交通运输工程学报

Comparison of numerical interconversion methods for relaxation modulus of asphalt mixture

GU Xing-yu, CUI Bing-yan, XING Shi-qin, HAN Dong-dong

Abstract: In order to obtain the relaxation modulus of asphalt mixture effectively, the numerical interconversion methods from dynamic modulus and creep compliance to relaxation modulus were compared. The conversion principle between linear viscoelastic parameters of asphalt mixture was investigated. The dynamic modulus and creep compliance of the same asphalt mixture were tested, respectively, and the master curves of experimental data were fitted. The function of relaxation modulus was obtained, and the possible reasons for the difference between two numerical interconversion methods were analyzed. The effects of different Maxwell element numbers on the calculation results of relaxation modulus were considered. The relaxation moduli of different asphalt mixtures were compared, and the applicability of two methods to different asphalt mixtures was validated. Research result shows that the fewer the number of Maxwell elements characterizing the asphalt mixtures' relaxation modulus is, the larger the fluctuation of master curves are. When the number of elements is greater than 11, the difference between master curves is lower than 5.26%. It is suggested to select about 11 elements to improve the computational efficiency. The relaxation modulus obtained from the conversion of dynamic modulus and creep compliance accords with the basic relaxation characteristics of materials. The coincidence degree of two master curves of relaxation modulus is high, and the correlation coefficient is greater than 0.99. In terms of different asphalt mixtures, the two methods are also applicable. In the linear viscoelastic range, the main difference of two methods is found in the lower time region that is from 10^-8 s to 10^-4 s. It is suggested to adopt the average value of two methods in practice to avoid the errors caused by the same test. Adding warm mixing agent may reduce the relaxation modulus of asphalt mixture to some extent. Compared to hot mix asphalt, the relaxation moduli of asphalt mixtures with foam agent and Evotherm warm mixing agent reduce by 14.69% and 13.61%, respectively. According to the influence degree on the relaxation modulus, the two kinds of warm mixing agents have the equivalent effect.More>

2019, 19(5): 1-10. doi: 10.19818/j.cnki.1671-1637.2019.05.001

FullText HTML (333) PDF (2770KB) (769)

Shrinkage and mechanical properties of UHPC with coarse aggregate

LI Cong, CHEN Bao-chun, WEI Jian-gang

Abstract: To reduce the shrinkage and cracking risk of ultra-high-performance concrete(UHPC), the autogenerous shrinkage, basic mechanical properties(compressive strength, tensile strength and elastic modulus), aggregate gradation and restrained ring shrinkage tests on five groups UHPCs with different coarse aggregate contents(mass fractions are 0, 12.5%, 22.5%, 32.5%, and 42.5%, respectively) were studied. The influences of coarse aggregate content and aggregate gradation on the autogenerous shrinkage and basic mechanical properties of UHPC were analyzed. The proposed relative difference of shrinkage-cracking stress was used to evaluate the effect of coarse aggregate incorporation on the shrinkage-cracking of UHPC. The cracking performances of UHPCs with and without coarse aggregate under restrained ring were tested and compared. The effectiveness of coarse aggregate incorporation on reducing the shrinkage-cracking of UHPC was verified, and the suggestions for the coarse aggregate content and the maximum particle size limitation in UHPC were given. Research result shows that the autogenous shrinkage of UHPC at early age reduces with the increase of coarse aggregate content, and the maximum decreasing amplitude is about 20%. The effect degrees of coarse aggregate on the elastic modulus, compressive strength and tensile strength of UHPC are depended on its content and gradation. When the coarse aggregate content is 22.5%, the aggeregate gradation curve is almost inside the range of Fuller's and Talbot's curves. It is the most closely packed group among the five material groups. The coarse aggregate of this group has the most remarkable effect on the improvement of elastic modulus and compressive strength of UHPC, and has the minimum effect on the reduction of tensile strength. The maximum relative difference of shrinkage-cracking stress of UHPC with the coarse aggregate content of 22.5% is up to 1.31 MPa. It is the optimal content in the tests and can effectively reduce the risk of shrinkage-cracking. Comparing the UHPC without coarse aggregate, the levels of residual stress and tensile stress of UHPC with the coarse aggregate content of 22.5% under the restrained ring decreases by 15.8% and 14.7%, respectively, indicating that its crack resistance improves. The UHPC with closely packing coarse aggregate is recommended to achieve the material properties as high as possible. The maximum particle size of coarse aggregate can relaxes to 9.5 mm for the UHPC with steel fiber length of 12-20 mm.More>

2019, 19(5): 11-20. doi: 10.19818/j.cnki.1671-1637.2019.05.002

FullText HTML (291) PDF (3983KB) (576)

Longitudinal deformation of expansion joint of suspension bridge under wind and random traffic flow

LI Guang-ling, HAN Wan-shui, CHEN Xiao, XU Xin, LIU Xiu-ping

Abstract: To dynamically simulate and evaluate the longitudinal deformation performance of expansion joint of long-span steel truss suspension bridge under the combining action of wind and random traffic flow at operation stage, an analysis system of wind-random traffic flow-steel truss suspension bridge was established. Based on the existing wind-vehicle-bridge coupling vibration analysis system of single beam, the spring element was introduced to simulate the expansion joint, and the analysis system was improved from the single beam to the grillage method via two aspects of the vehicle-bridge coupling relationship and the fine loading of wind on steel truss girder section. The traffic flow load was simulated and reproduced based on the monitoring data. The dynamic displacement time history response of expansion joint of a typical long-span steel truss suspension bridge under the action of random traffic flow was calculated through the established analysis system. The correlation between the cumulative displacement and traffic flow weight was obtained and verified. Taking the thickness of wear-resisting material of sliding support as the evaluation indicator, the critical value of cumulative displacement of expansion joint was determined, and the normal service life of expansion joint was evaluated. The parameter sensitivity analysis on the longitudinal deformation performance of expansion joint under different wind speeds and random traffic flow was carried out. Analysis result shows that the hourly maximum displacement of expansion joint under the random traffic flow is far less than the designed allowance-880-880 mm. The cumulative displacement of expansion joint is positively correlated with the traffic flow load in corresponding period. Under the combining action of wind and random traffic flow, when the wind speed is less than 15 m·s^-1, the main load factor affecting the longitudinal deformation of expansion joint is random traffic flow load. When the wind speed is greater than 15 m·s^-1, the main load factor is wind load. Both the hourly maximum displacement and hourly cumulative displacement of expansion joint increase with the increase of wind speed. When the wind speed increases to 20 m·s^-1, the longitudinal deformation of expansion joint generated by the wind load is approximately 2 times of that under the traffic flow load. The established wind-random traffic flow-steel truss suspension bridge analysis system can provide a numerical analysis platform for dynamic simulation and performance evaluation on the longitudinal deformation of expansion joint under operation loads.More>

2019, 19(5): 21-32. doi: 10.19818/j.cnki.1671-1637.2019.05.003

FullText HTML (139) PDF (8624KB) (508)

Impact signal dynamic characteristics of energy dissipation shed tunnel

WANG Lin-feng, ZHU Hong-zhou, SONG Nan-nan, ZOU Zheng, YAO Chang-yin

Abstract: Considering the rockfall falling height, mass, shape and cushion thickness, the impact signal dynamic characteristics of energy dissipation shed tunnel were studied by the indoor model test. The spectrums and autocorrelation curves of impact signal were obtained. The time-frequency characteristics of impact signal and the vibration frequency and its change law corresponding to the maximum spectrum were analyzed, and the impact signal of each frequency band was extracted based on the wavelet analysis method. The main energy distribution range of impact signal was obtained. Research result shows that the spectrum magnitude of impact signal at the center of shed tunnel roof increases as the rockfall falling height increases, and this spectrum of impact signal has four peaks with a symmetric distribution. When rockfalls with different shapes impact the shed tunnel, the order of spectrum magnitudes of impact signals from big to small is spherical, cuboid, cube and cylindrical. The thicker the ordinary shed tunnel roof cushion and the smaller the rockfall mass, the smaller the spectrum magnitude of impact signal at the center of shed tunnel roof. When a 5 kg spherical rockfall falls from the height of 0.5 m to impact the shed tunnel without cushion at the top, the maximum spectrum of impact signal of energy dissipation shed tunnel and the peak of autocorrelation curves are 60.98% and 82.57% lower than those of ordinary shed tunnel, respectively. When a 5 kg spherical rockfall falls from the height of 2.0 m to impact the shed tunnel without cushion at the top, the rockfall impact energy of energy dissipation shed tunnel mainly distributes in the frequency range of impact signal from 15.625 to 62.500 Hz, accounting for 63.73% of total energy. The rockfall impact energy of ordinary shed tunnel mainly distributes in the frequency range of impact signal from 0 to 15.625 Hz, accounting for 74.30% of total energy. Thus, the medium-frequency impact should be considered priorly when designing an energy dissipation shed tunnel, and the low-frequency impact should be considered priorly when designing an ordinary shed tunnel.More>

2019, 19(5): 33-41. doi: 10.19818/j.cnki.1671-1637.2019.05.004

FullText HTML (105) PDF (3757KB) (485)

Dewatering model test of advanced deep hole in deep-buried tunnel

ZOU Chong, LEI Sheng-you, ZHANG Wen-xin

Abstract: For the problem of tunnel palm surface collapsing and initial support cracking deformation caused by the sudden water surge in the weak glue-rich water powder fine sandstone, the advanced deep hole dewatering method in deep-buried tunnel was studied. A solid model for simulating the tunnel advanced dewatering was established. The water level surface changes of the model at each moment under 3 kinds of dewatering tubes and 3 kinds of pumping powers were analyzed. The three-axis test was used to analyze the failure state of powder fine sandstone with high water content. Research result shows that the water head in the middle part of dewatering test model at the same elevation measuring point on the tangent section is low, rises gradually on both sides, and is in a parabola form, reflecting the dewatering laws of advanced deep hole. The powder fine sandstone failures plastically under high and low water contents, and the axial strain at the failure is less than 5%. In the dewatering process, when the stratum water content decreases from 20% to 11%, the strength, cohesion, and internal friction angle of powder fine sandstone reach the optimal stable states, realizing the waterless state of excavation surface. The advanced dewatering parameters in the tunnel should be the vacuum dewatering pipe with the diameter of 65 mm and the vacuum pump with the pumping power of 7.5 kW. The dewatering pipe should be arranged at the side walls on both sides of tunnel and at 20 m ahead of tunnel palm surface. The advanced deep hole dewatering supplemented with grouting reinforcement in the deep-buried tunnels with rich water powder fine sandstone can achieve the stability of powder fine sandstone during the excavation. It lays the foundation for the smooth construction of tunnel, and avoids the difficulty of deep well dewatering from the surface in the deep-buried tunnels.More>

2019, 19(5): 42-52. doi: 10.19818/j.cnki.1671-1637.2019.05.005

FullText HTML (144) PDF (11025KB) (521)

Optimization on motion sequence of alignment platform between sensor intelligent chip and fiber array

TANG Hao, ZHANG Zi-lin, ZHOU Bi-feng, TANG Guo-ning

Abstract: Starting with the 720 types of possible motion sequence configurations of spatial motion of motion platform, the sensitivity of geometric error generated by each moving unit during the alignment process between the intelligent chip and fiber array was analyzed. Through distinguishing and classifying the sensitive and insensitive error of each motion unit, the number of motion sequence configurations was reduced to 90. Considering the uniform, decentralized, neat, and comparable characteristics of each motion unit, the orthogonal test design method was used to determine the sensitive and insensitive errors into 3 levels, and determine the 6 motion units into 6 influencing factors. The corresponding orthogonal test table was established, and 5 test paths of motion sequence configurations were obtained. The 5 test paths of motion sequence configurations were simulated through the MATLAB simulation platform, and the optimal motion sequence configuration of motion platform was obtained. The field test was conducted on the multi-degree-of-freedom precision motion platform of packaging system, and the simulation results were verified. Test result indicates that the optimal motion sequence of motion platform for docking the sensor intelligent chip and fiber array in space rectangular coordinates is moving along the horizontal axis first, then rotating around the horizontal axis, and then rotating around the vertical axis, and finally moving along the vertical axis. This method can not only optimize the spatial motion sequence of motion platform aligned by fiber scanning radar sensor smart chip and array optical fiber, but also can predict and plan the registration paths of other multi-degree-of-freedom motion platforms.More>

2019, 19(5): 53-63. doi: 10.19818/j.cnki.1671-1637.2019.05.006

FullText HTML (127) PDF (5098KB) (486)

Influence of slip frequency on running performance of maglev vehicle

ZHANG Min, FAN Yi-li, MA Wei-hua, LUO Shi-hui

Abstract: The longitudinal and vertical components of air gap magnetic field of linear induction motor(LIM) were solved by the two-dimensional electromagnetic field theory, and the analytical expressions of traction force and normal force of LIM were obtained. The analytical calculation method was tested by using the test bench for LIM, and the variations of traction force and normal force with speed under the constant slip frequency range of 6-18 Hz were compared. The dynamics model of a single maglev vehicle with three levitation frames was built. The vibration responses of car body and levitation frame under the impact forces of 1, 3, 5 and 8 kN were simulated and compared. The traction performance of a single middle-low speed maglev vehicle was calculated, the influence of slip frequency on the traction performance of vehicle was analyzed. Considering the influence of normal force on the levitation system and the traction demand of vehicle comprehensively, the variable slip frequency control(VSFC) strategy was proposed. Research result shows that the traction characteristic of LIM generally contains the constant force zone(CFZ) and constant power zone(CPZ). The primary current in the CFZ reaches a maximum of 390 A, and the voltage in the CPZ reaches a maximum of 212 V. The traction force in the CFZ changes little, and decreases rapidly in the CPZ. The smaller the slip frequency is, the greater the starting traction force and normal force of the motor are, and the shorter the CFZ is. When the normal impact force is less than 8 kN, the vehicle stability index grades are all excellent. However, in order to reduce the load of levitation system, the normal force of LIM should be as small as possible. The traction performance of vehicle in the low speed zone under lower slip frequency is better than that under higher slip frequency, but the higher slip frequency is beneficial to improve the traction performance in the full speed range. In the VSFC strategy, the selection of starting slip frequency takes into account the traction performance and levitation ability of vehicle, and the slip frequency gradually increases after the speed reaches the turning point of constant power. Under the VSFC strategy, the traction force is moderate in the CFZ, and is always the maximum value that the motor can exert in the CPZ.More>

2019, 19(5): 64-73. doi: 10.19818/j.cnki.1671-1637.2019.05.007

FullText HTML (144) PDF (4175KB) (462)

Frequency domain calibration and establishment method for load spectrum of bogie frame

ZHANG Zi-fan, LI Qiang, DING Ran, LIAN Qing-lin

Abstract: The quasi-static load-stress transfer relationship in time-domain was analyzed. The parameters of cross-spectral density between loads were taken as the characterization of load coupling effect. Based on the basic theory of multi-axis frequency-domain fatigue, the expression of equivalent stress in frequency-domain was derived. The expressions of discrete load system equivalent to the multi-axis load were obtained. In order to ensure that the damage calculated by load spectrum can cover the measured damage of test line, the 0-order spectral moment of the self-power spectral density of stress signal was taken as the parameter to characterize the damage, and the contribution ratio of load to the damage at the measured point was restrained. According to the principle of damage consistency, load calibration was carried out by using the NSGA-Ⅱ multi-objective optimization algorithm. The line test on a domestic metro bogie frame was carried out, the load and stress data were obtained, and the data processing and analysis were carried out. Analysis result shows that in the load system, the measured variance of the transverse load of frame is the largest, which is 5.08, and the variance of the transverse load of motor is the smallest, which is 0.02. The damage calibration accuracy considering load coupling effect in frequency domain is 1.08×10^-5, while the damage calibration accuracy using the time-domain discrete spectrum is 2.91×10^-3. The calibration accuracy of frequency-domain method is 99.63% higher than that of time-domain method. The comprehensive adjustment multiple of load calibration coefficient considering the coupling effect in frequency domain is 31.81, which is 41.71% lower than that using the discrete spectrum in time domain. The maximum coefficient adjustment multiple of frequency-domain method is 6.99, and the multiple of time-domain method is 15.68, the former is 55.42% lower than the latter. So the calibration method considering the load coupling effect in frequency domain is superior to the calibration method using the discrete spectrum in time domain in terms of error accuracy. The dispersion of coefficient adjustment ratio of frequency-domain method is lower than that of time-domain method. The calibration load is closer to the measured load, and the reliability of calibration results is high. Because the correlation between loads is taken into account in the calibration process, the load system can be applied to multi-axis loading of test bed and independent loading of simulation. The unification of the two loading modes is realized, which provides a new idea for the establishment of frame load spectrum.More>

2019, 19(5): 74-83. doi: 10.19818/j.cnki.1671-1637.2019.05.008

FullText HTML (183) PDF (1181KB) (465)

Evolutionary structure topology optimization method of rail wheel web plate considering UIC strength criterion

ZHENG Xiao-ming, WEN Yong-peng, SHANG Hui-lin, LIU Yue-jie

Abstract: To improve the structural performance of rail wheels, a structural optimization model of rail wheels was established by using the evolutionary structure topology optimization method. The double S-shaped rail wheel was used as the design blueprint, the design field of rail wheel web plate was analyzed, and the evolutionary structure topology optimization method of the rail wheel web plate was put forward under multi-working conditions. The optimization idea using the evolutionary structure topology optimization method to achieve the structural stress homogenization was introduced. According to the standard Overall Wheel Technical Inspection(UIC 510-5: 2003), considering the rail wheels in linear working condition, curved working condition and passing working condition of ballast, respectively, not only was the topology optimization structure obtained under the joint action of 3 typical working conditions, but also six topology structures were obtained under the action of 3 typical working conditions in turn. The stress conditions of wheel web plate before and after optimization were compared, and the web plate stress features of the optimized wheels were verified by using the finite element tool. The correctness and effectiveness of evolutionary structural topology optimization method were proved. Research result shows that the evolutionary structure topology optimization method is suitable for the topology optimization of rail wheels. Under the premise that the wheel weight does not increase, the thickness of wheel web plate increases and is unequal, the stress concentration reduces effectively, and the structural stress reduces. Compared with the original double S-shaped wheels, the structural performances of the optimized six wheel models improve by 16.6%, 20.7%, 22.5%, 21.3%, 20.1%, and 19.5%, respectively. The maximum structural stresses of the optimized wheel web plates of scheme 3 reduce by 4.0%, 14.5%, and 6.7% under 3 working conditions, respectively. The research contributes to the improvement of structural strength of the rail wheels, and has important reference value for the optimization of rail wheel structure under the multi-working coupling condition.More>

2019, 19(5): 84-95. doi: 10.19818/j.cnki.1671-1637.2019.05.009

FullText HTML (190) PDF (3320KB) (480)

Aerodynamic performance of high-speed train under heavy rain condition

YU Meng-ge, LI Tian, ZHANG Qian, LIU Jia-li

Abstract: In order to study the influence of heavy rain on the aerodynamic performance of a high-speed train, the aerodynamics computation model of high-speed train under heavy rain was established based on the Euler-Lagrange method. The air was modelled as the continuous phase, which was described by the Euler method. The raindrop was modelled as the discrete phase, which was described by the Lagrange method. The two-way coupled method was used to simulate the rainfall environment. The calculation of train aerodynamic performance and raindrop simulation were carried out, respectively, and the accuracy of the calculation method was verified by comparing with the experimental data. The flow field structure and aerodynamic performance of a high-speed train under heavy rain conditions were simulated numerically. Calculation result shows that with the increasing of rainfall intensity, under the impact of raindrops, the positive pressure on the front-end area of streamlined head increases, and the negative pressure on the back-end area of streamlined head decreases. As a result, the aerodynamic drag of head car increases. The rainfall intensity has great influence on the aerodynamic drag coefficient of the head car of a train, while has little influence on the aerodynamic lift coefficient. Compared with the aerodynamic drag coefficient under no rain conditions, when the rainfall intensity is 100-500 mm·h^-1, for the train speed of 200 km·h^-1, the aerodynamic drag coefficient increases by 0.004 0-0.020 4, the aerodynamic drag increases by 85-432 N, and the increasing percentage is 2.64%-13.46%. For the train speed of 300 km·h^-1, the aerodynamic drag coefficient increases by 0.002 7-0.013 7, the aerodynamic drag increases by 129-652 N, and the increasing percentage is 1.78%-9.05%. For the train speed of 400 km·h^-1, the aerodynamic drag coefficient increases by 0.002 3-0.009 8, the aerodynamic drag increases by 195-829 N, and the increasing percentage is 1.52%-6.49%. Therefore, the aerodynamic drag coefficient increases with the rainfall intensity at different train speeds, and there is an approximately linear relationship between the coefficient and the rainfall intensity. Under the train speed of 300 km·h^-1 and the raindrop intensity of 100 mm·h^-1, when the raindrop diameter increases from 2 mm to 4 mm, the aerodynamic drag coefficient increases from 0.152 0 to 0.154 9, the aerodynamic drag increases by 138 N, and the increasing percentage is 1.91 %. Therefore the aerodynamic drag coefficient of a high-speed train increases with the increasing of raindrop diameter, and there is an approximately linear relationship between the coefficient and the raindrop diameter.More>

2019, 19(5): 96-105. doi: 10.19818/j.cnki.1671-1637.2019.05.010

FullText HTML (154) PDF (5333KB) (525)

Residual life prediction of aeroengine based on multi-scale permutation entropy and LSTM neural network

CHE Chang-chang, WANG Hua-wei, NI Xiao-mei, FU Qiang

Abstract: Aiming at the change point of aeroengine performance degradation failure and the time series prediction of multi-state parameters, the residual life prediction model based on the multi-scale permutation entropy(MPE) algorithm and long-short term memory(LSTM) neural network was constructed. The change points in time series were analyzed by the MPE algorithm, and the mutation points in the process of performance degradation were solved. The starting point of performance degradation with fault symptoms was obtained. The LSTM neural network model with multi-variables was constructed, and the corresponding residual life was obtained by introducing the multi-state parameter data into the model.The aeroengine multi-state parameters and residual life after the change point were taken as samples and substituted into the LSTM neural network model, the multi-step and multi-variable time series prediction was carried out.The final residual life prediction results were obtained by integrating the state parameter change point analysis method and time series prediction model of aeroengine. Research result shows that the MPE algorithm can monitor the changes of state parameters in time. When abnormal state parameters are found, the value of permutation entropy will jump, which is helpful to discover the fault symptoms in time. The LSTM neural network model selects the information of long time series data through the gated units, and the effective information can be fully reserved for the time series prediction. The multi-variable LSTM neural network can synchronously analyze the multi-state parameters, and directly correspond to the residual life, which improves the efficiency of the model. The combination of MPE algorithm and LSTM neural network model can take the multiple degradation modes of aeroengine into account, and the residual life prediction results of aeroengine are more in line with the actual degradation process. After an example analysis, the root mean square error of the proposed residual life prediction method is 5.3, which is 63%, 72% and 78% lower than that of LSTM neural network, back-propagation neural network and support vector machine, respectively.More>

2019, 19(5): 106-115. doi: 10.19818/j.cnki.1671-1637.2019.05.011

FullText HTML (124) PDF (1095KB) (516)

Emission inventory of ship based on navigation data in Arctic region

MOU Jun-min, ZHANG Xin-sheng, YAO Xin, LI Meng-xia

Abstract: Based on the automatic identification system(AIS), polar ship navigation data were analyzed, and the estimate model of main engine power was established by considering the ice force on the ship. The feasibility and credibility of the main engine power model were verified based on the database of the Lloyd's Register. The dynamic emission model of ship was established by taking account of three navigate states, emission factors and load factors. The navigation data of five ships, including Yongsheng Vessel, ect., which crossing the Arctic region, were selected by the China COSCO Shipping Co., Ltd., and the ship emission estimation model was validated by the fuel consumption method. The emission inventory in the Arctic region was calculated by the emission estimation model, and the temporal and spatial patterns of the emissions were demonstrated on the ArcGIS. Analysis result shows that, among all kinds of ship exhaust emissions in the Arctic region, the CO₂ emission is the largest, about 69.7%, followed by NO_x and SO_x about 13.3% and 12.0%, respectively, and CH₄ is the least, only 0.4%. The emission share ratio of container ships is the largest, reaching 29.3%, and the ratio of icebreakers is the second, reaching 28.8%. Container ships and bulk carriers account for 50.4% of exhaust emissions. The emissions of CH₄, CO₂, CO, HC, NO_x, SO_x, and PM in the Arctic region are 504.85, 82 545.63, 1 645.90, 562.54, 15 711.47, 14 232.54, and 3 263.15 t, respectively, which is generally consistent with the density of vessel traffic. In 2016, the emissions from bulk carriers, container ships, tankers, and fishing boats were the largest in September, and gradually decreased in October and November, which is more related to the icebound condition. Within a day, the emissions of ro-ro ships, fishing boats, and icebreakers have a peak range from 11:00 to 18:00, which may be caused by their work natures.More>

2019, 19(5): 116-124. doi: 10.19818/j.cnki.1671-1637.2019.05.012

FullText HTML (197) PDF (2702KB) (579)

Research progress on car-following models

YANG Long-hai, ZHANG Chun, CHOU Xiao-yun, LI Shuai, WANG Hui

Abstract: The researches on the car-following models in the past 70 years were reviewed. According to the modeling methods, car-following models were divided into two types: theory-driven model and data-driven model, and the hotspots were summarized. The theory-driven car-following model was reviewed from five aspects: human factor, infrastructure, traffic information, heterogeneous traffic flow, and new modeling theory. According to different machine learning algorithms, the data-driven car-following model was also reviewed from five aspects: fuzzy logic, artificial neural network, instance learning, support vector regression, and deep learning. Analysis result shows that the theory-driven car-following model can theoretically deduces the traffic phenomenon. But it is difficult to comprehensively consider the influencing factors, and some human factors are difficult to quantify, and the explanation of driver decision-making process is not accurate enough. The car-following model of heterogeneous traffic flow lacks effective theoretical basis and formal proof under general traffic conditions. The data-driven car-following models summarize the traffic rules by traffic phenomenon. Due to different of data sources, evaluation indicators and methods, the models based on machine learning algorithms cannot be systematically compared. The data-driven models focuse on micro-angles to study driving behavior characteristics, but are not very explanatory for complex traffic phenomena(such as traffic oscillation, hysteresis, etc.). The research of the car-following models should innovate the data collection method, and capture the drivers' psychological tendencies, perceptual characteristics and cognitive abilities, as well as quantify the influence of human factors and make full use of big data. The data-driven car-following models should provide technical support for the development of driverless technology. Before the automatic driving is fully popularized, the characteristics of drivers' car-following behaviors in the mixed scene of manual driving and automatic driving need to be further studied.More>

2019, 19(5): 125-138. doi: 10.19818/j.cnki.1671-1637.2019.05.013

FullText HTML (533) PDF (842KB) (1180)

Coordinated optimization of operation routes and schedules for responsive feeder transit under simultaneous pick-up and delivery mode

WANG Zheng-wu, CHEN Tao, SONG Ming-qun

Abstract: The coordinated optimization problem of operation routes and vehicle schedules for responsive feeder transit under the simultaneous pick-up and delivery mode was studied. The vehicle route representation method based on passengers rather than demand points was devised by considering the personalization of passenger travel time window. The objective function that represents system benefit was constructed by combining the costs of vehicle departure and travel, penalty costs of vehicle early and late arrival, and ticket fares. The vehicle capacity, passenger time window, vehicle running time, vehicle holding quantity and departure time were all taken as constraints, and the integrated optimization model of departure interval, vehicle type and running route was constructed. The double genetic algorithm was designed to solve the integrated optimization model. In this algorithm, the chromosome was coded by multi-chain coding structure, and the chromosome chiasma included two ways of inter-individual and intra-individual. In order to validate the superiority of the simultaneous pick-up and delivery mode, and the effectiveness of the integrated optimization model and the algorithm, several examples were designed to compare the calculation results of the simultaneous pick-up and delivery mode and the separate pick-up and deliver mode. The effects of vehicle speed, single trip running time limit and vehicle composition on the operation efficiency of responsive feeder transit were analyzed. Calculation result shows that under the same passenger demand, compared with the separate pick-up and deliver mode, under the simultaneous pick-up and delivery mode, the departure times reduce by 1, the number of required vehicles reduce by 2, the average seat utilization rate increases by 8.3%, the average vehicle distance required to transport unit passenger reduces by 11.0%, and the operation cost reduces by 15.9%. Therefore, the simultaneous pick-up and delivery mode can effectively improve the operation efficiency. At the same time, under the simultaneous pick-up and delivery mode, when vehicle speed, single trip running time limit, and small vehicle ratio fluctuate by 15.0%, 15.0%, and 12.5% near the reference values, respectively, the maximum change rate of the departure times, the average seat utilization rate, and the objective value reach 20.0%, 15.7%, and 27.1%, respectively, therefore, these parameters have a significant impact on the system operating efficiency.More>

2019, 19(5): 139-149. doi: 10.19818/j.cnki.1671-1637.2019.05.014

FullText HTML (152) PDF (1211KB) (671)

Stochastic dynamic user equilibrium assignment model considering penetration of electric vehicles

XUN Ning, YAO En-jian, YANG Yang, LI Bin-bin, ZHANG Qian

Abstract: To analyze the impact of dynamic charging demand of electric vehicles(EVs) on the service level of public charging facilities(CFs), and provide guidances for the planning and operation of public charging network, a nested Logit model was employed to describe the joint choice behavior of EV travel including charging demand judgment, charging facility and path choices under considering the behavioral differences between EV and gasoline vehicle travelers, congestion state of road section, energy consumption of vehicle, location and service level of charing station. A dynamic traffic flow assignment model considering users' en-route fast charging behavior was developed. The stochastic dynamic user equilibrium condition under the hybrid traffic conditions and an equivalent variational inequality model were proposed, and a dynamic traffic flow iterative algorithm containing the charing queuing simulation on EVs was designed. The effectivenesses of the model and algorithm were verified through a numerical example, and the influences of some key indicators regarding charging demand and supply on the service level of CFs in different promotion phases of EVs were further discussed. Research result shows that affected by the distributions of traffic flow and CFs, the utilization ratios of CFs are unevenly distributed from both space and time perspectives. The average charging waiting time tends to increase with the rise of EV penetration rate(PR). The rise of PR also affects the temporal distribution during the charging peak. The initial state of EV battery charge and queuing length at CF have significant negative effects on users' charging demand judgement. The mismatch between the number of CFs and the demand scale in the road network may lead to a sharp decline in the service level and can easily induce local congestion. For most users, the dwell time at CF is within 15-20 min, and the waiting times of approximate 90% users are less than 9 min. Therefore, the proposed model is consistent with the reality and can fully reflect a series of influences caused by the charging behavior in the hybrid traffic network.More>

2019, 19(5): 150-161. doi: 10.19818/j.cnki.1671-1637.2019.05.015

FullText HTML (260) PDF (1280KB) (671)

Differential pricing decision-making on spot space under dynamic game of air freight transport companies

YU Shu-nan, YANG Zhong-zhen, CHEN Kang, ZHANG Wei, YAO Yuan-yuan

Abstract: In order to study the differential pricing decision-making problem of two air freight transport companies on the same air route, the space pricings of two companies at each sale stage were determined by building a dynamic game pricing model. Dalian Airport and China Southern Airlines on the Dalian-Guangzhou air route were selected to analyze the space pricing and revenue in spot market. Analysis result shows that when the two companies adopt the differential pricing mode, the prices at stages 1-4, 5-6, 7-10, and 11 are 9.7, 12.6, 13.6 and 15.2 yuan·kg^-1, respectively. When the two companies adopt the single pricing mode, the prices at each stage are 12.1 yuan·kg^-1. Therefore, under the dynamic game pricing mode, no matter the differential pricing mode or single pricing mode is adopted, the prices of the two companies at each sale period are completely the same. When the two companies adopt the differential pricing mode and single pricing mode, respectively, the revenues at all stages are 50 928 and 49 519 yuan, respectively, which indicates that in the spot market, adopting the differential pricing mode to sell spaces will make more revenue than adopting the single pricing mode. When the ratio between the levels of the booking demands influenced by own pricing and opponent's pricing is 1.5, 2.0 and 2.5, respectively, the space prices of the two companies in each period decrease gradually, which indicates that the closer the impact of the pricing of the two companies on the booking demands is, the more the room for increasing space prices for the two companies is, and the more the revenues are.More>

2019, 19(5): 162-169. doi: 10.19818/j.cnki.1671-1637.2019.05.016

FullText HTML (176) PDF (805KB) (539)

Information acquisition method of three-dimensional intersection spatial structure based on vehicle GPS trajectory

TANG Lu-liang, YU Zhi-wei, REN Chang, YANG Xue, ZHANG Ya-tao

Abstract: In order to identify different driving rules at the three-dimensional intersections, the features of vehicle trajectory data were analyzed by using random forest feature selection algorithm, and features were clustered according to the importance scores. The clustered results were measured by Davies-Bouldin index to obtain each driving rule cluster under the optimal clustering result, and Delaunay triangle network was constructed based on the cluster range. The skeleton line extraction and common sequence combination method were used to obtain the geometric structure and topological connectivity relationship of three-dimensional intersection. Finally, the spatial structure information of three-dimensional intersection was obtained. Taking the taxi trajectory data of Wuhan in 2016 as data source, the spatial structure information acquisition experiment of three-dimensional intersection in Wuhan was conduct. Analysis result shows that the top four items of vehicle GPS trajectory feature importance scores are the angle of ending point, the angle of starting point, the difference of starting and ending point angles, and the mean angle of middle points. The clustering result using the characteristics combination of terminal angle and starting angle is optimal. The recognition precision rates of the spatial structure information acquisition method in the directions of straight, left and right turning are 85.7%, 85.4%, and 87.5%, respectively, and the comprehensive precision rate is 86.2%. The information recall rates in the directions of straight, left and right turning are 91.5%, 87.2%, and 85.9%, respectively, and the comprehensive recall rate is 88.2%. The higher precision rates and recall rates indicate that the proposed method can accurately identify the spatial structure information and extract the geometric and topological connectivity relationship of driving rules at three-dimensional intersection.More>

2019, 19(5): 170-179. doi: 10.19818/j.cnki.1671-1637.2019.05.017

FullText HTML (97) PDF (9747KB) (704)

Interpolation method of traffic volume missing data based on improved low-rank matrix completion

CHEN Xiao-bo, CHEN Cheng, CHEN Lei, WEI Zhong-jie, CAI Ying-feng, ZHOU Jun-jie

Abstract: An improved low-rank matrix completion method was proposed to study the interpolation problem of road traffic volume missing data. The missing data in the traffic volume data matrix were interpolated in the first round by the low-rank matrix completion based on the nuclear norm. Hierarchical clustering algorithm was applied to classify traffic volume data into different clusters so that the data in the same cluster had strong correlation, while the data in different clusters had weak correlation. Low-rank matrix completion method was applied to each cluster to complete the second round interpolation for missing data. In order to reduce the impact of clustering number, the least square regression ensemble learning approach was proposed to combine the interpolation results under different clustering numbers, so as to obtain the final traffic volume data interpolation results. The interpolation errors of five methods were compared based on the highway traffic volume data in Portland, Oregon, USA, and the influences of different clustering numbers and distance metrics methods were analyzed. Analysis result shows that under the completely random missing pattern, when the missing rate is 10%-60%, the interpolation error reduces by 5.93%-9.11% compared with the traditional low-rank matrix completion model. Under the random and mixed missing patterns, the interpolation errors reduce by 8.32%-9.55% and 8.14%-9.20%, respectively. The integration of multiple interpolations under different clustering numbers can reduce the interpolation error by 2.62%-4.76% compared with the results under single clustering number. Therefore, under three data missing modes, the improved low-rank matrix completion method can reduce the interpolation error of traffic volume data effectively, and improve the effectiveness of traffic volume data after interpolation.More>

2019, 19(5): 180-190. doi: 10.19818/j.cnki.1671-1637.2019.05.018

FullText HTML (233) PDF (2149KB) (539)

User Center

Journal Info

Collection of Source JournalsMore>

2019 Vol. 19, No. 5

Search Type

Search Journal

LinksMore>