交通运输工程学报

Cover and Contents of Vol.26, No.7, 2026

Cover Contents

Influence of a centralized air chamber on the evolution of compression waves in high-speed railway tunnels

LIU Feng, MA Jian-bin, ZHANG Yang-tai, WEI Meng-jie, CHEN Da-wei

Abstract: To investigate the regulation mechanism of pressure gradients during the propagation of initial compression waves in high-speed railway tunnels by centralized air chambers, and to analyze the influence laws of the air storage effect of the air chamber and the geometric characteristics of the connection on the propagation characteristics of compression waves, numerical calculation studies on the tunnel-chamber coupled dynamic interaction were carried out based on an inviscid, compressible, and two-dimensional axisymmetric model, and a pressure regulation theory involving the synergistic effect of the air storage effect and damping characteristics was proposed. Based on the flow similarity criterion, scaled model tests of an adjustable high-pressure pulse generator were conducted. By comparing the numerical simulation results with the model test results, it was verified that the adopted calculation method could accurately simulate the dynamic response characteristics of the initial compression wave passing through the air chamber. The influence laws of key dimensional parameters of the air chamber on its air storage capacity and the damping characteristics of the connection were deeply analyzed, and the flow response mechanism driven by the pressure difference between the air chamber and the tunnel was explored. By comparing different air chamber sizes and connection sizes, the relief laws of the air chamber structure on the pressure gradient of the initial wavefront were further studied. The influences of wavefront parameters, such as the length and amplitude of the initial compression wave, on the relief effect of the air chamber were analyzed. Research results indicate that the pressure gradient relief mechanism of the centralized air chamber stems from its ability to absorb air when the initial compression wave passes through, and the connection size is a key factor in regulating the air flow response speed from the tunnel into the air chamber and the damping characteristics. When the connection size is 0.40R, and the air chamber size is R, the air chamber exhibits characteristics of high flow accompanied by moderate reflux, corresponding to an optimal relief rate of about 40%. Changing the amplitude of the initial compression wave does not affect the relief rate of the air chamber, while the length of the compression wave is strongly negatively correlated with the relief rate. When the wavelength is shortened, the relief rate of the air chamber significantly increases, and when the wavelength is constant, the relief rate remains stable.More>

2026, 26(7): 1-14. doi: 10.19818/j.cnki.1671-1637.2026.012

FullText HTML (88) PDF (4379KB) (61)

Diagnosis method for periodic irregularity of track slab on high-speed railway driven by multi-source dynamic inspection data

WEI Zi-long, SUN Xian-fu, YANG Fei, ZHANG Hang

Abstract: To achieve precise identification and scientific evaluation of periodic irregularity of track slabs on high-speed railway, based on the track geometry and vehicle dynamic response data collected from high-speed comprehensive inspection trains, a method for synchronous mileage treatment and periodic feature extraction from multi-source dynamic inspection data was proposed, and the time-domain and frequency-domain characteristics of multi-source dynamic inspection data in the periodic irregularity section of track slab were obtained. A method for identifying periodic irregularity of track slabs was constructed by integrating synchrosqueezed wavelet transform and deformation counting method. A comprehensive evaluation method for the periodic irregularity of track slabs was proposed by combining the vertical acceleration of the frame. Research results indicate that the proposed identification method can effectively capture the periodic irregularity of track slabs and evaluate the severity. As the sliding window length is set to 50 m, the threshold values for periodic irregularity deformation of track slabs are set to 0.8 mm for level Ⅰ and 1.5 mm for level Ⅱ, with the corresponding proportional parameters both set to 3/4; a good balance can be achieved between identification accuracy and missed detection rate, and the identification accuracy can reach 92%. Based on measured data from three high-speed railway lines with significant periodic irregularity of track slabs, it is concluded that there is a strong correlation between the vertical acceleration of the frame and periodic irregularity of track slabs in the 4-7 m wavelength band. The 95% quantile of the vertical acceleration of the frame in the 4-7 m wavelength band is between 3.9 and 8.5 m·s^-2. The vertical acceleration of the frame in the 4-7 m wavelength band can be evaluated for the impact of periodic irregularity of track slabs by exceeding the threshold of 5 m·s^-2 for 5 consecutive waves. This method provides guidance for accurately diagnosing the service status of high-speed railway tracks and scientifically conducting maintenance and repair operations.More>

2026, 26(7): 15-26. doi: 10.19818/j.cnki.1671-1637.2026.101

FullText HTML (103) PDF (6753KB) (65)

Intelligent detection method for metro track defect targets based on improved YOLOv8s algorithm

CUI Guang-yan, LI Yu-jie, WANG Yan-hui, AN Jie, ZHAO Wen-yue, ZHANG Tai

Abstract: For the problems of small-sized targets, sample imbalance, low detection accuracy, and high real-time requirements in track state detection, a detection method for metro track defect targets fusing a small target detection algorithm and the Focal Loss function was proposed. The TrackScanner defect dataset was constructed through data cleaning and image annotation. For the problems of "small target samples" and "unbalanced positive and negative samples" existing in the dataset, targeted improvements were made on the basis of the YOLOv8s algorithm. To address the problem of "small target samples", a lightweight dual attention mechanism module, an involution module, and a small target detection head with a detection scale of 160 × 160 were added. The attention of the algorithm to key details was enhanced through joint improvement measures. To address the problem of "unbalanced positive and negative samples", the Focal Loss function was introduced. This improvement regulated the predicted result value of the loss function through a weight factor and a loss factor. Through model training and ablation experiments, the effects and advantages of the proposed algorithm in the field of track state detection were compared and analyzed. Research results indicate that the mAP₅₀, F₁, and defection efficiency of the improved YOLOv8s algorithm are 71.70%, 73.43%, and 46.84 frame·s^-1, respectively, and the detection accuracies for defect targets such as missing fastener, reversed fastener, shifted fastener, broken fastener, foreign object on the track bed, and rail surface abrasion are 82.6%, 65.8%, 72.3%, 25.3%, 95.6%, and 88.6%, respectively. This research is applicable to the detection work of metro track defect targets, can effectively enhance the work efficiency of track inspection personnel, and is crucial for ensuring the safe operation of track trains.More>

2026, 26(7): 27-38. doi: 10.19818/j.cnki.1671-1637.2026.049

FullText HTML (114) PDF (4356KB) (59)

Calculation model for vertical stress of shallowly buried pipe-roofing tunnels and its engineering application

HUANG Ming, CHEN Jun-ning, LAI Feng-wen, LU De-chun, XIAO Xiong, LI Hong-jiang

Abstract: The traditional horizontal differential element method used to characterize the soil arching effect neglects the contribution of inter-layer shear forces to vertical stress transfer, leading to conservative calculation results. In view of this, the pipe-roofing tunnel was first simplified into a "trapdoor" model. A finite element limit analysis (FELA) method was adopted to systematically investigate the variation of the slip angle of the shallow trapdoor under different depth-to-width ratios and soil internal friction angles. Accordingly, based on the trajectory of the major principal stress, a vertical stress calculation model of an arched differential element was established, and the vertical stress solution for the shallow trapdoor was derived. Furthermore, the influences of soil internal friction angle, dilation angle, and surface surcharge on the normalized vertical stress of the trapdoor were clarified. Finally, the proposed model was applied to a pipe-roofing tunnel project to explore its engineering applicability. Research results indicate that the calculation results of the proposed model show good agreement with existing solutions, FELA solutions, and laboratory model test results; compared with the classical Terzaghi theory and code methods, the proposed model reduces the normalized vertical stress by 30.7% and 45.3%, respectively, and it reduces carbon emissions by 9.1% and 16.6%, respectively.More>

2026, 26(7): 39-52. doi: 10.19818/j.cnki.1671-1637.2026.273

FullText HTML (86) PDF (6269KB) (40)

Mechanical properties and micro-mechanisms of tunnel spoil roadbed improved by calcium carbide slag and phosphogypsum

JIAO Ning, ZHANG Li-shi, DING Jian-wen, XUE Chuan-rong, WANG Shou-jie

Abstract: To solve the problems of land resource waste and potential environmental threats caused by the stockpiling of abandoned tunnel spoils and industrial solid wastes, an attempt was made to synergistically improve abandoned tunnel spoils using industrial solid wastes, namely calcium carbide slag and phosphogypsum, to transform them into high-quality engineering fillers. A series of laboratory tests were conducted to investigate the influence patterns of different solid waste ratios on the physical and mechanical properties of the improved soil, and the optimal ratio was selected. Microscopic tests were performed to analyze the microscopic improvement mechanisms and action mechanisms. Road mechanical property tests were carried out to verify the feasibility of using the improved soil as roadbed fillers. The results indicate that the combined addition of calcium carbide slag and phosphogypsum exhibits a significant synergistic effect, especially in enhancing the mechanical properties of abandoned spoils, with an effect far exceeding that of adding calcium carbide slag alone (taking a total mass fraction of 4% as an example, the average compressive strength of the C3P1 improved soil is 1.35 times that of the soil with calcium carbide slag alone). The optimal ratio of calcium carbide slag to phosphogypsum is 3∶1, under which the improvement effect on abandoned spoils is the most significant, and both the mechanical properties and stability of the improved soil reach the optimal state. Hydration reactions and ion exchange reactions occur among calcium carbide slag, phosphogypsum, and the active components in the soil, generating cementitious products such as calcium silicate hydrate (C-S-H), calcium aluminosilicate hydrate (C-A-S-H), and ettringite (AFt). These products significantly increase the compactness of the soil through filling, bonding, and adsorbing soil particles, thereby enhancing the macroscopic mechanical properties of the improved soil. Attributed to the double-edged nature of the expansion effect of the product ettringite and the combined action of other cementitious products, under the optimal ratio (3∶‍1), the mechanical properties of the improved soil do not show a linear growth trend with the increase of the total additive content. The improved soil with an appropriate proportion of calcium carbide slag and phosphogypsum meets the requirements of highway subgrade design specifications, preliminarily verifying the feasibility of using the improved soil as roadbed fillers.More>

2026, 26(7): 53-68. doi: 10.19818/j.cnki.1671-1637.2026.111

FullText HTML (85) PDF (7968KB) (34)

Dynamic stress propagation characteristics and calculation methods for soft soil foundation reinforced by high-speed hydraulic compaction in Yellow River alluvial plain

LI Chao, ZHANG Yu-xuan, LI Wen-jie, MA Long, LIU Qian, ZHOU Chong, SONG Shu-guang

Abstract: To investigate the dynamic stress propagation characteristics of soft soil foundation reinforced by high-speed hydraulic compaction in the Yellow River alluvial plain, field compaction tests at three energy levels (70, 110, and 150 kJ) were conducted based on actual projects. The propagation laws of dynamic stress generated by high-speed hydraulic compaction in the vertical and radial directions were clarified by monitoring indicators such as excess pore water pressure in the soil and ground vibration velocity during the compaction process. On this basis, a theoretical calculation method for dynamic stress propagation characteristics was established, and the effective reinforcement range and the vibration influence range of high-speed hydraulic compaction were evaluated. The research results indicate that the vertical propagation of dynamic stress generated by high-speed hydraulic compaction follows an exponential decay pattern, which exhibits a distinct secondary decay trend. However, the transmission exponent is smaller than the theoretical value in elastic statics. The radial propagation of dynamic stress conforms to a negative power function decay pattern, and the dynamic stress value is positively correlated with the rammer area. The effective reinforcement depth of high-speed hydraulic compaction is positively correlated with the rammer weight and drop height, and negatively correlated with the rammer area and soil unit weight. The effective reinforcement depths for 70, 110, and 150 kJ energy levels are 5.8, 6.0, and 6.6 m, respectively. The effective reinforcement radius of high-speed hydraulic compaction is related to the rammer diameter, approximately 1.1 times the rammer diameter. The dominant frequency of ground vibrations induced by high-speed hydraulic compaction ranges from 5 to 30 Hz, which belong to medium-to-low-frequency vibrations. The vibration velocity decays according to the law of a negative power function with the distance from the tamping point to the observation point (tamping detection distance). For general industrial and public buildings, the safe construction distances for 70, 110, and 150 kJ energy levels are 7.2, 7.8, and 8.4 m, respectively. The safety construction distances for residential buildings, vibration-sensitive buildings, etc. should be increased successively with reference to this standard. The research findings can provide theoretical and practical references for the design and construction of high-speed hydraulic compaction.More>

2026, 26(7): 69-80. doi: 10.19818/j.cnki.1671-1637.2026.050

FullText HTML (77) PDF (5072KB) (31)

Axial compression performance of CFST pier column under long-term load reinforced by encased UHPC

KONG Wen-yuan, XING Zhi-quan, CHEN Li-bo, ZHENG Li, WU Xiao-lei, CHEN Yu

Abstract: To investigate the influence of long-term loading on the post-reinforcement performance of pier columns, 20 axial compression specimens were designed with varying long-term load ratios and section reinforcement configurations. A prestressed long-term loading device was employed to apply sustained axial compression to concrete-filled steel tube (CFST) specimens, and the evolution of strain and deformation was monitored. The specimens were subsequently reinforced by encasing ultra-high performance concrete (UHPC) to form composite pier columns, which were then tested under ultimate axial compression. The effects of long-term loading on failure modes and mechanical behavior were systematically analyzed. A finite element model incorporating core concrete shrinkage and creep was established, and parametric studies were conducted to evaluate the influence of sectional parameters on reinforcement efficiency. Based on experimental and numerical results, a predictive formula for the ultimate axial bearing capacity of UHPC-reinforced CFST pier columns accounting for long-term loading effects was proposed. The results indicate that under low long-term load ratios, splitting failure occurs at the ends of the reinforced layer, whereas under high ratios, crushing failure predominates at the mid-height section. Long-term loading induces additional load transfer to the steel tube due to concrete shrinkage and creep, leading to local deformation at the column base and increased mid-span deflection, which impairs deformation compatibility and reduces reinforcement effectiveness. The reinforcement ratio of the encased layer and the long-term load ratio significantly influence the ultimate capacity. A moderate long-term load ratio can moderately improve the mechanical response, while the inclusion of reinforcement in the UHPC layer effectively mitigates the adverse effects of long-term loading, limiting the capacity reduction coefficient to within 10%. A reinforcement ratio exceeding 1% is recommended for practical applications. The proposed formula offers a reliable reference for the design of CFST member reinforcements under long-term loading.More>

2026, 26(7): 81-97. doi: 10.19818/j.cnki.1671-1637.2026.011

FullText HTML (78) PDF (3712KB) (34)

Effects of environmental temperature field on bridge weigh-in-motion methods

CHEN Shi-zhi, XU Jia-lin, ZHAO Meng-xiang, CHEN Jia-qi, YANG Gan, WANG Tao, HAN Wan-shui

Abstract: Bridge weigh-in-motion (B-WIM) technology is an important means for vehicle load monitoring, and its accuracy is affected by the environmental temperature field; however, existing studies mostly overlook the dynamic interference of the environmental temperature field to it. Taking a concrete T-beam as the object, outdoor vehicle-induced vibration tests were conducted in summer and winter to simulate the coupling effect of multiple conditions including temperature field, vehicle speed, and vehicle weight, and vehicle-induced strain responses were collected. Based on two classical B-WIM methods, i.e., the Moses algorithm and the strain area method, the influence of the environmental temperature field on the identification accuracy of gross vehicle weight was quantified. The results show that high temperature in summer causes the extreme value of strain response of the T-beam to increase by 11.3% compared with winter; under the winter calibration condition, the maximum identification errors of the Moses algorithm and the strain area method for the gross weight of passing vehicles in the same season are 3.68% and 1.41%, respectively, but when the same calibration parameters are applied to the summer tests, the gross weight errors of the two algorithms are significantly expanded to 14.08% and 13.84%, respectively; the analysis shows that the environmental temperature difference between the calibration moment and the testing moment contributes far more to the error than the vehicle speed and vehicle weight, and it is further found that the B-WIM method calibrated in autumn can reduce the annual average error to 3.8%, which is lower than the annual average errors calibrated in other seasons. It can be seen that the environmental temperature during calibration is the key control factor affecting the year-round accuracy of B-WIM. It is recommended that in practical applications, a time period with a stable temperature field and close to the annual average state should be selected as the optimal calibration time according to the climatic characteristics of the bridge location, so as to effectively reduce the temperature-induced systematic error and improve the long-term reliability of bridge weigh-in-motion.More>

2026, 26(7): 98-110. doi: 10.19818/j.cnki.1671-1637.2026.100

FullText HTML (58) PDF (8619KB) (7)

Motion characteristics and trajectories of rockfalls under successive rockfall-vegetation collisions on slopes

DOU Hong-qiang, XIA Zhi-hao, GUO Shu-cheng, WANG Hao, GUO Chao-xu, JIAN Wen-bin

Abstract: To investigate the interception effect of vegetation on rockfalls, reveal interaction patterns between rockfalls and vegetation, and explore rockfall motion characteristics and trajectory distribution under vegetation effects, a laboratory model test of continuous rockfall-vegetation collisions was designed and conducted based on similarity principles of rockfall motion. Based on mathematical and statistical methods, the distribution of rockfall trajectories and the changes in velocity and angle after collisions between rockfalls and vegetation were statistically evaluated and analyzed. Factors with significant effects on rockfall motion characteristics were quantitatively analyzed. The results show that, under non-vegetated conditions, rockfall trajectories are symmetrically distributed along the slope centerline. Under vegetated conditions, rockfall trajectories are characterized by an approximately normal distribution. After collisions with vegetation, rockfall velocity and energy are attenuated to varying degrees. The mean value of the velocity restitution coefficient is 0.497. The original motion direction is basically maintained by about 74.44% of rockfalls. Obvious rebound is observed in about 25.56% of rockfalls. The dispersion degree of rockfall landing points under non-vegetated conditions is significantly lower than that under vegetated conditions. Rockfall velocity at the slope-exit reference line is negatively correlated with the number of vegetation rows and the slope-surface friction coefficient. It is positively correlated with the slope gradient. Under the test conditions, when the number of vegetation rows is greater than 14, rockfall motion can be effectively blocked by vegetation. When the slope-surface friction coefficient is 0.087, the critical slope gradient for effective rockfall blocking is 5°. When the slope gradient is 20°, the critical slope-surface friction coefficient for effective rockfall blocking is 0.35 - 0.40. Rockfall motion characteristics and trajectories under continuous collisions with vegetation can be effectively reflected by the laboratory model test designed for continuous rockfall-vegetation collisions.More>

2026, 26(7): 111-121. doi: 10.19818/j.cnki.1671-1637.2026.015

FullText HTML (43) PDF (2479KB) (6)

Impact of forward and reverse water capacity difference on open lock operation conditions of Chaohu ship lock

WU Jun, CHEN Yun-ji, CHI Ying-hua, ZHOU Shi-liang, SUN Teng-fei

Abstract: To investigate the influence mechanism of the difference in upstream and downstream water capacity of a ship lock on open lock operation conditions, a two-dimensional hydrodynamic numerical model integrating the ship lock and the upstream and downstream river reaches was developed, with Chaohu ship lock in Anhui Province taken as a case study. The model was used to examine the hydraulic characteristics and conditions of open lock operation of the ship lock, alongside prototype observation experiments. Open lock simulation experiments were conducted on the ship lock at a forward and reverse water level difference of 5–30 cm. The experimental results indicate that because of the forward and reverse water capacity difference, the change in water level difference after stabilization of forward and reverse open lock operations at Chaohu ship lock ranges from 2.7 to 7.1 cm and from 2.8 to 16.6 cm, respectively when the initial water level difference is between 5 and 30 cm. During forward open lock operation, the longitudinal flow velocity ranges from 0.71 to 2.15 m·s^-1, and the transverse flow velocity ranges from 0.11 to 0.48 m·s^-1. During reverse open lock operation, the longitudinal flow velocity ranges from 0.68 to 1.63 m·s^-1, and the transverse flow velocity ranges from 0.09 to 0.30 m·s^-1. For both forward and reverse operations, the flow velocities tend to increase with the increasing water level difference. Nevertheless, the flow velocities during forward open lock operation are significantly higher than those during reverse open lock operation, with the longitudinal and transverse flow velocities during forward operation being 1.04–1.32 times and 1.22–1.60 times those during reverse operation, respectively. Based on a combination of prototype observation, numerical model experiments, and ship lock passage tests, with the maximum transverse and longitudinal flow velocities as discriminant indicators for the safe open lock operation of the ship lock, it is found that the maximum permissible water level difference for forward open lock operation at Chaohu ship lock is 11 cm, while that for reverse open lock operation is 18 cm. This difference in permissible water level differences for open lock operation demonstrates that the forward and reverse water capacity difference has a significant impact on open lock operation conditions.More>

2026, 26(7): 122-133. doi: 10.19818/j.cnki.1671-1637.2026.054

FullText HTML (59) PDF (1927KB) (3)

Influence of wheel flat of EMUs on vehicle-track dynamic characteristics at sections in permafrost region

WU Dan, WANYAN Bo, YAO Yun, LONG Ye, LI Xian-feng

Abstract: To study the wheel-rail vibration response characteristics of EMUs under the combined action of sections in a permafrost region and wheel flat, a rigid-flexible coupling vehicle-track dynamic model considering flexible wheelsets and rails was established based on multi-body dynamics theory and finite element method. With the wheel flat and uneven frost heave of sections in the permafrost region as the wheel-rail excitation, the influence law of the joint action of the two on the dynamic performance of the vehicle and the safety threshold of each parameter were studied. The research results show that both wheel flat and frost heave deformation increase the wheel-rail vertical force to varying degrees. When the vehicle running speed is 300 km·h^-1 and the wheel flat length is greater than 30 mm, a brief separation phenomenon occurs between the wheel and rail. Through a comparative analysis of 9 kinds of composite irregularity conditions formed by the combination of wheel flat length (10, 20, and 30 mm) and uneven frost heave deformation amplitude (10, 20, and 30 mm), it is found that the longitudinal creep force is mainly influenced by the wheel flat length, while the lateral creep force is influenced by both the wheel flat length and the frost heave amplitude. In terms of vibration response, the vertical acceleration of the frame is influenced by both the irregularity combination and the vehicle running speed. The vertical acceleration transmitted to the vehicle body is attenuated by the suspension system, and its response level is mainly positively correlated with the vehicle running speed. In terms of operation safety, the rate of wheel load reduction and wheel-rail vertical force are mainly influenced by the wheel flat length. Combined with the vehicle performance safety index, it is recommended that the wheel flat limit be controlled within 30 mm when passing through sections in the permafrost region. The research results can provide some theoretical reference value for the operation and maintenance of high-speed trains.More>

2026, 26(7): 134-144. doi: 10.19818/j.cnki.1671-1637.2026.064

FullText HTML (56) PDF (3340KB) (4)

Review of occupant injury mechanism and integrated protection in frontal collision under AEB conditions

GUO Jian-bao, CAO Li-bo, KANG Wei, LIANG Zheng, ZHANG Jia-fei

Abstract: To further investigate the research progress and future trends of integrated active-passive occupant protection under AEB intervention, an in-depth analysis of the injury characteristics and underlying mechanisms experienced by occupants during AEB events was provided. Based on human biomechanical tests, the applicability and limitations of crash test dummies and human body models in integrated active-passive safety scenarios were evaluated. The protective effects of AEB braking parameters and active pretensioning seatbelts on mitigating occupant out-of-position conditions were also examined. The results show that AEB intervention alters the initial conditions of traditional crashes and triggers instinctive muscular tension responses, thus changing occupants' biomechanical characteristics. The risk of head and chest injuries increases due to forward inertial motion. Although AEB has a relatively small effect on lower-limb displacement, occupants' active muscular responses prior to impact may lead to lower-limb injury patterns that differ from those observed in conventional crashes, and the corresponding injury mechanisms require further study. Significant discrepancies exist between the pre-crash kinematic responses of dummies and those of real humans. Limited research has been conducted on improving dummies for out-of-position conditions. The development of crash test dummies capable of delivering realistic mechanical responses in both low-speed pre-crash and high-speed crash phases will be a key topic for future integrated active-passive safety development. Human body models exhibit good biofidelity during the AEB braking phase and can serve as a complementary tool to crash test dummies for safety condition development. However, evaluation standards for the biofidelity of crash test dummies and human body models in pre-crash conditions remain absent and require further industry standardization. Intelligent restraint systems centered on active pretensioning seatbelts have been proven effective in correcting occupant out-of-position postures and reducing occupant injury. With continuous advancements in intelligent driving technologies, smart cockpit technologies, and AI, integrated active-passive occupant protection will evolve toward diversified protection strategies. By deeply fusing multi-domain sensor data through AI algorithms and combining individualized occupant information, such as weight, height, gender, age, and seating posture obtained via in-cabin cameras, customized occupant protection strategies can be generated, breaking the boundary between active and passive safety and achieving integrated safety, which represents the future development trend.More>

2026, 26(7): 145-164. doi: 10.19818/j.cnki.1671-1637.2026.084

FullText HTML (86) PDF (3026KB) (15)

Analysis of wind and rain load characteristics on high-speed train in heavy rainfall environment based on different rain load calculation models

SHENG Xu-gao, MAO Jun, YU Meng-ge

Abstract: To investigate the rain load characteristics on high-speed trains in a heavy rainfall environment, a numerical calculation model for aerodynamic characteristics of a three-car high-speed train under heavy rainfall was built with the Euler-Lagrange method, and the interaction between raindrops and airflow was established using an interphase coupling approach. Wind tunnel tests were conducted to validate the accuracy of the airflow field model, while raindrop falling tests were performed to verify the accuracy of the rain field model. The Lagrange method was employed to capture the state information of each raindrop impacting the train body within a unit time, including the impact velocity between raindrops and the train as well as the raindrop mass flow rate. The impulse of raindrops was converted into impact loads on the train by utilizing the conservation of momentum, and the impact loads of all raindrops on the train body within a unit time were integrated to establish rain load calculation model A. Based on raindrop impact tests, a normalized time-history curve of raindrop impact force was plotted. Integration was performed for the impact forces of all raindrops striking the train body within a unit time to establish rain load calculation model B. The wind and rain loads on the train were analyzed using the established numerical calculation model for aerodynamic characteristics and rain load calculation models. The results indicate that under the same rainfall intensity, the mass of raindrops falling onto the head train is the largest, followed by that onto the middle train, and the tail train receives the smallest raindrop mass. The aerodynamic drag, longitudinal total load, and the longitudinal rain-wind load ratio of the three cars all increase with the increasing rainfall intensity. When the train speed is 350 km·h^-1 and the rainfall intensity is 500 mm·h^-1, the mass flow rates of raindrops falling onto the head train, middle train, and tail train is 19.43, 9.08, and 5.94 kg·s^-1, respectively. Compared with that in the condition without rain, the aerodynamic drag of the head train, middle train, and tail train increases by 7.39%, 7.69%, and 5.43%, respectively. The longitudinal total loads on the head train, middle train, and tail train are 16 695.07, 7 463.84, and 7 591.44 N, respectively, and their longitudinal rain-wind load ratios are 0.37, 0.38, and 0.22, respectively. This study can provide a reference for the operational safety assessment of high-speed trains under heavy rainfall conditions.More>

2026, 26(7): 165-175. doi: 10.19818/j.cnki.1671-1637.2026.017

FullText HTML (68) PDF (3678KB) (4)

Reinforcement learning anti-rollover control for distributed electric drive multi-axle emergency rescue vehicles

LI Jian, LI Shuai, SU Yan-zhao, CHENG Kun, XU Hai-jun, HUANG Jin, CHEN Jun-jie

Abstract: When emergency rescue vehicles perform urgent dispatch tasks on downhill and sharp-curve road sections, the gravitational component caused by the slope and the lateral inertia induced by sharp turns are likely to trigger vehicle rollover. To address this issue, an anti-rollover control strategy suitable for distributed electric-driven multi-axle emergency rescue vehicles is proposed. The strategy consists of a mode selection layer and an integrated control layer. In the mode selection layer, the lateral load transfer ratio (LTR) was used as an objective evaluation index. The vehicle rollover risk was judged according to the preset threshold, and the system was switched to the anti-rollover control mode. In the integrated control layer, a reinforcement learning controller based on the soft actor-critic (SAC) algorithm was designed to realize the coordinated control of the rear-wheel driving/braking force and the front-wheel fine-tuning angle, thereby improving the vehicle anti-rollover capability. A Python/MATLAB/TruckSim co-simulation was conducted for verification. The results show that the proposed SAC reinforcement learning anti-rollover control strategy reduces the LTR by approximately 9.7% and increases the speed by 3.9% under conditions of high initial speed and large slope; reduces the LTR by about 4.1% and increases the speed by approximately 4.5% under conditions of medium initial speed and medium slope; and reduces the LTR by around 7.8% and increases the speed by about 3.2% under conditions of low initial speed and small slope. A theoretical reference is provided for the learning-based anti-rollover control of distributed electric-driven multi-axle heavy-duty vehicles under extreme operating conditions.More>

2026, 26(7): 176-189. doi: 10.19818/j.cnki.1671-1637.2026.297

FullText HTML (71) PDF (3305KB) (11)

Optimization of testing and evaluation methods and recommendations for decision-making and planning technologies of autonomous driving driven by competition data

TIAN Ye, LI Rui, GUO Hui-jie, ZHOU Xin, FENG Hua-yue, SUN Jian

Abstract: An autonomous driving algorithm competition was carried out, and an evaluation system for autonomous driving algorithms was established. The strengths and weaknesses of different types of algorithms in different scenarios were identified, and an "evaluation-diagnosis-optimization" closed-loop feedback system was formed. The defects of autonomous driving algorithms overly customized for specific scenarios were revealed. The results indicate that the highest score in the first track of the second competition is 72.79; the lowest score is 5.83, and the standard deviation of scores is 18.19. These results indicate that the test questions can distinguish the performance of different algorithms. Based on the deep mining of 582 versions of algorithm submissions, it is identified that Chinese autonomous driving algorithms perform relatively poorly in motor and non-motor mixed traffic scenarios, and the ability to handle the uncertainty of non-motorized behavior still has room for improvement. By comparing different algorithms, it is found that moderate customization for key scenarios is beneficial to improving generalization and stability of algorithms. By comparing pure rule-based architectures and pure learning-based architectures with rule-learning hybrid architectures, it is found that the rule-learning hybrid architecture achieves a better balance between safety and efficiency. Through the test analysis of segmented end-to-end architectures, it is found that such architectures need to incorporate physical constraints and safety models to break through the bottleneck of safety problems. The constructed "evaluation-diagnosis-optimization" closed-loop system provides three core implications for the research and development of decision-making and planning technologies of autonomous driving in China: establishing a benchmark scenario library to solve the "algorithm mismatch" problem, developing a data-rule hybrid architecture to cope with the safety/efficiency game, and formulating dynamic optimization standards for scenario classification to guide the transformation of technological paradigm.More>

2026, 26(7): 190-203. doi: 10.19818/j.cnki.1671-1637.2026.022

FullText HTML (85) PDF (3974KB) (17)

Trajectory planning and hierarchical robust control for autonomous driving on unstructured roads

CHEN Zheng, LI Chun-yu, CHEN Bo-wen, ZHANG Jing, GUO Feng-xiang

Abstract: To address the problems of inaccurate trajectory planning and reduced tracking control stability of autonomous vehicles caused by the typical characteristics of urban unstructured roads, such as severe mixed traffic among traffic participants, high randomness of traffic behaviors, poor functional continuity of road boundaries caused by the above factors, and consequent difficulty in traffic flow data collection, a collaborative optimization method based on a dynamic feasible region and hierarchical robust model predictive control was proposed to improve the safety and stability of autonomous driving. First, for trajectory planning, the road scenario was divided into multiple static feasible regions. A dynamic feasible region was constructed by integrating driving intention analysis, travel trend prediction, and the proportion of space occupied in each static feasible region. The trajectory was parameterized and modeled by piecewise Bézier curves. For tracking control, a hierarchical robust model predictive control (HRMPC) architecture was adopted. The upper layer optimized and generated a nominal system based on the vehicle dynamics model to determine the nominal state. The lower layer obtained sample trajectories by adding finite random factors. A tube constraint based on the 95% confidence interval was then constructed, and the actual system was used to optimize the actual control input. A two-layer collaborative optimization model was thus formed. The effectiveness of the method was demonstrated through proofs of robust stability and asymptotic stability. The results show that, in urban unstructured experimental scenarios, the maximum acceleration is 1.812 m s^-2 and the minimum acceleration is -2.103 m s^-2. The velocity and acceleration curves remain continuous without abrupt changes. The influence of finite random disturbances can be effectively addressed, and safe and smooth trajectories can be planned. Compared with existing robust model predictive control (RMPC) and tube-based robust model predictive control (Tube-RMPC) methods, the proposed method reduces the maximum tracking error by more than 7.4% and the sideslip angle of the center of mass by 26.3%. The results confirm that the proposed dynamic feasible region combined with HRMPC has good stability and can effectively improve the tracking control performance of autonomous vehicles in urban unstructured scenarios.More>

2026, 26(7): 204-219. doi: 10.19818/j.cnki.1671-1637.2026.134

FullText HTML (82) PDF (4738KB) (13)

Multi-parameter detection robot system for highway corrugated beam steel guardrail based on linear structured light scanning

WANG Hui-feng, GUAN Yue-yuan, BAI Jia-di, ZHANG Heng, CAO Jing-ming, WANG Rui, MA Xue-yan, HUANG He, JIAO Yun-mei

Abstract: In order to overcome the limitations of traditional manual detection methods and realize the automatic measurement of structural parameters of highway corrugated steel guardrails, a vehicle-mounted structured light scanning detection system based on an autonomous mobile robot platform was proposed. A direct line structured light scheme was adopted to construct the system detection model. To solve the problem of strong reflection interference easily generated on the guardrail plate surface under complex illumination, an adaptive exposure highlight suppression method was proposed, which effectively suppressed specular reflection noise and eliminated background interference. On this basis, the traditional slope threshold method was improved by combining the line structured light model with the characteristics of light stripe images, which improved the feature recognition accuracy of light stripes. Finally, algorithms suitable for measuring the crossbeam center height, plate shape, deformation, and inclination of corrugated beam guardrails were proposed. Laboratory and field test results show that the system achieves an accuracy of 98.5% in guardrail plate shape recognition; in inclination detection, the relative errors of two-wave and three-wave guardrails are better than 20% and 19%, respectively, with an overall accuracy of more than 80%, which meets the practical engineering requirements; in crossbeam center height measurement, the errors of two-wave and three-wave guardrail plates are 1.439-4.029 mm and 1.137-4.387 mm, respectively, which satisfies the requirements for high-precision detection and automatic inspection of guardrails. The system can realize the efficient, high-precision, and automatic detection of the structural parameters of corrugated beam steel guardrails, which has strong practicability and provides reliable support for the condition detection and maintenance of highway guardrails.More>

2026, 26(7): 220-234. doi: 10.19818/j.cnki.1671-1637.2026.278

FullText HTML (57) PDF (9227KB) (6)

User Center

Journal Info

Index byMore>

Current Issue
2026 Vol. 26, No. 7

Search Type

Search Journal

LinksMore>