交通运输工程学报

Cover and Contents of Vol.24, No.4, 2024

Cover Contents

2024, 24(4): .

FullText HTML (75) PDF (840KB) (16)

Potential assessment of photovoltaic power in expressway area in China

HU Li-qun, HUANG Hong-xin, SHA Ai-min

Abstract: More> In order to promote the application of photovoltaic (PV) in the expressway area, the OpenCV library in the Python programming language was used as a tool, and the route images in the expressway route map were extracted and projected to the light radiation distribution map. The length of the expressway in each radiation area and its proportion were analyzed. The land areas of various types of expressway infrastructures were calculated. The JNMM60 PV module parameters were adopted to calculate the PV power potential in the expressway area. The power demands of expressway operation period and intelligent expressway roadside equipments were compared with the PV power potential, and the PV construction and investment costs were calculated. Analysis results show that the annual average radiation in the expressway area is 1 523.865 kW·h·m^-2, and the annual power generation per square meter of PV land is 63.27 kW·h. The land area of the nationwide expressway can reach about 4.9×10⁵ hm² by the end of 2020 and 6.4×10⁵ hm² by the end of 2025, indicating great PV power potential. Only 75% of the expressway service management area and 10% of the roadside area require the installation of PV equipment to meet all the power demands during the expressway operation period and those of intelligent expressway roadside equipments, and it only takes 4-6 years to recover the construction cost. The large-scale installation of PV equipment in the expressway area still faces difficulties, such as huge initial construction and investment costs, insufficient of PV production capacity and technology, large demand for supporting facilities, unknown interaction between PV infrastructure and road traffic environment, poor space matching between supply and demand, and lack of large-scale PV planning and design management system methods. 14 tabs, 4 figs, 42 refs.

2024, 24(4): 1-13. doi: 10.19818/j.cnki.1671-1637.2024.04.001

FullText HTML (84) PDF (6617KB) (54)

Energy management strategy of integrated photovoltaic-storage-swapping on highways considering influence of photovoltaic uncertainty

WANG Biao, LU Jie, SHA Ai-min, JIANG Wei, LIU Zhuang-zhuang, KE Ji

Abstract: More> In view of the energy management of highways under the influence of uncertain factors of photovoltaic power generation, the issue of swapping electric vehicles in the service area ensuring integrated photovoltaic-storage-swapping was studied under three scenarios, featuring summer sunny days, golden weeks, and winter snowy days in terms of the grid-connected single electricity price and grid-connected time-of-use electricity price from two dimensions: deterministic and uncertain photovoltaic power generation. Taking the maximum photovoltaic self-consistency rate and highest economic benefit as the objective function, constrained by the microgrid power balance and energy consumption characteristics at both supply and demand ends, an optimization model for energy management of integrated photovoltaic-storage-swapping on highways was established by considering the uncertain factors of photovoltaic power generation. To address the shortcomings of traditional genetic algorithms, such as slow convergence rate, poor local search ability, and easy falling into prematurity, an improved multi-objective quantum genetic algorithm based on elite retention strategy and fast non-dominated sorting strategy was proposed. Research results show that the three scenarios featuring summer sunny days, golden weeks, and winter snowy days, can ensure the charging and swapping demands of electric vehicles with consideration for both deterministic and uncertain photovoltaic power generation. Under the constraint of weather conditions in each scenario, the renewable energy utilization rates can reach 10.31%-78.27% with better daily economic benefit. In addition, CO₂ emissions in service areas ensuring integrated photovoltaic-storage-swapping on three scenarios reduce by 62.5%, 41.3%, and 10.3%, respectively. From the perspective of grid-connected electricity consumption mode, the photovoltaic self-consistency rate and carbon emission reduction have no significant difference in terms of single electricity price and time-of-use electricity price. However, the daily economic benefit of time-of-use electricity price increases by 10%, so the time-of-use electricity price scheme has a higher cost performance than the single electricity price scheme. 9 tabs, 11 figs, 31 refs.

2024, 24(4): 14-30. doi: 10.19818/j.cnki.1671-1637.2024.04.002

FullText HTML (56) PDF (11113KB) (40)

Planning method of highway-traffic self-contained microgrid system orientedto multiple energy demand scenarios

SHI Rui-feng, TANG Ke-yi, GAO Yu-qin, JIA Li-min

Abstract: More> To address the demand for green and clean energy consumption in transportation, a method for planning a green, resilient, self-contained, and sustainable highway-traffic self-contained microgrid system was proposed. Based on the endowment of wind and solar renewable resources within transportation infrastructure, initially, the architecture of a highway-traffic self-contained microgrid system was constructed around the renewable energy generation devices, energy storage devices, energy consumption loads, and the microgrid system within the spatial scope of highway transportation infrastructure. The architecture facilitated a synergistic integration of "source-grid-load-storage". Subsequently, based on varying regional characteristics, natural resource endowments, and power grid support conditions in China, the basic models of highway-traffic self-contained microgrid under three typical scenarios of Xizang, Jilin and Zhejiang were proposed. Building upon this, considering the factors such as the economic feasibility of microgrid construction and the renewable energy consumption rate within the microgrid, a system planning model was developed. The goal of the model was to minimize the average annual comprehensive cost of the microgrid, subject to the output constraints of various energy and storage devices and the power balance constraints within the microgrid. Finally, the particle swarm optimization algorithm was employed to optimize the capacity configuration of each energy device within the microgrid. To validate the global search capability and convergence characteristics of the particle swarm algorithm, and to reduce the random factors in the solving process, 50 independent simulation experiments were conducted on the planning schemes for three typical scenarios. Research results indicate that the proposed model and solution method enable a highway-traffic self-contained microgrid system without major grid support to achieve a renewable energy self-contained rate of 99.95% through the coordinated operation of batteries and hydrogen energy storage, while the abandonment rate of wind and solar energy is only 1.34%. It demonstrates that a appropriate storage configuration can effectively mitigate the intermittency and variability of wind and solar outputs, ensure adjustability and reliability of the microgrid's power supply, and provide decision supports for the planning and design of China's highway-traffic self-contained microgrid system. 5 tabs, 6 figs, 31 refs.

2024, 24(4): 31-42. doi: 10.19818/j.cnki.1671-1637.2024.04.003

FullText HTML (47) PDF (8749KB) (40)

Scheme design and configuration optimization of self-consistency energy systems for rail transit

XU Chun-mei, ZHANG Yi-fei, LIU Su-yao, MA Xin-ning, DIAO Li-jun

Abstract: More> To improve the rationality of the construction of self-consistency energy systems for rail transit, the EMU trains on the Beijing-Zhangjiakou Railway were taken as the research objects. According to operating scenarios and line conditions, the topology scheme of a wind-photovoltaic-storage microgrid self-consistency energy system based on the railway power conditioner was determined. Through traction calculation for the longitudinal operation of the EMU trains, the energy flow relationship during train operation was analyzed, and a real-time energy management strategy was designed. Under the premise of reasonable power distribution, with the economical and lightweight wind-photovoltaic-storage self-consistency energy system as optimization objectives, multi-objective optimization technology for the configuration scheme of the self-consistency energy system was studied. Under the conditions of the established line and power constraints, the particle swarm optimization algorithm was used to optimize and calculate control variables of the self-consistency energy system such as the number of photovoltaic cells in series and parallel, the number of storage batteries in series and parallel, and the scale of wind turbines, so as to achieve the optimal configuration scheme of the wind-photovoltaic-storage self-consistency energy system under the given line conditions and objectives. With the actual line conditions of 16 CR400BF trains on the downline of Beijing-Zhangjiakou Railway as an example, the proposed optimized configuration scheme of the wind-photovoltaic-storage self-consistency energy system was verified through MATLAB/Simulink software. Combining the two optimization objectives of the total lifecycle cost of the system (including initial purchasing cost, replacement cost, and electricity purchasing cost) and the total volume of the occupied area, the configurations were optimized by using three different sets of weight coefficients. Analysis results show that with the increase in the weight of the economic objective, the total lifecycle costs of the corresponding optimized configuration schemes reduce by 191.649 million yuan (about 49.1%), 188.258 million yuan (about 48.2%), and 179.911 million yuan (about 46.0%). As the weight of the lightweight objective increases, the total volumes of the optimized self-consistency energy system along the line reduce by 3 377.2 (about 50.4%), 3 393.7 (about 50.6%), and 3 446.9 m³ (about 51.4%). 6 tabs, 15 figs, 31 refs.

2024, 24(4): 43-55. doi: 10.19818/j.cnki.1671-1637.2024.04.004

FullText HTML (48) PDF (5975KB) (36)

Economic characteristics of highway self-consistent energy system planning

HUANG Xian, YE Xiao-rong, JI Wen-tong, FENG Zhang-jie

Abstract: More> To promote the integrated development of transportation and energy, an architecture of highway self-consistent energy system was constructed, with wind-photovoltaic-storage as the power supply side and highway electricity equipment as the demand side. The architecture was equipped with reasonable operation rules. Besides, a planning model for the self-consistent energy system was built with the number of wind-photovoltaic-storage equipment as the planning variable, the equivalent annual cost of system as the optimization objective, the probability of power shortage and the amount of curtailed wind and photovoltaic power as the constraints. Simulations were performed with wind and solar historical data and load demand data in a certain western region as inputs after the operation rules being transformed into specific strategies. Research results indicate that the upper limit increase in microgrid interconnection power can significantly reduce the cost of the self-consistent energy system. When the upper limit of microgrid interconnection power increases from 0 to 1 000 kW, the equivalent annual cost of the system decreases by 8.53%. A new energy accommodation rate increase will lead to rising cost. Too high pursuit of new energy accommodation rates can even result in a sharp cost rise. The strengthening of constraints on curtailed wind and photovoltaic power leads to a 9.05% increase in new energy accommodation rate, while the equivalent annual cost increases by 73.86%. The application of hierarchical load management can significantly reduce the economic cost of the system. When the upper limit of the loss of load probability for primary load remains unchanged, and tertiary loads change respectively from 0 to 0.05 and 0.20, the equivalent annual cost of self-consistent energy system even reduces by 90.97%. Based on these findings, it can be concluded that the application of microgrid interconnection and hierarchical load management, and the appropriate new energy accommodation rate requirements can make the cost of highway self-consistent energy system planning more reasonable. 10 tabs, 11 figs, 46 refs.

2024, 24(4): 56-70. doi: 10.19818/j.cnki.1671-1637.2024.04.005

FullText HTML (60) PDF (10602KB) (32)

Optimization on scheduling decision-making for wind/solar/hydrogen storage highway microgrid based on improved Pareto algorithm

HAO Xue-li, ZHAO Mei-xuan, PEI Li-li, LI Wei, LIU Zhuang-zhuang

Abstract: More> To address the challenges of large loads and frequent sudden load fluctuations during peak hours in highway energy microgrid, an intraday scheduling decision-making optimization model for wind/solar/hydrogen storage highway microgrid was proposed based on the improved Pareto algorithm, to ensure the power peak shaving and valley filling in the microgrid during the intraday operation cycle, help the power system maximize the consumption of wind/solar unit output, and achieve the complementary and optimal utilization of different types of energies. The objective function was constructed by using the lowest intraday operating cost, the lowest carbon emission, and the highest wind/solar consumption rate of the microgrid system as the main criteria. In view of various constraints such as the electric power balance, wind/solar energy output, hydrogen storage energy, and interaction with the external grid, the output strategy was formulated according to the policy of preferential consumption of renewable energy, and the optimal scheduling decision-making results of the microgrid in the region were output. To verify the validity, accuracy, and practicality of the proposed model, the meteorological data and electric load data of Jili Lake section of Xinjiang S21 Highway were analyzed. Research results show that the wind/solar/hydrogen storage highway microgrid system constructed based on the proposed optimization model can effectively improve the consumptions of wind/solar energies. Compared with the traditional Pareto algorithm and multi-objective particle swarm optimization algorithm, the improved Pareto algorithm reduces the total intraday operating cost of the system by 8.5% and 3.7% under the same microgrid structure, increases the renewable energy consumption rate by 3.6% and 10.1%, and lowers the carbon emission by 14.4% and 23.9%, respectively. Thus, the wind/solar/hydrogen storage intraday scheduling decision-making optimization model based on the improved Pareto algorithm can improve the reliability of the wind/solar/hydrogen storage system while ensuring the smooth operation of the highway microgrid system. 5 tabs, 5 figs, 30 refs.

2024, 24(4): 71-82. doi: 10.19818/j.cnki.1671-1637.2024.04.006

FullText HTML (73) PDF (4710KB) (33)

Review on multi-energy integration systems in ports

YUAN Yu-peng, XU Chao-yuan, LI Na, TANG Dao-gui, YUAN Cheng-qing, ZHONG Xiao-hui, YAN Xin-ping

Abstract: More> The current application statuses of wind, solar, hydrogen, and other clean energy in global ports were investigated. A variety of natural resource endowment characteristics were assessed. Based on the types of energy consumption and load characteristics at ports, three microgrid system architectures including DC grid connection, AC grid connection, and AC/DC hybrid grid connection were analyzed, as well as the characteristics of various energy storage methods. In view of the comprehensive architecture of a multi-energy integration system featuring wind, solar and hydrogen storage and the characteristics of its "source-grid-load-storage" network architecture, the key technologies of integration modes, matching methods, energy capture, security guarantees, and operational controls for the multi-energy integration system were summarized. Research results show that the application of clean energy at ports is characterized by single application form and low utilization rate. The energy demand forms of port load are diverse. The application of a multi-energy integration system composed of wind, solar and hydrogen storage units can satisfy the load demand at ports and overcome the shortcomings of single energy source. The mode and architecture of multi-energy integration at ports need to be designed according to the actual situation of the port. Generally, the multi-criteria decision-making approach can be used to determine the optimal energy integration mode, and the multi-objective optimization is adopted to determine the capacity of energy by comprehensively considering security, environment, economy, and other factors. In terms of the selection of specific energy types, wind and solar power generation combined with hydrogen production through the electrolysis of water is partly suitable for practical application scenarios at ports. However, it is necessary to study the adaptation of key equipment and security technologies for hydrogen production, refueling, storage, and supply. Both load and power generation from wind and solar energy are stochastic. Therefore, it is necessary to study new multi-level energy management strategies to optimize the energy dispatching and load balance of the multi-energy integration system and ensure that the system can operate safely, stably, and economically.

2024, 24(4): 83-103. doi: 10.19818/j.cnki.1671-1637.2024.04.007

FullText HTML (83) PDF (14234KB) (28)

Decoupling effect and peak prediction of carbon emission in transportation industry under dual-carbon target

CHEN Tao, LI Xiao-yang, CHEN Bin

Abstract: More> To help the transportation industry achieve the strategic development goals of carbon peaking and carbon neutrality, the change trend and influencing factors of carbon emission in China's transportation industry were analyzed from two perspectives of historical verification and future prediction. The logarithmic mean Divisia index (LMDI) model was used to decompose the influencing factors of CO₂ emission change in China's transportation industry from 2000 to 2020. The decoupling state of carbon emission and economic development in the industry and the driving factors of decoupling were analyzed by combining the Tapio decoupling model. The decomposition results of influencing factors were used as the basis for the selection of the indicators in the scenario analysis method, and the variations of prediction indicators under different scenarios were set. A prediction model of stochastic impacts by regression on population, affluence, and technology (STIRPAT) was constructed by using ridge regression. Analysis results show that the total CO₂ emission exhibits an increasing trend year by year during the study period, with a cumulative increase of 694 million tons from 2000 to 2020. The decrease in transportation intensity is the main inhibiting factor for the increase in carbon emission, with a cumulative effect of -626 million tons. The growth of per capita GDP is the most important factor promoting the increase in carbon emission, and the cumulative effect is 1 294 million tons. The energy consumption is still dominated by fossil fuels, and the energy structure is not significantly optimized. The decoupling index of industrial carbon emission is in a stable decline stage, and the decoupling state improves, mainly manifesting in three states, such as the expansion negative decoupling, growth connection, and weak decoupling. The optimization of energy structure is the most potential factor to help the decoupling. In the future, the carbon emission in China's transportation industry will rapidly grow at first, slow down near the peak, reach a plateau for a short period after the peak, and finally decline. In the baseline, pessimistic, and optimistic scenarios, the peak CO₂ emission in China's transportation industry will occur in 2040, 2045, and 2035, respectively, with peaks of about 1.210 billion, 1.263 billion, and 1.130 billion tons, respectively.

2024, 24(4): 104-116. doi: 10.19818/j.cnki.1671-1637.2024.04.008

FullText HTML (60) PDF (2667KB) (31)

Performance prediction of hydrogen enriched compressed natural gas engine based on IMPSO-BPNN

DUAN Hao, ZHANG Meng, WANG Jin-hua, ZHANG Feng-qi, ZENG Ke

Abstract: More> To further improve the performance of traditional particle swarm optimization back-propagation neural network (PSO-BPNN) model, based on the influence mechanisms of inertia weight and acceleration factor on particle swarm optimization, an improved particle swarm optimization back-propagation neural network (IMPSO-BPNN) method adopting non-linear decreasing inertia weight and non-linear acceleration factor was proposed. The IMPSO-BPNN method was applied to the regression analysis and prediction of performance parameters such as torque, equivalent brake specific fuel consumption, and brake specific NO_x emission of a hydrogen enriched compressed natural gas (HCNG) engine. It was also compared with other neural network methods in terms of prediction accuracy, generalization ability, and convergence speed, including PSO-BPNN, genetic algorithm optimized back-propagation neural network (GA-BPNN), and back-propagation neural network (BPNN) methods. Research results show that the fuel-air ratio and spark advance angle can significantly affect the torque, equivalent brake specific fuel consumption, and brake specific NO_x emissions of the HCNG engine. With torque as the predictive variable, the average absolute percentage error of the optimal IMPSO-BPNN model is 5.85%, 12.62%, and 17.96% smaller than those of PSO-BPNN, GA-BPNN, and BPNN models, respectively, and the correlation coefficient of the optimal IMPSO-BPNN model is 0.999 86, also the highest among these models, which indicates that the prediction performance and generalization ability of the model established by the IMPSO-BPNN method are generally superior to those established by other methods. With brake specific NO_x emission as the predictive variable, the CPU running times reduce by 95% in both the optimal PSO-BPNN model and the optimal IMPSO-BPNN model compared with the optimal GA-BPNN model, which demonstrates the superiority of PSO-BPNN and IMPSO-BPNN methods to the GA-BPNN method in terms of time dimension. Therefore, compared with PSO-BPNN and GA-BPNN methods, the proposed IMPSO-BPNN method has significant advantages in prediction performance and generalization ability, and ensures high computing efficiency.

2024, 24(4): 117-128. doi: 10.19818/j.cnki.1671-1637.2024.04.009

FullText HTML (36) PDF (3733KB) (20)

Review on pavement power generation technologies

ZHOU Yu-ming, DENG Yao, LIU Yu-qin, PENG Zhu-yi, ZHA Xu-dong, LI Ping, WEI Jian-guo, LIU Zhao-hui

Abstract: More> To systematically understand the development of pavement power generation technologies and promote the rapid development of green and smart roads enabling energy saving and emission reduction, the CiteSpace software was used to conduct a quantitative analysis of relevant literature on the pavement power generation technologies from 2012 to 2022. The research progresses, advantages, disadvantages, and applicabilities of three main technologies, namely the photovoltaic power generation, thermoelectric power generation, and piezoelectric power generation, were compared. The fundamental theories of converting solar energy, thermal energy, and mechanical energy into electrical energy were introduced. The pavement design methods of photovoltaic and thermoelectric power generation technologies were summarized. The selection of power generation materials for piezoelectric power generation technology, the design of piezoelectric transducer device, and the structural design of integrated power generation pavement system were discussed. The future research trends of pavement power generation technologies were prospected. Based on the existing research foundations of pavement power generation technologies, some suggestions were put forward for the development of green and smart roads and the demand for the integrated development of transportation and energy from the perspectives of material, structure, construction, operation and maintenance. Research results show that most studies on the photovoltaic power generation focus on the macro-level analysis of the feasibility of solar pavements, providing power supply for transportation infrastructures and alleviate the heat island effect. However, there still is much room for optimization in the research on mechanical properties and power conversion efficiency of photovoltaic pavement. Thermoelectric power generation mainly relies on the temperature difference of pavement structure. It can realize all-weather power generation with stable energy harvesting. However, it currently has a disadvantage of low efficiency. At the same time, it is necessary to focus on the problem of mismatch between mechanical properties of thermoelectric heat conduction devices and asphalt pavement. Piezoelectric power generation has a high energy harvesting density, good sustainability, and promising prospects. However, some key issues have not been well solved, such as the durability of piezoelectric material, compatibility and stiffness matching between piezoelectric transducer elements and pavement, and structural stability and durability of the integrated pavement of piezoelectric power generation system, which still require further research.

2024, 24(4): 129-147. doi: 10.19818/j.cnki.1671-1637.2024.04.010

FullText HTML (30) PDF (16637KB) (27)

Influence of wheel-rail excitation on contact characteristics and stress intensity factors of cracked gears for locomotives

ZHU Hai-yan, TAO Ze-yu, WANG Yu-hao, WANG Meng-wei, ZHANG Wei-hua, XIAO Qian, YI Yong

Abstract: More> The time-varying meshing stiffness of gears with root crack faults was solved by the numerical method, a 6-DOF gear dynamics model was established, and the Newmark method was used to solve the dynamic responses of the gear transmission systems with crack depths of 1, 2, and 3 mm. The time domain signal was analyzed, and the sensitivities of different statistical indexes to different fault degrees were calculated. A multi-body dynamics model of the locomotive was built based on the multi-body dynamics. The contact characteristics of the gear transmission system with tooth root cracks were solved under wheel polygon excitations of 18th, 19th, and 24th orders and rail corrugations of 0.5, 1.0, and 1.5 mm wave depths. The contact states of tooth surfaces under different excitation conditions were simulated. Analysis results indicate that the time-varying meshing stiffness gradually decreases with the increase of crack propagation angle and crack depth. With the deepening of crack depth, the vibration and shock of the transmission system become increasingly severe. The pulse factor is sensitive to the crack fault characteristics and is suitable as an evaluation index of crack fault characteristics. With the increase of wheel polygon order and rail corrugation depth, the maximum contact force of a single tooth surface with cracks is about 3.0 times and 6.4 times that under the no excitation, respectively. When the wheel polygon order is 24, the contact resultant force of the tooth surface and the contact force of the single tooth surface both reach maximum values, which are 4 125 and 1 178 N, respectively. The value of the mode Ⅰ crack stress intensity factor is much larger than that of the mode Ⅱ crack stress intensity factor. Mode Ⅰ crack is dominant in crack propagation, and the stress intensity factor increases with the increase of load and expansion degree, indicating that the existence of wheel-rail excitation can increase the crack propagation rate of cracked teeth, shorten its service life, and affect the safe operation of locomotives.

2024, 24(4): 148-160. doi: 10.19818/j.cnki.1671-1637.2024.04.011

FullText HTML (30) PDF (14353KB) (22)

Metro vehicle lateral vibration characteristics based on vehicle-equipment coupling under service conditions

WEN Yong-peng, WU Jun-han, ZHONG Shuo-qiao, ZONG Zhi-xiang, ZHOU Hui

Abstract: More> To reduce lateral vibrations of vehicle body and improve the comfort of metro vehicles during long-term service, a lateral dynamics model of metro vehicles with underframe equipment was built. The dynamic change rules in the distribution of vehicle speed and the equivalent conicity reflecting the wheel-rail contact state under service conditions were investigated. The lateral vibration characteristics of vehicle-equipment coupling were obtained, and a parameter design method for reducing lateral vibrations in vehicle-equipment coupling was formulated. Analysis results indicate that the service conditions of metro vehicles change dynamically with the variations in vehicle speed and equivalent conicity. The critical speed of the vehicles decreases and the lateral vibrations intensifies with an increase in the equivalent conicity of wheel-rail contact. The characteristic frequency of lateral vibration increases linearly with the increase in vehicle speed and equivalent conicity, and its value is concentrated in the low-frequency range of 1-3 Hz. When vehicle body and equipment undergo lateral vibration coupling, an optimal matching relationship exists between equipment mass and elastic suspension stiffness. Selecting an appropriate suspension stiffness can significantly reduce the lateral vibrations of vehicle body. Furthermore, increasing the width and roll inertia of underframe equipment helps reduce the lateral vibrations of vehicle body. To suppress the lateral vibration of vehicle-equipment, attention should be paid to the selection of lateral suspension stiffness, which is recommended to be 6.0×10⁴ N·m^-1. The lateral vibration frequency of equipment is made close to that of vehicle body through the adoption of a proper elastic suspension connection between underframe equipment and vehicle body. The lateral vibration of vehicle body can be reduced by utilizing the same frequency coupling vibration between vehicle and equipment, thereby improving the quality of vehicle operation.

2024, 24(4): 161-170. doi: 10.19818/j.cnki.1671-1637.2024.04.012

FullText HTML (30) PDF (10037KB) (22)

Multi-objective equilibrium optimization model and improved NSGA-Ⅲ algorithm of railway construction

ZHANG Yan, LIU Ji-zhen, QIN Jia-liang, YANG Lan, ZHANG Hong

Abstract: More> The characteristics, optimization models and optimization algorithms of railway infrastructure construction schemes were analysis, the double code-network diagrams were drawn, the time required for the construction process was taken as independent variable, and a method for calculating the construction cost was proposed under considering the time cost of capital. The system reliability theory was introduced to quantitatively assess the construction quality, the interrelationship between the safety level, time and cost of construction quality was explored, the safety level was calculated, and a multi-objective equilibrium optimization model of quality-safety-duration-cost for railway infrastructure construction was put forward. The NSGA-Ⅲ algorithm was improved by introducing the random integer genetic coding method and penalty function method to solve the Pareto solution set of the model, the solution performance of the improved algorithm was compared with the NSGA-Ⅱ algorithm, and the model was verified by using a railway construction case. Analysis results show that when the population number is 140, the iteration number is 900, and the test number is 40, the average coverage rate per generation of the improved NSGA-Ⅲ algorithm to the NSGA-Ⅱ algorithm is nearly 27 times higher than that of the NSGA-Ⅱ algorithm to the improved NSGA-Ⅲ algorithm, and the mean supervolume per generation of the improved NSGA-Ⅲ algorithm is nearly 54% higher than that of the NSGA-Ⅱ algorithm, therefore, the improved NSGA-Ⅲ algorithm is obviously superior to the traditional NSGA-Ⅱ algorithm. The proposed model and improved NSGA-Ⅲ algorithm are well applied to the multi-objective equilibrium optimization of railway construction management. In the construction case of track engineering, when the population number is 140, the iteration number is 900, and the reference point number in each dimension is 8, 140 Pareto solutions are obtained, and the maximum optimizations of quality level, safety level, duration and cost of the engineering are 0.1121, 0.1073, 36 days and nearly 7.2 million yuan, which can better guide the decision makers to arrange the construction.

2024, 24(4): 171-183. doi: 10.19818/j.cnki.1671-1637.2024.04.013

FullText HTML (43) PDF (2673KB) (28)

Importance evaluation of edges in transportation network under dynamic and randomly disruptive event

DU Yong-jun, WANG Ning, ZHANG Pan, CAI Zhi-qiang, QIAO Xiong

Abstract: More> A dynamic Bayesian importance measure method was proposed for edge importance evaluation of transpartation network. The random process theory was applied to characterize the generation process of external disruptive events, and a transportation network reliability model was established. Probabilistic techniques were utilized to obtain a formula of dynamic Bayesian importance measure of each network edge, and a maximum value for the importance measure and corresponding maximum edges were determined. Based on the formula, an numerical algorithm was developed to evaluate the Bayesian importance measure of each edge at different times. An actual case of a transportation network was introduced, and the dynamic random disruptive shock process incurred by the edges was a saturated non-time homogeneous Poisson counting process with given scale parameters and shape parameters. The calculation method of dynamic Bayesian importance measure was demonstrated, and the sensitivity analysis of the scale parameter and shape parameter of the importance ranking of the connected edges was made. Research results show that, regardless of the changes in random disruptive events from the external environment, the one-edge cut in the network remains the most important edge, which verifies the correctness of the theoretical analysis. Considering external random disruptive events and the network structure, the Bayesian importance measure can timely and accurately identify the importance of all edges, which fills the gap left by the traditional static measure for edge importance that only considers the position of an edge. As the values of the scale parameters and shape parameters become larger, the importance ranking of the two edges changes faster.

2024, 24(4): 184-194. doi: 10.19818/j.cnki.1671-1637.2024.04.014

FullText HTML (42) PDF (2878KB) (29)

Typical feature recognition of dynamic anti-migration for wireless charging vehicles in road traffic systems

ZHOU Xi-wei, SHI Wen-shuai, WANG Hui-feng, DAI Liang, WU Qi-sheng, BAI Ye-hong

Abstract: More> In view of the dynamic feature recognition and classification problem of wireless charging vehicles, a composite induction device based on vehicle-road cooperation with a hybrid electromagnetic induction unit and a geomagnetic field induction unit was designed. In the hybrid electromagnetic induction unit, to realize the induction recognition under dynamic conditions, a detection method with the effective value of resonant current as the feature of wireless charging vehicles was put forward. Two forms of electric field coupling and magnetic field coupling were introduced into the circuit topology, alongside a high-order bilateral capacitor-inductor compensation structure. Additionally, to quantify the coupling degree in semi-open field scenarios, the power ratio parameters between coil transmission and plate transmission were defined, thus realizing the induction recognition under migration conditions. Geomagnetic field disturbance signals captured by geomagnetic sensors during different vehicles passage were taken as an example, and an ensemble empirical mode decomposition (EEMD) method for geomagnetic signals was applied to enhance the extraction effect of nonlinear and non-stationary signals. Furthermore, by introducing a feature vector extraction method for curves and taking small three-box sedans, medium two-box sedans, medium van trucks, and large carriages as typical test samples, the geomagnetic curve signals of various vehicles were transformed into feature vector spectra to determine the vehicle shape types. Research results show that under test conditions, the recognition length of the hybrid electromagnetic induction unit in the direction of coil migration is about 220 mm, the recognition length perpendicular to the direction of outward coil migration is about 170 mm, and the recognition range increases by approximately 62.8% compared to the single magnetic field coupling. Meanwhile, the geomagnetic field induction unit can detect the characteristics of vehicles with lengths ranging from 3.7 to 12.0 m and speeds ranging from 2.78 to 16.67 m·s^-1. The reliability of dynamic classification and recognition for wireless charging vehicles can be effectively enhanced by the synergistic cooperation between the geomagnetic field induction unit and the hybrid electromagnetic induction unit, thus promoting the application and development of wireless charging technology in road traffic electrification facilities.

2024, 24(4): 195-207. doi: 10.19818/j.cnki.1671-1637.2024.04.015

FullText HTML (38) PDF (14017KB) (26)

Review on pilot-in-the-loop modeling techniques facing integrated operation

WEI Lin, YANG Ji-rui, LI Xiu-yi, XIAO Yue, ZHENG Yuan, LI Cheng-long

Abstract: More> The pilot models facing integrated operation environments involving manned aerial vehicle (MAV) and unmanned aerial vehicle (UAV) were studied. The development history of pilot modeling technologies and the characteristics of each model were introduced according to the two types of control, including cockpit piloting of MAV and remotely controlled piloting of UAV. The pilot-in-the-loop response characteristics of MAV and UAV were analyzed by simulation software, and the influence of communication delay on the piloting loop of UAV was discussed. The applicabilities and shortcomings of various pilot models in different task scenarios were summarized. Analysis results show that the bottlenecks of pilot modeling facing integrated operation environments mainly lie in the heterogeneous control loops between MAV and UAV pilots, the lack of situational awareness of UAV pilots, and the uncertainty of delay in command and control link (C2 link) of UAV system. MAV pilot modeling generally describes the human body structure and the pilot's control characteristics according to control theory methods. Moreover, with the development of artificial intelligence, pilot model design methods based on intelligent control theory can better describe the pilot's control characteristics. Due to the different positions of UAV pilots in the control loop, remotely controlled piloting relies on the C2 link to perceive UAV status and upload control commands. Therefore, UAV pilot modeling focuses on the expression of human-in-the-loop, and the matching pilot model should be designed according to the application scenario, so as to more truly reflect the pilot's control and decision-making behavior characteristics in a specific scenario. For future integrated operation environments, the human-in-the-loop model construction should focus on how to describe decision-making differences caused by the lack of situational awareness of UAV pilots during remotely controlled piloting, how to better fit human-machine system control characteristics, and how to build an efficient and reliable pilot model with low-delay C2 link.

2024, 24(4): 208-227. doi: 10.19818/j.cnki.1671-1637.2024.04.016

FullText HTML (30) PDF (6434KB) (25)

Operating speed prediction models of trucks at interchange ramps based on high-frequency GPS data

ZHANG Min, LIU Kai, ZHANG Chi, XI Sheng-yu, NIE Yu-han

Abstract: More> To clarify the operating speed rules for trucks at interchange ramps, operating speed models for trucks on interchange ramps were constructed based on the analysis of the measured high-frequency GPS data of trucks from the mainline to the end of the ramp on an expressway. Through the analysis of the measured speeds of trucks, the characteristic points of the truck operating speeds at the ramp were determined, and correlation analysis was carried out for design elements that might be related to the speed at the characteristic points. Multiple operating speed models for trucks at characteristic points were established by using all subsets regression methods. By comparing the Akaike information criterion, mallows's C_p statistic, and model test values of different models, the parameters of the independent variables were determined, and prediction models for truck operating speed at various locations were developed, including the small nose point of interchanges, the midpoint of the ramp curve, and the merging nose of the ramps and the connecting lines. These models were validated by using data from four ramps. Research results show that the small nose point of interchanges, the midpoint of the ramp curve, and the merging nose of the ramps and the connection lines can be considered as the characteristic points of truck operating speed at ramps. The radius of the ramp curve, the gradient rate of the exit, and the operating speed of the diversion point have significant effects on the operating speed at the small nose point. The radius, the distance from the midpoint to the merging nose of the ramp, and the operating speed at the small nose point have significant effects on the operating speed at the midpoint of the ramp curve. The longitudinal slope in the first half of the midpoint, the curvature of the curve, and the operating speed at the midpoint have significant effects on the operating speed at the merging nose. The correlation coefficients of the operating speed prediction models at the three characteristic points are 0.988, 0.993, and 0.990, respectively. The mean absolute percentage errors between the predicted and measured values are less than 10%, which meet the requirements of model accuracy.

2024, 24(4): 228-242. doi: 10.19818/j.cnki.1671-1637.2024.04.017

FullText HTML (38) PDF (15722KB) (24)

Spatio-temporal traffic data prediction based on low-rank tensor completion

ZHAO Yong-mei, DONG Yun-wei

Abstract: More> To dynamically evaluate traffic condition in real time, a traffic speed prediction model based on autoregressive regularization terms and Laplacian regularization terms was proposed. To improve the model's expression capability in global dimensions, a Laplace convolutional regularization term based on a low-rank tensor completion framework was introduced to represent the correlations of road segments. To improve the model's expression capability in local spatial dimensions, the time series trend-capturing capabilities of autoregressive models were utilized, and the short- and long-term expression capabilities of the models in the time dimension were improved to capture the spatio-temporal information of traffic data more effectively. The implementation of the truncated kernel norm as the low-rank tensor approximation model and the conversion of time- and frequency-domain signals leaded to improve the computation efficiency. An efficient low-rank Laplacian autoregressive tensor completion (LLATC) prediction method was developed by using the alternating direction multiplier method. Based on taxi speed data set and expressway traffic speed data set, the completion performances of the LLATC algorithm under different missing rates were systematically analyzed, and the prediction accuracy of the LLATC algorithm was compared with other baseline prediction algorithms. Research results show that under the random missing pattern with a missing rate of 20% to 70%, the mean absolute error (MAE) of the LLATC algorithm reduces by 2% to 6% compared to the traditional low-rank tensor completion models, and the MAE reduces by 4% to 22% compared to the traditional prediction methods. Under the non-random missing pattern, the MAE of the LLATC algorithm reduces by 2% to 6% compared to the traditional low-rank tensor completion models, and the MAE reduces by 13% to 25% compared to the traditional prediction methods. The finding indicates that the LLATC algorithm effectively reduces the completion error of traffic volume data, significantly enhances the prediction accuracy of traffic volume data under two kinds of missing data patterns, and simplifies the data processing workflow.

2024, 24(4): 243-258. doi: 10.19818/j.cnki.1671-1637.2024.04.018

FullText HTML (39) PDF (16169KB) (36)

User Center

Journal Info

Collection of Source JournalsMore>

Current Issue
2024, Volume 24, Issue 4

Search Type

Search Journal

LinksMore>