Archive

20 September 2024, Volume 43 Issue 9 Previous Issue   
---
Transportation Infrastructure Engineering
Transportation+Big Data & Artificial Intelligence
Transportation Equipment


For Selected:

Download Citations EndNote Ris BibTeX

Toggle Thumbnails

Transportation Infrastructure Engineering

Select

Highway Route Selection Decision Auxiliary Model Based on Improved A* Algorithm CHEN Yuren, LI Dongfeng, YU Bo, GAO Jianqiang 2024, 43(9): 1-8.  DOI: 10.3969/j.issn.1674-0696.2024.09.01 Abstract ( )   PDF (4314KB) ( )   In order to improve the intelligence level of route selection of mountainous highway, the analysis on highway route selection decision auxiliary model was conducted on the basis of the improved A* algorithm. Firstly, the plane constraint indexes and vertical cross-section constraint indexes for highway route selection were proposed. Then, according to the actual situation of mountainous highway route selection, the traditional A* algorithm was improved in terms of valuation function, search range, search direction, data structure, optimization methods, etc., and the highway route selection decision auxiliary model was established. Finally, the validity of the proposed model was confirmed through case study. Research results show that compared with the traditional A* algorithm, the improved A* algorithm reduces search time by about 75%, search iterations by about 48%, and the number of search nodes by about 44%. The total turning angle and length of the route after optimization are reduced by about 83% and 17%, respectively. The highway route selection decision auxiliary model is capable of effectively addressing the route selection problem in complex natural environmental conditions and possesses high practical value. References | Related Articles | Metrics

Select

Finite Element Model Updating of Bridge Based on Response Surface Method and Artificial Rabbits Optimization YANG Hongyin1, 2, JIANG Liangwei1, GU Jianfeng1, CAO Hongyou3, WU Nanhao1 2024, 43(9): 9-17.  DOI: 10.3969/j.issn.1674-0696.2024.09.02 Abstract ( )   PDF (3762KB) ( )   Aiming at the problems of high computational complexity and difficultty in global optimization during the model updating process of bridge structure, a bridge finite element model updating method based on response surface method and artificial rabbits optimization (ARO) algorithm was proposed. The proposed method employed the F-test to identify parameters significantly affecting structural responses and constructed response surface models as surrogate models for the bridge finite element model for iterative calculations. Then, the static-dynamic combined updating method was used to establish an objective function of the calculated values and experimental values of the response surface model based on the characteristic response. The objective function was iteratively calculated by use of the ARO algorithm to obtain accurate finite element model parameters. The proposed method was applied to update the finite element model of the Huanghe Extra-Large Bridge, which was a continuous rigid-frame bridge with variable cross-sections. The results show that the correlation coefficient of the bridge response surface model established on the basis of the significance analysis results is greater than 0.98, with an average absolute error lower than 0.05. After model updating, the maximum relative error of vertical displacement at critical cross-sections of the main beam is reduced from 32.97% to 5.72%, and the average relative error is reduced from 20.67% to 3.61%. The average relative error of the bridge first three-order vertical natural frequencies is only 0.18%. Therefore, it can be seen that the model updated by the proposed method can be used to evaluate the actual stress performance of bridge structures. References | Related Articles | Metrics

Select

Dual Encoder Pavement Crack Detection Network Based on Sobel Operator Bridging LAN Zhangli1, XU Yuantong1, ZHAO Shengwei1, ZHANG Hong2, HUANG Darong1,3 2024, 43(9): 18-24.  DOI: 10.3969/j.issn.1674-0696.2024.09.03 Abstract ( )   PDF (8303KB) ( )   In order to improve the detection accuracy of road pavement cracks, a dual encoder pavement crack detection network based on Sobel operator bridging was proposed to address the issues such as the polymorphism of road cracks and the complexity of noise. The dual encoder consisted of two parts: original image encoding and gradient encoding, in order to solve the problem of single encoder easily losing gradient information. Firstly, the original image encoding results were computed by bridging the Sobel operator in eight directions to generate the encoding information for gradient encoding. Then, the original image encoding results and the gradient encoding results were passed through a multi-scale edge information compensation module to enhance the edge information of the cracks. Finally, dynamic channel graph convolution was introduced to obtain the topological relationships existing between channels to highlight the semantic features of important channels. The research results indicate that the proposed method achieves better results on three benchmark datasets such as DeepCrack, CamCrack789 and CFD. The composite metric ODS is 87.75%, 85.05% and 78.83% for DeepCrack, CamCrack789 and CFD datasets, respectively. References | Related Articles | Metrics

Select

Test and Finite Element Simulation of Bearing Characteristics of Large Diameter Rock-Socketed Cast-in-Place Pile BAI Xiaoyu1, DIAO Haojie1, YIN Jichao1, SANG Songkui1, ZHANG Yamei1, ZHANG Shengkai2 2024, 43(9): 25-33.  DOI: 10.3969/j.issn.1674-0696.2024.09.04 Abstract ( )   PDF (8206KB) ( )   Based on four large-diameter rock-socketed cast-in-place piles in a project in Qingdao, the vertical compressive static load test for single pile and the pile body integrity test were carried out on site, to clarify the bearing characteristics of large-diameter rock-socketed cast-in-place piles under vertical loads and reveal the bearing capacity mechanism of rock socketed cast-in-place piles under different load levels. Combined with the finite element simulation of the bearing capacity of a single pile with different rock-socketed depths, the vertical compressive bearing characteristics and pile-soil displacement distribution characteristics of a single pile at different rock-socketed depths were mainly analyzed. The research results show that under the test condition of the writer, the vertical compressive capacity of single large diameter rock-socketed cast-in-place pile is discrete, and the damage of pile head will greatly affect the effective performance of the bearing capacity of single pile. After loading, the settlement of pile top obtained by the numerical simulation is 31.56mm, which is about 1.1 times of the measured value on site. The simulated results are in good agreement with the measured results. By comparing the simulation results of different rock-socketed ratios, it is found that the larger the rock-socketed depth, the smaller the ratio of pile tip resistance to pile top load, and the change rate of pile tip resistance ratio gradually slows down with the increase of rock-socketed ratio. It is advisable to take 2.5 times the pile diameter for the rock-socketed depth. References | Related Articles | Metrics

Select

Current Situation of Development of Large-Scale Navigation Tunnels and Evolution Trend of Towing Navigation WANG Qi1, WU Jun2, WANG Wei3, HUANG Longjiang1, ZHOU Shiliang2 2024, 43(9): 34-42.  DOI: 10.3969/j.issn.1674-0696.2024.09.05 Abstract ( )   PDF (6973KB) ( )   The navigation tunnel, as a key gateway to open up mountainous waterways and achieve effective connection between navigation buildings, has become the preferred solution for upgrading and expanding the capacity of multiple high-grade waterway hubs such as the Wujiang River. The industry demand and development status of large-scale navigation tunnels were investigated, and the technical bottlenecks faced by self-propelled navigation tunnels, such as high requirements for cross-sectional dimensions, difficulty in construction excavation and uncontrollable ship navigation safety were analyzed. In view of the technical bottlenecks of self-propelled navigation tunnels, the top towing, the shore side towing and other towing navigation tunnel schemes were summarized. A comparative analysis was conducted on the applicable characteristics of various schemes in terms of tunnel section size, safety risks, navigation efficiency, and environmental friendliness. In response to the analysis of the complex fluid-structure coupling between ship and water in a towing navigation tunnel, four key technical problems that urgently needed to be solved were proposed. The four “standardization” requirements brought about by navigation tunnels transforming ship from “self-propelled mode” to “non-self-propelled towing mode” were outlooked. Finally, through the 1:40 physical model test of the towing navigation tunnel, it is concluded that the total resistance of ship towing navigation increases with the increase of Froude number under the conditions of 30 m tunnel width and different water depth, while at the same towing speed, the total resistance of ship towing navigation decreases with the increase of water depth. Some ideas for the expansion of navigation tunnels in the Wujiang waterway are provided. References | Related Articles | Metrics

Transportation+Big Data & Artificial Intelligence

Select

BRP Problem Considering Stowage Plan Based on Improved DQN Algorithm LIANG Chengji, HUA Yue, WANG Yu 2024, 43(9): 43-49.  DOI: 10.3969/j.issn.1674-0696.2024.09.06 Abstract ( )   PDF (1131KB) ( )   In order to meet the requirements of ship stowage plans, reduce the number of crane movements, and enhance terminal operational efficiency, the block relocation problem (BRP) considering stowage plan was studied, which was based on the traditional BRP and also considered the impact of stowage plan on block relocation. To minimize the number of container movements for this problem, a DQN algorithm was designed for solving it. Meanwhile, in order to improve the solving performance of the algorithm, the threshold values based on heuristic algorithms and a new reward function were designed on the basis of original algorithm to improve the algorithm. By comparing with experimental results in other literature, the results show that in terms of computational results, the improved DQN algorithm outperforms the optimal results of current heuristic algorithms in all cases and the larger the scale, the better the results. Moreover, in terms of training time, the improved DQN algorithm significantly outperforms the unimproved one, and the larger the scale, the more significant the time saved. References | Related Articles | Metrics

Select

Reinforcement Learning Algorithm for Airport Flight Delay Recovery DING Jianli, LIU Dekang 2024, 43(9): 50-58.  DOI: 10.3969/j.issn.1674-0696.2024.09.07 Abstract ( )   PDF (2146KB) ( )   Flight delays at airports resulted in aircraft and passengers being stranded at the airport, and improper recovery and scheduling of flight delays can exacerbate the losses caused by delays. Aiming to the issue of minimizing losses in flight delay recovery scheduling, a target function was formulated to calculate the total delay loss, a Markov decision-making process was constructed for flight delay recovery, and an airport flight delay recovery rescheduling model was established. To address computational complexity, a deep learning neural network parameterized policy function was employed to parameterize the strategy of reducing the delay loss objective function value, which was trained by the reward function and advantage function. A reinforcement learning algorithm for airport flight delay recovery was proposed. The research results show that the proposed model can reduce the total loss of flight delays by 7.83% and the duration of passenger delays by 7.23%. The proposed deep reinforcement learning algorithm outperforms other algorithms in both time and performance. References | Related Articles | Metrics

Select

Exploration of Characteristics of Dangerous Goods Road Transport Accidents in China from 2013 to 2022 REN Cuiping, YANG Mengyao, LIU Jiashuo 2024, 43(9): 59-67.  DOI: 10.3969/j.issn.1674-0696.2024.09.08 Abstract ( )   PDF (6500KB) ( )   In order to comprehensively and continuously explore the characteristics of dangerous goods road transport accidents in China, the dangerous goods transport accidents on road from 2013 to 2022 were taken as the research object. By integrating mathematical statistics and event trees, the research was carried out from the dimensions of time and space distribution, morphological characteristics and chain reaction, causes of accidents and types of dangerous goods. The results show that the number of accidents varies significantly among different months, with frequent accidents occurring from June to August. The consequences of the accident are most severe from 2 to 8 oclock. There is a strong positive correlation between the number of regional chemical enterprises and accident rates, as well as between the number of accidents and the number of injuries. There is a medium positive correlation between the number of accidents and the number of deaths. Accidents in developed chemical regions are prominent, but the consequences of accidents caused by individual chemical enterprises are not significant. Single-vehicle and leakage accidents are most common, with leakage being a key node in the chain reaction of accidents. Personnel are the primary factor resulting in accidents of the road transport of dangerous goods. Dangerous goods are not easy to cause accidents by themselves, but the consequences of accidents caused by Class 3 flammable liquids represented by diesel and petrol, and Class 2 gases represented by LPG and natural gas, are more serious. References | Related Articles | Metrics

Select

Variable Speed Limit Control Method Based on Cooperative Vehicle Infrastructure System WU Lan1,2, REN Siqi1, CHEN Qian2, HAN Chenqiu3 2024, 43(9): 68-77.  DOI: 10.3969/j.issn.1674-0696.2024.09.09 Abstract ( )   PDF (4476KB) ( )   To address the issue of low traffic efficiency in the merging area of expressways, real-time information collection and interaction technology in the cooperative vehicle infrastructure environment were used. Based on the Gipps safety distance model, the METANET model was enhanced by incorporating dynamic safety distance into the variable speed limit control. The control method and model for variable speed limit control on the mainline were established and ramp vehicles were guided to merge into the mainline by use of available gaps, which improved the traffic efficiency in the merging area of expressway and the success rate of ramp merging. Finally, VISSIM and MATLAB were used to jointly build a simulation control environment to simulate and evaluate the proposed control strategy under different flow rates. The research results show that under medium and high flow rates, the proposed variable speed limit control strategy increases the speed by 8.75% and 9.28%, reduces the density by 9.45% and 9.77%, reduces the travel time by 9.29% and 8.88% and reduces the delay by 9.21% and 8.84%, respectively. It is demonstrated that the proposed variable speed limit control strategy can enhance the traffic efficiency in the merging area and improve the operational state of traffic flow. References | Related Articles | Metrics

Select

Traffic Flow Prediction Based on Multi-head Attention Spatiotemporal Graph Neural Network XIAO Lin1, 2, CHEN Hongchao1, 2, ZOU Fuming3 2024, 43(9): 78-85.  DOI: 10.3969/j.issn.1674-0696.2024.09.10 Abstract ( )   PDF (2629KB) ( )   Accurate traffic prediction is pivotal for intelligent transportation systems (ITS). However, due to complex time and space dependencies, the existing traffic flow prediction methods have failed to effectively capture the spatiotemporal characteristics of the road network and also ignored the strong dynamic nature of the correlation of road network traffic data in both spatial and temporal dimensions. In order to furtherly improve the prediction accuracy, a spatiotemporal graph neural network model based on multi-head attention was proposed. Firstly, the proposed model constructed an adaptive graph structure learning component that could effectively capture the dynamic spatiotemporal correlation of the graph structure. Secondly, the proposed model respectively designed a temporal multi-head attention module and a spatial multi-head attention module, which were based on the attention mechanism. The designed spatiotemporal multi-head attention module could effectively extract the spatiotemporal features of the road network. Finally, the stacked spatiotemporal convolutional layers were used to predict the future traffic conditions. The results, derived from open-source datasets, demonstrate that the proposed model performs well in spatiotemporal feature extraction and long-term prediction, and achieves more accurate prediction results than baseline methods. References | Related Articles | Metrics

Select

Networked Heterogeneous Traffic Flow Characteristics Based on Improved NaSch Model ZHANG Mengmeng1,2, SONG Jiashu1,2,3, XIE Shukun1,2 2024, 43(9): 86-91.  DOI: 10.3969/j.issn.1674-0696.2024.09.11 Abstract ( )   PDF (7625KB) ( )   To study the evolution law of heterogeneous traffic flow in the intelligent network connection environment, typical scene driving simulation experiments were designed. The driving behavior characteristic indicators of human driven vehicle (HDV) and connected vehicle (CV) were collected, key parameters of heterogeneous traffic flow were analyzed and calibrated. The heterogeneous traffic flow cellular automata model that took into account the differences in driving behavior between HDV and CV was established. Simulation experiment was carried out on the basis of the improved NaSch model, the basic traffic flow diagram in the intelligent connected environment and the characteristics of heterogeneous traffic flow were analyzed. The research results show that: compared to HDV, CV drivers capture of road information and reaction time increases by about 11.4%. Under free flow conditions, the speed of CV increases by 7.4% compared to HDV, and the safe following distance is shortened by 18.2% at the same speed. As the proportion of CV increases from 20% to 80%, the basic traffic flow map shows a significant increase in average speed, and the traffic flow rate increases. The spatiotemporal trajectory map shows a significant improvement in local congestion. References | Related Articles | Metrics

Select

Parking Lot Management Strategy Based on Stackelberg Game LI Jie1, HU Yancheng1, QIN Chenyang2 2024, 43(9): 92-101.  DOI: 10.3969/j.issn.1674-0696.2024.09.12 Abstract ( )   PDF (4167KB) ( )   To address the difficulty of urban parking, a management strategy for urban parking lots based on the Stackelberg game theory was proposed, which involved the joint decision-making of parking lot operators and traffic managers. Firstly, to tackle the intricate interplay among parkers, parking lot operators and traffic managers, the influence among these three participants was analyzed from the perspective of game theory, and the corresponding utility functions for the three participants were established. Secondly, to minimize the conflict of interests between traffic managers and parking lot operators, a game model of these three parties was established based on Stackelberg game theory, and the principles of real-time parking lot access management were determined. Then, based on numerical simulation, the differential evolution algorithm and Kriging agent model were combined to solve the proposed game model, and the corresponding parking reservation management plan and tripartite utility were obtained. Finally, the scheme based on the Stackelberg game was compared with the scheme of maximizing the interests of parking lot operators through numerical simulation. The research results show that in the three scenarios of high, medium and low traffic flow of mainline, the Stackelberg game scheme is the one with the smallest conflict of interests among all parties. References | Related Articles | Metrics

Select

Optimization of Urban Regional Charging Station Layout Considering Service Radius CHEN Weiwei1, ZHANG Ximei1, WEI Yudong2, YANG Wenjing1, JIANG Yuan1 2024, 43(9): 102-108.  DOI: 10.3969/j.issn.1674-0696.2024.09.13 Abstract ( )   PDF (2435KB) ( )   The optimal layout of charging stations plays an important role in the sustainable and high-quality development of the new energy automobile industry. Based on the maximum coverage model, fully considering the problem that the possibility of charging demand coverage was reduced due to the shortening of the service radius of charging stations, the probability of charging station coverage of user charging needs was used as a segmented function of charging service radius. The probability coverage was introduced into the optimal layout of charging stations, and the new random probability coverage model for charging stations was established. Taking Nanan District of Chongqing as a case, by crawling POI data, the genetic algorithm was used to solve the model, and the sensitivity analysis on the comprehensive utility of candidate charging stations was carried out. The research results show that as the service radius reduces, the demand coverage also decreases correspondingly. The charging demand coverage can be effectively improved by building new charging stations. References | Related Articles | Metrics

Transportation Equipment

Select

PMSM Sliding Mode Active Disturbance Rejection Decoupling Control Based on Current Deviation DU Haiming, CHEN Qi, HU Zhihong, FAN Mingjie 2024, 43(9): 109-117.  DOI: 10.3969/j.issn.1674-0696.2024.09.14 Abstract ( )   PDF (5415KB) ( )   A sliding mode active disturbance rejection decoupling control method of permanent magnet synchronous motor (PMSM) based on current deviation was proposed to address the problem of poor dynamic decoupling effect when there was current coupling and inductance parameter mismatch during the operation of PMSM. The proposed method was based on the mathematical model of the motor in the coordinate system of synchronous rotation of d and q axes. The deviation decoupling control was used to compensate the current coupling terms of d and q axes of the motor. In order to accelerate the time for the system to stabilize during startup and reduce output chattering, a new approach law including system state was designed by introducing hyperbolic tangent function through the integral sliding mode control. The difference between the estimated current value and the actual value was used as an expansion observation to comprehensively observe and compensate for disturbances, and the stability of the proposed controller was theoretically analyzed and proved. The simulation and experiment results show that the proposed method improves the rapidity and dynamic control performance of the motor, and thus realizes the approximate complete decoupling of d and q axis currents. References | Related Articles | Metrics

Select

Lithium Battery SOC Estimation Method Based on CSO-AUKF WU Huawei1,2, HONG Qiang1,2, CHEN Yunxing1,2, MA Yubo1,2 2024, 43(9): 118-126.  DOI: 10.3969/j.issn.1674-0696.2024.09.15 Abstract ( )   PDF (2229KB) ( )   Battery state of charge (SOC) estimation is one of the key technologies of battery management system (BMS). A battery SOC estimation method based on the combination of cat swarm optimization (CSO) algorithm and adaptive unscented Kalman filtering (AUKF) algorithm was proposed for lithium-ion batteries. The state equation of lithium battery based on the second-order RC equivalent circuit model was established, and the CSO algorithm was used to improve the identification accuracy of battery. The AUKF algorithm was combined to estimate the SOC. Based on the data of hybrid pulse power characterization test (HPPC) condition and intermittent constant current discharge condition, the effectiveness of the proposed method was verified. The results show that based on CSO-AUKF estimation, the maximum error of SOC is less than 1.64%, and the estimation accuracy and stability are both better than those of the genetic algorithm. References | Related Articles | Metrics

Select

Anti-pitch Interconnection Mode of Hydraulic Suspension for Tracked Vehicles LIU Gang, CHEN Tianyu, ZHANG Baolai, ZHANG Hongyang 2024, 43(9): 127-134.  DOI: 10.3969/j.issn.1674-0696.2024.09.16 Abstract ( )   PDF (5433KB) ( )   Aiming at the problem of poor vibration damping performance of independent suspension tracked vehicles under pitching conditions, a scheme using oil pipes to interconnect the shock absorbers of different axles on the same side of the vehicle body was proposed. Firstly, three interconnection modes under the pitching condition were proposed. Modes 1, 2, and 3 respectively refer to the interconnection of the first and second axis shock absorbers, the interconnection of the first and sixth axis shock absorbers and the interconnection of the second and sixth axis shock absorbers. Secondly, the mechanical-hydraulic-thermal model of 1/2 tracked vehicle shock absorber was established. Then, the E-level pavement model was compiled according to the harmonic superposition method by using MATLAB and RecurDyn. Lastly, the mechanical-hydraulic-thermal joint simulation platform of the hydraulically interconnected suspension model and the whole vehicle dynamics model was established by use of AMESim and RecurDyn. And the anti-pitch performance of the vehicle was compared and analyzed with that of the tracked vehicle equipped with an independent suspension under the E-level road surface. The simulation results show that the hydraulic interconnected suspension tracked vehicles with the installation three modes reduce the root-mean-square value of the body pitch angle by 3.97%, 5.96%, and 3.53%, respectively, and the maximum value of the body pitch angle of mode 2 reduces by 0.0104 rad compared with that of the independent suspension, with a reduction of 14.11%, which has the best anti-pitch performance. References | Related Articles | Metrics