: 10.19818/j.cnki.1671-1637.2019.04.002

FullText HTML (116) PDF (4749KB) (1344)

Solar radiation time-varying temperature field and temperature effect on small radius curved rigid frame box girder bridge

SHENG Xing-wang, ZHENG Wei-qi, ZHU Zhi-hui, YANG Ying, LI Shuai

Abstract: The real time position of the sun was determined by using the theory of solar physics. Combining with the ray tracing algorithm, the time-varying light surfaces of structure were selected in real time, and the time-varying thermal boundary conditions of structure were obtained. Taking a small radius curved rigid frame box girder bridge in Shijiazhai Interchange of Yongshun to Jishou Expressway as an engineering background, a summer day with the highest temperature was chosen by referring to the local historical meteorological data, and the finite element simulation of three-dimensional transient solar radiation time-varying temperature field of small raduis curved rigid frame box girder bridge was realized with the consideration of solar radiation, long wave radiation, convective heat transfer, wind speed and other environmental conditions. The solar radiation time-varying temperature effect on the small radius curved rigid frame box girder bridge was obtained through the thermal-structural coupling analysis. Research result shows that under the action of time-varying solar radiation, due to the covering effect of flange plate of small radius curved rigid frame box girder bridge, the direct sunlight times of box girder webs are different. The maximum temperature difference at the webs of each section of box girder is 1.3 ℃. The vertical temperature gradient variation law of roof of small radius curved rigid frame box girder bridge is similar to that in the General Specifications for Design of Highway Bridges and Culverts (JTG D60—2015). The maximum temperature difference between the upper and lower surfaces of roof is 14.3 ℃, and the temperature change of lower surface of box girder roof is about 3 h behind that of upper surface of box girder roof. The maximum transverse tensile stress at the lower surface of small radius curved rigid frame box girder bridge roof appears to be 3.13 MPa, and the transverse tensile stresses of upper surface of roof and the outer surface of web also appear to be more than 2 MPa. The displacement change trend of girder end opposites to that of mid-span of small radius curved rigid frame box girder bridge, preliminary revealing a serpentine movement law of small radius curved rigid frame box girder bridge under the action of time-varying solar radiation.More>

2019, 19(4): 24-34. doi: 10.19818/j.cnki.1671-1637.2019.04.003

FullText HTML (186) PDF (4541KB) (1268)

Effect of web hole on vibration and noise of rail transit box girder

ZHANG Tian-qi, LUO Yan-yun, ZHOU Li

Abstract: The vibration and near field noise of each plate of a simply-supported box girder in the urban rail transit were measured, and the relationship between the vibration radiation noise of box girder structure and the box girder vibration was studied by combining with the acoustic radiation theory of plate. Based on the low-frequency characteristics of easily generating diffraction of box girder structure noise, the structure noise change situations of rectangular concrete plates under different opening conditions were calculated. On the basis of considering the web holes of box girder, the vehicle-track-box girder coupled finite element model and the box girder vibration-structure noise finite element-infinite element model were built. The vibration and structure radiation noise change situations of each plate before and after the opening of box girder web were analyzed. Research result shows that the acoustic radiation efficiency of box girder plate does not show a linear relationship with the increase of frequency. The near field low-frequency (less than 250 Hz) radiation noise of each box girder plate is not a simple linear relationship with the structure vibration acceleration level. The noise radiated by the box girder is determined by the vibration of box girder and the acoustic radiation efficiency of each box girder plate together. For an opening plate simply supported at both ends, when the opening ratio is basically the same (about 0.4%), the smaller the diameter of opening hole, the smaller the sound pressure level of plate vibration radiation noise is. Using the finite element-infinite element method to simulate the near field low-frequency noise of box girder structure can solve the influence of sound field boundary when using the finite element method only, and can avoid the inconvenience caused by the cross-use of multi-software in the finite element-boundary element method. Although the web hole increases the vibration response of box girder plate at some frequencies (100-125 Hz), the acoustic radiation efficiency of plate and the noise in corresponding frequency band range reduces, due to the connection of inside and external sound fields of box girder increasing the acoustic short circuit effect. After the opening of web, the total sound pressure levels near the top plate, bottom plate and web plate in the frequency band of 1-250 Hz reduce by 9.43, 2.74 and 1.63 dB, respectively. Thus, the box girder structure noise is controlled.More>

2019, 19(4): 35-46. doi: 10.19818/j.cnki.1671-1637.2019.04.004

FullText HTML (102) PDF (3028KB) (1226)

Simplified impact force model of pier under heavy vehicle collision

ZHAO Wu-chao, QIAN Jiang, WANG Juan

Abstract: To explore the safety of reinforced concrete piers under heavy vehicle collision, a refined finite element (FE) model of heavy vehicle-pier collision was established. The effects of impact velocity, pier diameter, superstructure boundary condition and cargo height on the failure mode and internal force distribution of pier were investigated. The characteristics of vehicle impact force under different conditions were analyzed, and a simplified impact force model was proposed based on the dissipation characteristics of vehicle initial kinetic energy. Analysis result shows that the heavy vehicle collision process can be divided into three stages such as the bumper, engine and cargo collide the pier. The impact force is mainly concentrated at the elevation of 0.9 m in the first two collision stages, and is distributed at the elevation of 2.7 m in the third collision stage. Under the heavy vehicle collision, not only will there be serious damage at the end of pier, but also there may be serious local punching shear damage near the collision site. Due to the neglection to the dynamic effect of impact load and the coupling effect of vehicle and pier, the equivalent static design approach recommended by the General Code for Design of Highway Bridges and Culverts (JTG D60—2015) is difficult to obtain an actual impact response of pier. The impact velocity has the most significant influence on the internal force and impact force of pier. The difference of cargo height changes the spatial distribution of impact force, but will not affect the maximum internal force response of pier. A threshold of 6.5 MJ exists in the initial kinetic energy of heavy vehicle. When the initial kinetic energy is less than the threshold, the impact actions from the vehicle engine and bumper play dominant roles in the dynamic response of pier. On the contrary, the impact action of cargo at the back determines the peak impact force. The relative error of the predicted maximum internal force responses of pier between the simplified impact force model and the refined vehicle model is less than 8%, and the computation time shortens from 6-7 h to 4 min.More>

2019, 19(4): 47-58. doi: 10.19818/j.cnki.1671-1637.2019.04.005

FullText HTML (129) PDF (6268KB) (1344)

Vibration characteristics of metro depot upper building under double vibration source excitation

FENG Qing-song, WANG Zi-yu, LIU Quan-min, LUO Xin-wei, LI Ji-yang

Abstract: Taking a certain metro depot in Guangzhou as the research object, the accelerations of ground vibration caused by the testing line and the repairing line in the depot were measured, and the attenuation laws and differences between the two kinds of vibration sources were analyzed. The finite element model of metro depot upper building was established, and the measured ground vibration data were processed by multi-support excitation of large mass method. The influence of double vibration source excitation on the vibration of upper building floor was analyzed. Analysis result shows that when the train passes through, the main frequency range of ground vibration of testing line is 60-80 Hz, and that of repairing line is 20-40 Hz. The vibration source strength of testing line is greater than that of repairing line, which is about 6 dB. The vibration attenuation rates of testing line and repairing line are about 1.07 and 1.69 dB·m^-1 respectively, which indicates that the attenuation speed of ground vibration strength caused by the repairing line is faster than the testing line. Compared with the non-uniform excitation, the uniform excitation has a significant effect on the vibration of below 10 Hz for the upper building floor. The acceleration level of each layer has obvious peak at 2.5 Hz, which related to the inherent frequency of building floor. Under the load excitation of testing line, the main vibration frequency ranges of low and top floors are 40-60 and 20-40 Hz, respectively, and the central frequency of peak is concentrated at 40.0 Hz. Under the load excitation of repairing line, the main vibration frequency range of each floor is 0-40 Hz, and the central frequency of peak is concentrated at 31.5 Hz. Compared with the single source excitation, the double vibration source excitation increases the degree of Z direction vibration level of building floor by 0-3.5 dB, which should be paid full attention in the environmental vibration evaluation for the upper building of metro depot.More>

2019, 19(4): 59-69. doi: 10.19818/j.cnki.1671-1637.2019.04.006

FullText HTML (99) PDF (6993KB) (1303)

Bearing mechanism of composite foundation with rigid-flexible piles in loess area

ZHANG En-xiang, HE La-ping, LONG Zhao, HU Zhi-ping

Abstract: In order to investigate the loading transfer mechanism of composite foundation with rigid-flexible piles in collapsible loess area, a field prototype test was conducted. The stress variation rules of piles and soil among piles with different loads and depths were analyzed. Taking the behaviors of rigid single pile as comparison, the characteristic of pile-soil interaction in composite foundation with rigid-flexible piles was summarized. Combined with existing literatures, the difference of mechanical performance between the composite foundation with rigid-flexible piles in collapsible loess area and soft soil area was analyzed. Analysis result indicates that the flexible pile of composite foundation with rigid-flexible piles in collapsible loess area is mainly used to compact the soil among piles and eliminate the collapsibility of soil, and the collapsibility coefficients are basically less than 0.015 after the treatment in the test site. As a result of the compaction of flexible pile, the bearing capacity of soil among piles can be fully developed and the loading transfer ability of rigid pile can be enhanced. The load-sharing ratio of flexible pile is always larger than that of soil among piles in soft soil area. As a result of the high bearing capacity of loess and small modulus ratio between the flexible pile and soil among piles, the load-sharing ratio of soil among piles in collapsible loess area is stable at about 26%, much higher than 7% of flexible piles. The rigid pile in composite foundation pertains to an end bearing friction pile. As the loading increases, the loading transfer ability of rigid pile improves gradually, and the load sharing ratio increases and finally settles at 67%. The improvement of the loading transfer ability of rigid piles is against the development of the bearing capacity of composite foundation with rigid-flexible piles, of which effects on the valid length of pure friction pile should be taken into account in the designing process, and its effects on the bearing capacity of soil at the bottom of the pile of end bearing friction pile should be also been considered.More>

2019, 19(4): 70-80. doi: 10.19818/j.cnki.1671-1637.2019.04.007

FullText HTML (145) PDF (4316KB) (1319)

Axle box spring load characteristics and fatigue damage of high-speed train

YANG Guang-xue, ZHANG Ya-yu, LI Guang-quan

Abstract: Taking the axle box spring of high-speed train as the research object, the spring load test sensor installed on the power bogie was made through the load calibration method. The load-time history of axle box spring was obtained from line test. Combined with a vehicular gyroscope signal, the spring load characteristics of axle box were analyzed under typical working conditions, such as traction, braking, straight line with high or low speed, in and out of ramp, and curve line. According to the change characteristics of axle box spring load, the test load was decomposed into the trend load and dynamic load. On the basis of statistics, the load spectrum of axle box spring under certain mileage was given, the corresponding relationship between load amplitude and its frequency was determined. According to the damage consistency criterion, the damage proportion of load spectrum at all levels and the variation rule of fatigue damage of axle box spring with the increase of train speed were analyzed. Analysis result shows that the axle box spring load is linear with strain, and its transfer coefficient is 94.5 kN^-1. Compared with the non-power side axle box spring, the load amplitude of power side axle box spring is greatly influenced by the torque load of motor. During train start-up phase, the motor output torque reaches the maximum value. The loads on the power and non-power side axle box springs are-7.42 and 1.26 kN. When the train speed increases from 240 to 350 km·h^-1, the axle box spring trend load varies from-0.6 to-2.0 kN, and the maximum dynamic load increases from 1.53 to 1.86 kN, which increases by 22%. The fatigue damage proportions of power side axle box spring are 0.79 and 0.75 when the train is in low and high speeds. The increase of train speed will slightly reduce the fatigue damage proportion caused by the high amplitude load of axle box spring, which is consistent with the distribution characteristics of non-power side fatigue damage proportion. The fatigue damages of both power and non-power side axle box springs tend to decrease first and then increase finally as the train speed increases. The fatigue damage reaches the minimum value when the train speed is around 300 km·h^-1, and the fatigue damages of power and non-power side axle box springs are 0.110 and 0.004 respectively.More>

2019, 19(4): 81-93. doi: 10.19818/j.cnki.1671-1637.2019.04.008

FullText HTML (116) PDF (1618KB) (1271)

Influence of welding residual stress on fatigue strength for EMU aluminum alloy carbody

LU Yao-hui, ZHANG De-wen, ZHAO Zhi-tang, LIU Jun-jie, LU Chuan

Abstract: The residual stresses of butt joint for EMU aluminum alloy carbody were calculated by using the thermal-elastic plastic method and inherent strain method, and the values were compared to verify the rationality of calculating the residual stress by inherent strain method. The shell finite element model of carbody was established, and with reference to the criteria, Railway Applications—Structural Requirements of Railway Vehicle Bodies (EN 12663), the fatigue load conditions of carbody service state were determined. The fatigue strengthes of carbody with and without residual stress were calculated by applying the inertial release method. According to the principle of maximum principal stress, the multi-axial stress of carbody was transformed into the uniaxial stress, and the average stresses and stress amplitudes of concerned points of the welds and the base metal were obtained. Combining with the performance parameters of aluminum alloy carbody material, the Goodman fatigue curve was drawn. The reliability safety coefficients of each concerned point were calculated and the influence of residual stress on the carbody fatigue strength was analyzed. Analysis result shows that the welding residual stress has little effect on the concerned points of base metal, and its reliability safety coefficient decreases by less than 5%. The average stress increasing amount of the concerned point of weld can be up to 25 MPa, and its reliability safety coefficient decreases by more than 50%, up to 54%, which makes the carbody prone to fatigue failure. The residual stress has a significant change in the direction of the maximum principal stress of the concerned point of weld.More>

2019, 19(4): 94-103. doi: 10.19818/j.cnki.1671-1637.2019.04.009

FullText HTML (212) PDF (2791KB) (1179)

High and low temperature characteristics of rubber component dynamic parameters of a bogie

TAN Fu-xing, SHI Huai-long, WANG Wei, LIU Shi-hui, LIU Hong-tao

Abstract: For the rubber components in bogie suspension of a high-speed train, in order to master the frequency-dependent, amplitude-dependent and temperature-dependent characteristics of nonlinear stiffness and damping coefficient, the high and low temperature characteristic tests of dynamic parameters were conducted. The test method for dynamic characteristic of rubber components parameters was introduced. Both the static and dynamic tests at axial and radial directions were performed for the layer rubber spring mounted on axle box and the rubber joint mounted on the swing-arm. The stiffness and damping coefficient were calculated through the load-deflection hysteresis curve. Test result shows that under the normal ambient temperature of 23 ℃, the stiffness and damping coefficient of rubber components only show frequency-dependent and amplitude-dependent characteristics, but their changes are strongly dependent on the temperature. Compared with the normal ambient temperature of 23 ℃, the stiffness and damping coefficient of the rubber components rise significantly under extremely low temperature environment of-60 ℃. In the case of an excitation displacement of 0.50 mm, the stiffness increases by more than 1 times, and the damping coefficient increases by 4-6 times. The higher the excitation frequency, the greater the increasing rate of stiffness and damping coefficient. In case of high temperature environment of 60 ℃, the stiffness and damping coefficient decrease only 5% and 25%, respectively, with respect to that at 23 ℃. The frequency-dependent and amplitude-dependent of rubber components are nonlinear weakened under high temperature environment. Low temperature causes the change of stiffness and damping coefficient of vehicle suspension system, which in turn affects the vehicle dynamics behaviour. With respect to that at ambient temperature, the running safety index like derailment coefficient increases slightly about 5%, whereas the car body vibration accelerations significantly rise around 17%.More>

2019, 19(4): 104-114. doi: 10.19818/j.cnki.1671-1637.2019.04.010

FullText HTML (89) PDF (5956KB) (1322)

Nonlinear dynamics characteristics of maglev vehicle under track random irregularities

CHEN Chen, XU Jun-qi, RONG Li-jun, PAN Hong-liang, GAO Ding-gang

Abstract: The dynamics characteristics of maglev vehicle caused by the random irregularity were studied based on the flexible track. Based on decomposing the track force into a segmented chain structure, an analysis method for the vertical suspension stability of maglev vehicle was proposed. The vibration natural frequencies and modal matrices of flexible track were defined when different suspension forces acted on their respective suspension points. The discrete form of track segmented chain structure and the motion equation of track structure were established. The random excitation generated by the track irregularity was transformed to the system input excitation by the virtual excitation method, and the vibrations of vehicle and track were controlled by taking the random unevenness irregularity of track as the vibration excitation source. The double-loop PID controller with the voltage feedback was used to numerically simulate the vehicle suspension state under different feedback control parameters, and the influences of feedback control parameters on the maglev system stability under the random track irregularity excitation were analyzed. Research result shows that when the maglev vehicle speed is 50-80 km·h^-1 and the displacement feedback parameter, speed feedback parameter and current feedback parameter are 140 000, 50 and 500, respectively, the vehicle can quickly locate from the initial gap of 16 mm to the equilibrium position (9 mm), and can achieve a stable suspension at 2.2 s. The overshoot and steady-state error of system are-1.50 and 0.13 mm, respectively, and the vibration frequency approaches zero. When the displacement feedback parameter, speed feedback parameter and current feedback parameter are 15 000, 50 and 400, respectively, the maglev vehicle suspension stability becomes worse under the action of track random irregularity. The system tends to be stable gradually around 9 s, but it still floats up and down at the equilibrium position, and the vibration frequency and amplitude of system are 7 Hz and 0.5 mm, respectively. When the maglev vehicle speed exceeds 50-80 km·h^-1, the first set of feedback control parameters are no longer applicable. The maglev system diverges around 1.7 s, and the train is unstable, indicating that under the actions of different vehicle speeds and feedback control parameters, the track random irregularity can significantly affect the suspension stability of maglev vehicle.More>

2019, 19(4): 115-124. doi: 10.19818/j.cnki.1671-1637.2019.04.011

FullText HTML (143) PDF (2496KB) (1267)

Influence of y⁺ value on calculation accuracy of aerodynamic parameters of MIRA model

LIU Hai-tao

Abstract: In order to study the influence of dimensionless parameter y⁺ value on the calculation accuracy of aerodynamic parameters of vehicles, based on the step-back MIRA model, the size of near-wall grids was adjusted under the condition that the number and quality of model grids were similar, and the flow field simulation models with different y⁺ values were constructed. Considering that different turbulence models have different applicable ranges of y⁺ values for the external flow field simulation of vehicles, two common turbulence models of shear stress transmission (SST) κ-ω and large eddy simulation (LES) were selected to simulate the steady and unsteady external flow field of the step-back MIRA model. The simulation results of aerodynamic parameter were compared with the experimental results, and the appropriate ranges of y⁺ value were obtained. Combining the velocity nephogram and the carbody surface stress curve from the flow field simulation results, the influence of grid thickness of the first layer of boundary layer on the simulation accuracy was analyzed. The external flow field simulation models for the square-back MIRA model under the two turbulence models were established, the aerodynamic parameters were calculated at different flow velocities, and the ranges of y⁺ value were verified. Analysis result shows that the appropriate average y⁺ value range of SST κ-ω model is 20-50 for the external flow field numerical simulation of vehicles, and the appropriate average y⁺ value range of LES model is 5-10. When the thickness of the first near-wall grid of boundary layer is too large, the numerical simulation can not accurately capture the change of velocity gradient in the boundary layer, which leads to the loss of flow information in the flow field of boundary layer. When the thickness of the first near-wall grid is too small, the boundary layer grid will be seriously distorted. In both cases, the calculation errors of aerodynamic parameter exceed 5%, which will affect the numerical simulation accuracy of external flow field of vehicles. According to the obtained ranges of y⁺ value, the errors of aerodynamic parameter calculated by the square-back MIRA model are less than 5%, which illustrates the correctness of ranges of average y⁺ value for the two turbulence models.More>

2019, 19(4): 125-136. doi: 10.19818/j.cnki.1671-1637.2019.04.012

FullText HTML (187) PDF (4191KB) (1165)

Calculation model of intersection capacity based on traffic flow survival function

HU Yao, WEI Wei, SHANG Ming-ju, LI Li, LI Yang

Abstract: The concept of stochastic traffic capacity of urban road was proposed for the disadvantage that the basic traffic capacity was unable to fully reflect the road traffic conditions. According to the evaluation system, the traffic breakdown and continuous breakdown were defined to quantify the degree of urban road traffic congestion. The existing estimation methods of traffic capacity were studied, and the product-limit and lifetime distribution were used to construct and estimate the traffic flow distribution function. The parameter model of traditional continuous traffic flow was improved by combining the characteristics of traffic flow data of each intersection entrance, and a calculation model of intersection capacity based on traffic flow survival function was proposed. The estimation result of the calculation model was compared with Highway Capacity Manual 2010 model and practical traffic flow of intersection to analyze the computation errors. Analysis result shows that the mean errors of intersection capacity with traffic breakdown and continuous breakdown calculated by the survival function model and HCM2010 model are 0.162 1 and 0.116 4, respectively, and the variances are 0.029 0 and 0.015 2, respectively, both have small error fluctuation. The relative errors between the results of the proposed calculation model and the measured greater traffic flow are 9.720%, 3.822% and 4.936%, 4.779%, respectively. The relative error of the proposed calculation model in a statistic sense is 5.871%, and the estimation effect is robust. There is a product-limit survival function between the traffic breakdown time, probability of acceptable breakdown, traffic flow, speed and traffic capacity. The traffic capacity of the researched intersection is 7 632 pcu·h^-1, so the estimation result of the proposed calculation model is more reliable. Therefore, the proposed calculation model has high practicability, especially in urban road traffic areas with different congestion degrees. By estimating traffic capacity of the acceptable breakdown probability, the optimization objective, scientific decision and acceptable theoretical basis can be provided for urban road traffic organization and management department.More>

2019, 19(4): 137-150. doi: 10.19818/j.cnki.1671-1637.2019.04.013

FullText HTML (95) PDF (2482KB) (1497)

Ordered choice model of travel information demand based on scale choice set

TANG Li, ZHOU Hou-qing, ZHANG Xue-jun

Abstract: The key information needed for urban travelers' daily activity and travel was studied, and a quantitative analysis method focusing on scale data was proposed. Setting traveler's information demand as the study subject, a survey based on Likert 5 scale questionnaire was designed. The rankings of travel individual's demand degrees for various types of travel information were summarized, and the ordered choice models for driving route, destination location and real-time traffic information demand were built and calibrated separately. The ordered choice model was compared with the multinomial Logit model in terms of parameter significance, Akaike information criterion and log likelihood function value so as to verify its effectiveness, and the partial effect analysis on key variables affecting the information demand was carried out. Research result shows that the bus operation change information is mostly required before the travelers' commuting trips, and the real-time traffic information is mostly required during commuting. Non-commuting travelers care destination location information most, and the demand probabilities for destination location information before departure and en-route are 31.08% and 29.25% higher than that before commuting, respectively. In general, workers have higher information demand degree than that of students and freelancers, indicating that people with high value of time are more expected to properly arrange their trips by comprehensive and timely information acquisition. Demand probability of real-time traffic information for female is 10.23% higher than that for male, indicating that females have stronger cognition in avoiding delay risks. The 20 s to 40 s have the strongest travel information needs that gradually decline with age increasing. All age groups show a higher demand for real-time traffic information, indicating that people are more sensitive to information with potential negative impact. Therefore, the scale choice set can be analyzed by using the ordered choice model.More>

2019, 19(4): 151-160. doi: 10.19818/j.cnki.1671-1637.2019.04.014

FullText HTML (98) PDF (963KB) (1730)

Operational benefit evaluation model of flexible parking incentive mechanism based on game theory

JI Yan-jie, GAO Liang-peng, CHEN Dan-dan, TANG Dou-nan

Abstract: The dynamic game process between car travelers and parking managers under the parking mechanism was analyzed. The quantitative relationships between the travelers' bidding behavior, individual travel cost and parking incentive intensity were studied. The operational benefit evaluation model for the flexible parking incentive mechanism was conducted. The sensitivity analysis method was used to explore the effects of various factors on the operational efficiency of flexible parking incentive mechanism, and an empirical study of the University of California, Berkeley was carried out. Analysis result shows that the flexible parking incentive mechanism can not only ensure the sustainability of daily operation of parking lot, but also promote the sharing of parking berths. In terms of economic benefit, the flexible parking incentive mechanism can adjust the number of reclaimed berths through the change of incentive intensity. When the maximum incentive intensity increases from $15 to $30, the number of reclaimed berths reduces from 17 to 8, showing that the mechanism can ensure the economic benefits of parking lot by selectively rejecting the high price berths. As the proportion of people who can accept the minimum incentive increases, the marginal benefit of reclaimed berth will increase. When the proportion increases from 5% to 55%, the marginal benefit increases from $14.4 per berth to $17.3 per berth, showing that with the increase of the number of bidders, the economic benefit of parking lot improves gradually. In terms of social efficiency, the implementation of flexible parking incentive mechanism can help to release the parking berth resources. For example, the daily velocity rate will decrease by about 9%, and the minimum vacancy rate will show an unbalanced increase of 0-8%, showing that the mechanism can effectively promote the travelers to share the parking berths actively.More>

2019, 19(4): 161-170. doi: 10.19818/j.cnki.1671-1637.2019.04.015

FullText HTML (141) PDF (2521KB) (1367)

Hand gesture recognition method in driver's phone-call behavior based on decision fusion of image features

CHENG Wen-dong, MA Yong, WEI Qing-yuan

Abstract: In order to detect drivers' phone-call behavior robustly in natural environment, a hand gesture recognition method was proposed. The Adaboost algorithm was used to detect driver's face region. In YC_gC_r color space, the brightness component and chroma component of facial skin were sampled by sparse grid, respectively, and a Gaussian distribution model of skin color was built. Considering the inhomogeneity of cab illumination, a skin color component drift compensation algorithm was proposed, and an online skin color model was established to adapt the changes of illumination, so that the skin color regions of right and left hands can be accurately segmented. The 2 376 dimensions HOG feature vector of hand skin region was extracted by HOG algorithm, and then PCA method was used to reduce HOG feature vector to 400 dimensions. Meanwhile, the PZMs features of hand skin region were extracted and 8 PZMs feature vectors with the largest weights were screened out by Relief algorithm. A support vector machine classifier decision for phone-call hand gesture was established based on the PCA-HOG and Relief-PZMs features. Experimental result shows that the hand gesture recognition rate based on the PCA-HOG features is 93.1%, and it has good robust to illumination changes but is easily disturbed by hand and head rotation. The hand gesture recognition rate based on the Relief-PZMs features is 91.9%, and it has good tolerance to head and hand gestures but has poor illumination robustness. The hand gesture recognition rate of the proposed multi-feature-fusion method combined with the PCA-HOG and Relief-PZMs is up to 94.5%, and it has good adaptability to illumination fluctuate, hand and head rotation, and other interference conditions.More>

2019, 19(4): 171-181. doi: 10.19818/j.cnki.1671-1637.2019.04.016

FullText HTML (148) PDF (5495KB) (1388)

Formation control model of airport pavement deicing vehicles

XING Zhi-wei, LI Si, LUO Qian

Abstract: In order to solve the problem that the scheme of the actual deicing operation of the airport pavement cannot be fully adapted to the environment, the operation mode and the airworthiness condition of the aircraft during deicing process were considered, and a two-stage deicing operation model with time constraints was constructed. Based on the operation ability of the airport deicing vehicle, the collaborative operation problem of multiple vehicles in the mechanical deicing operation method was studied, and the formation control model based on the complex Laplacian matrix was designed. In order to reduce communication consumption and ensure communication stability, the Henneberg sequence operation method was used to generate the optimal communication diagram of the airport pavement deicing operation vehicles, and the generated optimal communication diagram satisfied the double root condition required by the formation control model. Analysis result shows that the two-stage deicing operation model can select different heterogeneous vehicles for formation work to achieve the optimal time and effect. The formation of control model based on the combination of the complex Laplacian matrix and the leader method is more stable than the traditional control model. The optimal communication diagram generated by the edge directed ensures the availability of communication between the leader and follower in the formation. Under the first-order kinematics model, the speed convergence can be achieved and the desired formation can be generated within 1 min based on the 5-agent "人" type formation. There is no winding or small angle turning in the motion trajectory, which conforms to the actual operation rules of vehicles and can maintain the desired formation in subsequent operations. Therefore, the formation control model can realize the formation control of large-scale heterogeneous deicing operation vehicles and meet the expected requirements.More>

2019, 19(4): 182-190. doi: 10.19818/j.cnki.1671-1637.2019.04.017

FullText HTML (140) PDF (560KB) (1111)

User Center

Journal Info

Collection of Source JournalsMore>

2019 Vol. 19, No. 4

Search Type

Search Journal

LinksMore>