Please wait a minute...
Information

Journal of Jilin University(Engineering and Technology Edition)
ISSN 1671-5497
CN 22-1341/T
主 任:陈永杰
编 辑:张祥合 曹 敏  程仲基
    赵莹莹 赵浩宇
电 话:0431-85095297
E-mail:xbgxb@jlu.edu.cn
地 址:长春市吉林大学南岭校区
    逸夫教育大楼B823室
WeChat

WeChat: JLDXXBGXB
随时查询稿件状态
获取最新学术动态
Table of Content
01 June 2025, Volume 55 Issue 6
Research progress on the performance changes of lithium⁃ion batteries after aging
Xue-wei SONG,Ze-ping YU,Yang XIAO,De-ping WANG,Quan YUAN,Xin-zhuo LI,Jia-wen ZHENG
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1817-1833.  DOI: 10.13229/j.cnki.jdxbgxb.20240763
Abstract ( 2 )   HTML ( 0 )   PDF (4643KB) ( 1 )  

To clarify the current research status of mechanical, electrical, and thermal performance changes of batteries after aging at home and abroad, and to provide reference for battery performance prediction and experimental design, battery aging is linked to battery performance. The internal mechanism of battery aging is summarized, and the performance changes of batteries under normal aging and abnormal aging conditions are sorted out. The characteristics of various performance parameters after aging are explored. From the perspectives of battery components and individual cells, it was found that the mechanical properties of lithium-ion batteries deteriorate to varying degrees with aging; Starting from the changes in battery capacity and impedance, describe the decrease in performance of aging batteries; Using the characteristic temperature of thermal runaway as an indicator of battery thermal safety performance, explain the changes in thermal safety of aging batteries; Finally, the future development direction of battery performance research was discussed.

Figures and Tables | References | Related Articles | Metrics
Research progress on modification mechanism, preparation and performance of waste rubber powder modified asphalt
Wan-feng WEI,Hong-gang ZHANG,Yang-peng ZHANG,Fan YANG,Bo-ming TANG,Ling-yun KONG
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1834-1853.  DOI: 10.13229/j.cnki.jdxbgxb.20240796
Abstract ( 2 )   HTML ( 0 )   PDF (5960KB) ( 1 )  

In order to further promote the application of waste rubber powder modified asphalt in road engineering, the composition and characteristics of waste rubber powder were systematically sorted and analyzed. It is believed that the physical and chemical properties of waste rubber powder have a significant impact on the modification of asphalt, and activation can greatly improve the compatibility between waste rubber powder and asphalt. It is pointed out that the modification of asphalt by waste rubber powder is the result of physical-chemical dual modification interaction, but the internal mechanism of the interaction between waste rubber powder and asphalt is not clear. The current research status on the main influencing factors of the performance of waste rubber powder modified asphalt, and the performance of activated waste rubber powder modified asphalt and polymer/waste rubber powder composite modified asphalt is sorted out. Finally, a summary and outlook were provided on the desulfurization and activation control of waste rubber powder, the utilization of multi-source waste tires, the modification mechanism of waste rubber powder, and the composite modification of asphalt with polymer/waste rubber powder, suggestions include: ① research on the factory activation method of waste rubber powder, focusing on the balance between the performance of waste rubber powder modified asphalt and the activation of rubber powder; ② study the interaction process between waste rubber powder-asphalt, as well as the molecular structure-activity relationship of waste rubber powder modified asphalt, to reveal the intrinsic mechanism of the interaction between waste rubber powder and asphalt; ③ further research on the compatibility of various types of waste tire rubber powder with asphalt; ④ development of rubber asphalt series products for multi-scenario applications; ⑤ in-depth study on the aging behavior of waste rubber powder modified asphalt.

Figures and Tables | References | Related Articles | Metrics
Air⁃fuel ratio control of gasoline engines based on Gaussian process regression intake prediction
Jing-hua ZHAO,Da LIU,Yu-qi ZHOU,Long WEN,Qian-yu LIU,Jie LIU,Fang-xi XIE
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1854-1861.  DOI: 10.13229/j.cnki.jdxbgxb.20240829
Abstract ( 3 )   HTML ( 0 )   PDF (1598KB) ( 2 )  

To address the problem of insufficient analysis of continuous changes and randomness in traditional model-based prediction techniques, a gasoline engine air-fuel ratio feedback control method based on Gaussian Process Regression (GPR) intake quantity prediction is proposed. The simulation analysis results show that compared with the real-time feedback control method of the intake air sensor, the lambda average error of the control method proposed in this paper is reduced by 12% and 29% respectively under the transient operating conditions of the two engines, effectively improving the control accuracy of the air-fuel ratio, while also having strong anti-interference ability.

Figures and Tables | References | Related Articles | Metrics
Multi⁃physics simulation method of vehicle motor under varying working conditions based on multi⁃software combination
Mei-xia JIA,Jian-jun HU,Feng XIAO
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1862-1872.  DOI: 10.13229/j.cnki.jdxbgxb.20230968
Abstract ( 1 )   HTML ( 0 )   PDF (3353KB) ( 1 )  

A multi software based multi physics simulation method for automotive motors under variable operating conditions was proposed. By combining finite element and lumped parameter methods, both finite element and variable operating condition simulations can be carried out simultaneously, making the simulation process closer to the actual working process. On the basis of analyzing the temperature field and electromagnetic field of the motor, a mathematical model coupling multiple physical fields was decomposed. A multi software joint simulation model was established using Maxwell Simplorer Simulink, and the finite element model was reduced using the equivalent current extraction method, greatly improving the computational speed of the model. The joint simulation model is a fully coupled model that can achieve data exchange and transmission between different software models, thereby realizing transient simulation of the motor under different operating conditions. By using the joint simulation method, synchronous simulations of multiple physical fields can be obtained simultaneously, enabling real-time data exchange of the joint simulation model and avoiding errors caused by asynchronous simulation of non fully coupled models. Finally, the output torque of the motor was verified through experiments, and the torque error between the model and the experiment was less than 3%, proving the effectiveness and accuracy of the multi field coupled multi software joint simulation method.

Figures and Tables | References | Related Articles | Metrics
Fuzzy energy management strategy of fuel cell electric vehicle based on improved pigeon⁃inspired optimization
Chun XIAO,Zi-chun YI,Bing-yin ZHOU,Shao-rui ZHANG
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1873-1882.  DOI: 10.13229/j.cnki.jdxbgxb.20230926
Abstract ( 2 )   HTML ( 0 )   PDF (1725KB) ( 1 )  

A composite fuzzy energy management strategy was proposed with the goal of improving the lifespan of auxiliary energy source power batteries. The improved pigeon swarm optimization algorithm (IPIO) was used to update the fuzzy membership function, while ensuring that the power battery operates in a suitable range for a long time and reducing equivalent hydrogen consumption. The existing ADVISOR model was developed to establish a simulation model for the FCEV hybrid power system, and was conducted simulation experiments under two operating conditions: NEDC and CLTC-P. The results show that the charging speed of the IPIO-enhanced energy management strategy is more than twice as fast as the power-following strategy when the initial State of Charge (SoC) is low, enabling a faster transition to the optimal SoC range and prolonging battery lifespan. When the initial SoC is high, the equivalent hydrogen consumption of the IPIO-enhanced composite fuzzy energy management strategy is reduced by 11.8% and 9.09% compared with before under two driving cycles, significantly reducing hydrogen consumption and enhancing the economy of hydrogen fuel cell vehicles.

Figures and Tables | References | Related Articles | Metrics
Rolling bearing fault diagnosis based on variational mode extraction and lightweight network
Zhi-gang FENG,Shou-qi WANG,Ming-yue YU
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1883-1891.  DOI: 10.13229/j.cnki.jdxbgxb.20231047
Abstract ( 2 )   HTML ( 0 )   PDF (2904KB) ( 1 )  

A rolling bearing fault diagnosis method combining variational mode extraction (VME) and lightweight convolutional neural network (CNN) was designed to solve the problems of low diagnostic performance of CNN in complex industrial environments as well as the problem of large number of parameters. VME was used to extract the desired modes in the vibration signals collected from multiple sensors and construct the multi-sensor grayscale feature maps to eliminate information interference while enabling data fusion. The residual structure and ultra-lightweight subspace attention module(ULSAM) are introduced on the basis of SqueezeNet to construct a lightweight residual attention convolutional neural network (LRACNN). The method has a high fault recognition rate and diagnostic stability in complex environments.

Figures and Tables | References | Related Articles | Metrics
Effect of parasitic load components with multi⁃directional motion on weight⁃sensing outputs
Gui-yong GUO,Jian-feng ZHONG,Qiu-kun ZHANG,Bao-jie CAI
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1892-1905.  DOI: 10.13229/j.cnki.jdxbgxb.20231023
Abstract ( 3 )   HTML ( 0 )   PDF (6526KB) ( 1 )  

For the parasitic load component generated by multi-component superposition effect has a great influence on the weight-sensing output, a variable parameter and multi-component output effect assessment method based on the overall structure was proposed. Subsequently, a mathematical theoretical model of the instantaneous characteristics of "strain-total weight-motion" was established and analyzed by finite element simulation. In addition, exploring the characteristics of the multi-directional motions with transverse, longitudinal and rotating around an axis that are commonly seen in the weighing process, which include parasitic effects such as changing the total weight, rotating speed and angular acceleration, etc. Eventually, the reasonableness of the theoretical model and simulation analysis was verified through experimental tests. The results showed that the parasitic load component of transverse motion had the smallest influence on the output, and the maximum deviation was no more than 0.001%FS ( Full Scale); followed by longitudinal motion, and the maximum deviation was no more than 0.015%FS; and the parasitic load component of themotion of rotating around an axis had the largest influence on the output, and the maximum deviation can be up to 0.5%FS, and it was changed with the differences of the total weight and the rotating speed, which provided a technical support and accuracy compensation technology for dynamic weighing application of the weight-sensing structure. This law provides the technical support of accuracy compensation and scientific evaluation basis for the dynamic weighing application of weight-sensing structure.

Figures and Tables | References | Related Articles | Metrics
Global reliability sensitivity analysis of deformation for machine tool rotary table
Xian-zhen HUANG,Ming-fei MA,Chao LI,Xu WANG,Zhi-ming RONG
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1906-1914.  DOI: 10.13229/j.cnki.jdxbgxb.20230947
Abstract ( 3 )   HTML ( 0 )   PDF (1407KB) ( 1 )  

In view of the problem that the rotary table will deform due to bearing large loads during machine tool processing, which will affect the processing accuracy and consistency,this paper proposes a reliability analysis model for the deformation behavior of the rotary table that considers the influence of random factors. The model treats the rotary table as a system, and uses the maximum deformation of the rotary table as the criterion for reliability analysis. The study explores the global reliability changes of the entire system under different load conditions and calculates the global sensitivity of each parameter to system failure. The results indicate that the material parameters of the table plate, rotary seat, and sliding seat have a significant impact on the reliability of the rotary table system, with the sensitivity index of the elastic modulus being greater than that of density and Poisson's ratio being the smallest.

Figures and Tables | References | Related Articles | Metrics
Influence of contact thermal resistance on temperature rise characteristics of high⁃speed ball screw
Ge-dong JIANG,Hao WANG,Ya-bin JING
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1915-1922.  DOI: 10.13229/j.cnki.jdxbgxb.20230938
Abstract ( 2 )   HTML ( 0 )   PDF (1923KB) ( 1 )  

Based on the MB fractal theory, the contact thermal resistance of the raceway contact surface was determined, and a finite element model for thermal analysis of high-speed ball screw pairs considering contact thermal resistance was established. The temperature rise characteristics of ball screw pairs under different contact thermal resistances were analyzed through simulation calculations. The results indicate that the highest temperature of the screw is concentrated at the joint between the front bearing and the ball raceway, while the peak temperature of the ball and nut is concentrated in the middle of the raceway. The change in contact resistance has a significant impact on the steady-state temperature of the ball and nut. As the contact resistance increases, the steady-state temperatures of the ball and nut both increase significantly. The above work provides a theoretical basis for studying the temperature rise control and cooling lubrication methods of high-speed ball screws.

Figures and Tables | References | Related Articles | Metrics
Development and application of online measurement device for water content of lubricating oil based on microwave resonance technology
Tao ZHANG,Qin JIANG,Jie LIU,Zi-jian DING,Xue-mei HU,Bing HAN
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1923-1930.  DOI: 10.13229/j.cnki.jdxbgxb.20230960
Abstract ( 3 )   HTML ( 0 )   PDF (1493KB) ( 1 )  

In view of the shortcomings of the current lubricating oil water content detection technology with expensive equipment and few online measurements, an online measurement device based on microwave resonance technology was designed. It consists of resonator, microcontroller unit, coaxial transmission line, voltage-controlled oscillator, logarithmic detector and PC. Combined with the finite element simulation software HFSS, a microwave resonator was designed, and the S-parameter response curve was simulated under the state of uniform and non-uniform distribution of water and oil, and the relationship between resonant frequency and water content was obtained. In the experiment, the measurement device was fabricated to measure the relationship between resonance frequency and water content in the range of 0~1.2%, and the results are consistent with the simulation. Temperature compensation was carried out within 16~30 ℃ to improve the measurement accuracy. Finally, the device was connected to the oil pipe for on-line measurement, and the results prove the effectiveness of the measurement device.

Figures and Tables | References | Related Articles | Metrics
Microstructure of titanium alloy heat affected zone repaired by laser deposition and properties of subsequent heat treatment
Jin-lan AN,Lan-bin WANG,Song ZHOU,Yan-qing HUANG
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1931-1939.  DOI: 10.13229/j.cnki.jdxbgxb.20231006
Abstract ( 1 )   HTML ( 0 )   PDF (5073KB) ( 1 )  

In view of the characteristics of titanium alloy parts with good performance but low material utilization rate and damage in harsh environments, laser deposition repair technology was used to quickly repair them. The effects of different heat treatments on the microstructure and mechanical properties of titanium alloy(Ti-6.5Al-2Zr-1Mo-1V) repaired by laser deposition were studied. The results show that the forging zone has a bi-modal microstructure, the laser repair zone has a basketweave microstructure, the Ghost α phase was formed in the bottom heat-affected zone, and a large number of massive β phases were formed in the top heat-affected zone. Tensile tests show that its mechanical properties were generally high strength and low plasticity, and were all ductile fractures. The heat treatment method of 920 ℃/1 h achieves a good match between strengthand plasticity. Through Digital image processing technology analysis, it was found that the laser repair zone determines the overall performance of the entire sample, and the thinner lamellar α and the short rod-shaped α phase precipitated in the β matrix help to improve the mechanical properties.

Figures and Tables | References | Related Articles | Metrics
Effect of inoculation treatment on thermal conductivity and tensile strength of high carbongray cast iron
Jin-guo WANG,Cheng-gang WANG,Tian-shi LU,Jian-dong WANG,Feng LI,Tie-fang CHENG,Rui-fang YAN
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1940-1947.  DOI: 10.13229/j.cnki.jdxbgxb.20230915
Abstract ( 2 )   HTML ( 0 )   PDF (6126KB) ( 1 )  

The effects of different inoculants on the thermal conductivity of high carbon gray cast iron were studied. DRPL-2C thermal conductivity tester was used to determine the thermal conductivity of gray cast iron at room temperature,WAW-200 tensile testing machine was used to test the tensile strength, XJG-0.5 optical microscope, TESCAN tungsten filament scanning electron microscopy were used for tissue observation, and graphite characteristic parameter statistics (5-8 photos of gray cast iron with different inoculant treatment gray cast iron ) were carried out by using Image-pro plus (IPP), Photoshop and other software. The results showed that different inoculants had different degrees of influence on the graphite microstructure characteristics (graphite tip morphology, graphite length, graphite quantity and graphite proportion) of high carbon ash cast iron, thereby affecting the thermal conductivity of high carbon ash cast iron. In addition to the content of graphite in gray cast iron, the length of graphite and the amount of graphite also have an important impact on the thermal conductivity of high carbon gray cast iron. Compared with the traditional inoculant, the new inoculant can effectively passivate the tip of the graphite, reduce the size and increase the amount of graphite in the gray cast iron structure, and synergistically improve the tensile strength and thermal conductivity of the gray cast iron, which provides some theoretical guidance for the future application of high-carbon gray cast iron in the field of high-strength and high-thermal conductivity materials.

Figures and Tables | References | Related Articles | Metrics
Strategies for controlling vehicle movements at signalfree intersections in intelligent networked environment
Fu-quan PAN,Yuan-zheng NIU,Li-xia ZHANG,Jin-shun YANG,Xiu-feng CHEN,De-qi CHEN
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1948-1962.  DOI: 10.13229/j.cnki.jdxbgxb.20221102
Abstract ( 3 )   HTML ( 0 )   PDF (3603KB) ( 1 )  

In order to realize the intelligent control of connected and autonomous driving vehicle in the intersection without signal in the intelligent network connection environment and improve the intersection passage efficiency, a vehicle passage control strategy based on the gap theory was proposed. According to the function and usage of the intersection area, the intersection area was divided into a change zone, a regulation zone, a buffer zone, a physical zone and a recovery zone. A vehicle conflict zone calculation model for the physical zone was established by considering the physical size of real vehicles, and a mathematical model for the clearance control of straight-straight, straight-left-turn and left-turn-left-turn vehicles was developed by optimising the trajectory of left-turn vehicles as an elliptical trajectory. A vehicle speed induction model for the regulation zone and buffer zone was established based on the trigonometric acceleration control strategy. The use of the efficiency and rationality of the control strategy and model were compared and verified by using joint simulation of Vissim and Matlab. The results show that the proposed control strategy and model can enable the conflicting vehicles to pass through the conflicting area sequentially without stopping; comparing with the signal control strategy, the average delay time of vehicles through the intersection is reduced by 55.97%, the average travel time is reduced by 41.87%, and the vehicle energy consumption is reduced by 33.31% under this control strategy and model at a traffic volume of 1 600 pcu/h, and the higher the traffic volume, the more significant the improvement effect is.

Figures and Tables | References | Related Articles | Metrics
Trajectory prediction model for intelligent connected vehicle
Jian WANG,Chen-wei JIA
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1963-1972.  DOI: 10.13229/j.cnki.jdxbgxb.20231046
Abstract ( 2 )   HTML ( 0 )   PDF (2214KB) ( 1 )  

In contrast to traditional single-vehicle intelligent autonomous driving systems, which can only make predictions about the future based on their own perception of the environment, intelligent connected autonomous driving systems have the capability to enhance predictions by incorporating additional dynamic information about the surrounding road environment through V2X technology. Building upon the foundation of single-vehicle intelligent trajectory prediction, a specialized encoder was employd to enable the trajectory prediction model to seamlessly fuse its own perceptual information with dynamic road data obtained via V2X communication. The experimental results on the CARLA simulation dataset demonstrate that using V2X technology to obtain dynamic information of the surrounding road environment can more accurately predict vehicle trajectories compared to trajectory prediction algorithms that do not use dynamic environment information.

Figures and Tables | References | Related Articles | Metrics
Highway infrastructure performance and traffic state prediction on road network
Zhen YANG,Rui-ping ZHENG,Zhe GONG
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1973-1983.  DOI: 10.13229/j.cnki.jdxbgxb.20230981
Abstract ( 3 )   HTML ( 0 )   PDF (2312KB) ( 1 )  

By establishing the prediction model for the highway infrastructure performance that considered the temporal variation of traffic status and the feedback model for traffic status that considered the decay of highway infrastructure performance, the mutual influence between the two was explored. By integrating the expression and storage of road network topology, the coupled simulation model was constructed. Using the variable step-size discrete event simulation method and Matlab Simulink for decoupling on PC platform, the coupled simulation of the highway infrastructure performance and traffic status was implemented. The application example demonstrates that the simulation model can predict highway infrastructure performance well, and the accuracy is higher than that of the traditional prediction without considering the coupling.

Figures and Tables | References | Related Articles | Metrics
Scheduling algorithm for battery electric vehicle in closed scenic area
Sheng-yu YAN,Ming-jie CHENG,Hong-ce TIAN,Hong-yu WANG,Yong-heng ZHOU,Bo-hao MA
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1984-1993.  DOI: 10.13229/j.cnki.jdxbgxb.20231277
Abstract ( 1 )   HTML ( 0 )   PDF (3072KB) ( 1 )  

To meet the scheduling needs of battery electric vehicle (BEV) in closed scenic areas, a multi-objective scheduling model was proposed. With the goal of optimizing the operating costs of BEV procurement, frequency of departure, stopping time, and charging price difference, a departure schedule solving algorithm was designed based on UI rules. Heuristic algorithms were used to solve the train number chain set, and a BEV performance testing plan was designed. By limiting the driving speed of the test sample vehicle, the single round trip time was obtained, and a maximum round trip calculation method combining CRUISE simulation and real vehicle testing was proposed. Taking the south line of Mount Wutai scenic spot as an example, the feasibility of BEV scheduling model and solution algorithm was verified. The results indicate that the UI rule-based time slot BEV scheduling algorithm can achieve minute level BEV departure schedules. The deviation rate between the calculated number of BEV cars purchased on the example route and the ideal minimum number of cars purchased is 2.99%, with a solution time of 0.89 seconds. When simulating a scheduling plan with a daily passenger flow from 3 000 to 30 000, the maximum deviation rate of actual transportation capacity supply and demand is 1.00%. The research results can be applied to the BEV dynamic scheduling algorithm and vehicle scale calculation model for enclosed scenic spots.

Figures and Tables | References | Related Articles | Metrics
Digital twin driven longitudinal and lateral control of truck platoon
Shu-you YU,Hua-cheng XIE,Wen-bo LI,Yong-fu LI,Hong CHEN,Bao-jun LIN
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  1994-2002.  DOI: 10.13229/j.cnki.jdxbgxb.20231367
Abstract ( 0 )   HTML ( 0 )   PDF (5971KB) ( 1 )  

A vehicle platoon cooperative control system based on digital twin was proposed, considering the longitudinal and lateral motion. The proportional-integral-derivative (PID) control was designed to ensure stable driving of the convoy. The linear quadratic regulator (LQR) control was designed to ensure the lateral lane tracking performance. A digital twin simulation scenario was build using Prescan, TruckSim, and Matlab/Simulink, and the longitudinal following and lateral lane tracking performance of vehicle queues was dynamically simulated based on longitudinal and lateral control. The digital twin comprehensively debugs and optimizes the control strategy and parameters by remotely controlling the vehicle platoon and monitoring indicators such as longitudinal and lateral velocity, lateral position and yaw angle deviation of the platoon. The simulation results show that the digital twin driven truck platoon control system has good tracking performance and lane-keeping performance.

Figures and Tables | References | Related Articles | Metrics
Expressway small object detection algorithm based on deep learning
Hui-zhi XU,Dong-sheng HAO,Xiao-ting XU,Shi-sen JIANG
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2003-2014.  DOI: 10.13229/j.cnki.jdxbgxb.20230939
Abstract ( 3 )   HTML ( 0 )   PDF (1889KB) ( 1 )  

To address the challenging issue of real-time detection of small distant pedestrians and vehicles in images captured by roadside cameras on expressways, an improved object detection algorithm YOLOv5s-3S-4PDH was proposed. Firstly, the Shufflenetv2-Stem-SPPF network structure was used to improve the running speed of the algorithm. Secondly, the accelerated normalized weighted fusion feature map and the 160×160 small object detection layer were introduced to optimize the performance of small object detection; Then, the improved decoupling head mechanism was introduced to improve the localization and classification accuracy of small object detection. Finally, Focal EIoU was used as the localization loss function of the algorithm to accelerate the training convergence speed of the algorithm. The results show that: compared with the YOLOv5s on the self-built pedestrian and vehicle dataset, the computation and parameter amount of the proposed algorithm are reduced by 10.1% and 24.6%, respectively, and the detection speed and accuracy are increased by 15.4% and 2.1%, respectively; Transfer learning experiment on the VisDrone2019 dataset shows that the proposed algorithm has better average precision for all categories. The proposed algorithm not only meets the real-time and accuracy requirements of small object detection, but also has generalization ability.

Figures and Tables | References | Related Articles | Metrics
Curve lattice model for connected commercial vehicles based on density dispersion and information transmission delay
Hong-zhuan ZHAO,Ze-jian WU,Xin ZHANG,Sheng-wen SHI,Wen-yong LI,Xin ZHAN,En-yong XU,Jia-ming WANG
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2015-2029.  DOI: 10.13229/j.cnki.jdxbgxb.20230987
Abstract ( 1 )   HTML ( 0 )   PDF (3287KB) ( 1 )  

In order to solve the traffic congestion problem of mixed traffic flow with connected commercial vehicles in the curve environment and improve the stability of mixed traffic flow with connected commercial vehicles in the curve environment, a lattice model of connected commercial vehicles based on density dispersion effect and information transmission delay effect (DDITD) in curve environment was proposed. The density dispersion effect was first proposed, which reveals the influence of the following characteristics of connected commercial vehicles on the density distribution of mixed traffic flow. The density dispersion effect and the information transmission delay effect were introduced into the curve lattice model at the same time, which expands the applicable scope of the lattice model. Firstly, the turn correction coefficient in DDITD model was calibrated through real vehicle test, and then the DDITD model was analyzed by linear stability analysis to study the influence of density dispersion and information transmission delay on the stability of traffic flow under the curve environment. Secondly, the reduced perturbation method was applied through nonlinear stability analysis. The mKdV (modified Korteweg de Vries) equation was derived to describe the critical point of traffic density wave evolution. Finally, the theoretical results were verified by numerical simulation, and the results show that the density dispersion effect and the information transmission delay effect can effectively alleviate the traffic jam on the curve. The research results provides a new method for studying the following characteristics of connected commercial vehicles and the stability of mixed traffic flow, and provides a new ideas and basis for traffic management and control in curved environments.

Figures and Tables | References | Related Articles | Metrics
Bus schedule optimization considering bus and metro interchange needs
Yuan-wen LAI,Yan-sheng CHEN,Shu-yi WANG,Yu-long ZHANG,Xin-yun ZHU
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2030-2037.  DOI: 10.13229/j.cnki.jdxbgxb.20230931
Abstract ( 1 )   HTML ( 0 )   PDF (804KB) ( 1 )  

An all-day bus schedule optimization method was proposed in consideration of passengers' demand between the bus and the metro. Firstly, with the goal of minimizing the cost of passenger waiting time and the operating costs of public transportation enterprises, an optimization model was constructed. Then, a genetic-simulated annealing hybrid algorithm was designed to solve the model. A case study of the upward direction of the No. 1 bus in Fuzhou was selected to verify the effectiveness and practicability of the proposed model and algorithm. After optimizing the schedule, the average waiting time cost of passengers in B-M mode, M-B mode, and non-transfer mode has been reduced by 19.30%. Public transportation enterprises' operating costs have increased by 4.95% due to increased bus departures, but the total system cost has decreased by 5.05%.

Figures and Tables | References | Related Articles | Metrics
Hybrid strategy improves WOA⁃BiLSTM speed prediction of expressway exit ramp
Qing-ling HE,Yu-long PEI,Lin HOU,Jing LIU,Sheng PAN
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2038-2049.  DOI: 10.13229/j.cnki.jdxbgxb.20230940
Abstract ( 2 )   HTML ( 0 )   PDF (1971KB) ( 1 )  

Aiming at the problem that the existing meta-heuristic algorithm has slow convergence speed and large error in the process of optimizing the neural network to predict vehicle speed, a vehicle speed prediction method for expressway off-ramp based on IWOA-BiLSTM was proposed. Firstly, the Circle chaotic map was used to replace the randomly generated initial population in the whale optimization algorithm to increase the diversity and quality of the population.Secondly, the elite opposition-based learning strategy was used to improve the diversity of the individual 's preferred position and reduce the risk of the algorithm falling into local optimum and premature convergence.Finally, the cosine function was used to change the adaptive convergence factor and introduce the inertia weight strategy. On the premise of retaining the advantages of the whale optimization algorithm, the global search and local development capabilities of the algorithm were balanced. The simulation results show that, compared with the existing meta-heuristic algorithm and vehicle speed prediction model, the IWOA algorithm has significantly improved in terms of optimization accuracy, convergence rate and prediction accuracy.

Figures and Tables | References | Related Articles | Metrics
Damage mechanism of FRP reinforced concrete under alkali freezing coupling effect
Wen-yuan XU,Wei LI,Da-yang WANG,Yong-cheng JI
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2050-2062.  DOI: 10.13229/j.cnki.jdxbgxb.20231016
Abstract ( 2 )   HTML ( 0 )   PDF (5318KB) ( 1 )  

The degradation law of carbon/basalt/glass/aramid fiber-reinforced concrete under strong alkali solution and freeze-thaw coupling was explored. Fiber reinforced polymer (FRP) was used for full reinforcement of cylindrical axial compression members, and local reinforcement was used for prismatic bending members. The mass loss rate, dynamic elastic modulus, pH value change, compressive and flexural strength of the specimens under alkali frost coupling were tested. The results show that carbon fiber and aramid fiber reinforced specimens have better quality loss rate, dynamic elastic modulus, compressive strength loss, plasticity, and flexural bearing capacity loss than glass fiber and basalt fiber reinforced specimens. Based on experimental data, the Lam Teng constitutive model of FRP reinforced concrete with alkali frost coupling effect was modified.

Figures and Tables | References | Related Articles | Metrics
Algorithm for detecting corrosion points in bridge steel structures based on synthetic aperture ultrasonic imaging
Feng SHI,Peng NIU
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2063-2068.  DOI: 10.13229/j.cnki.jdxbgxb.20240556
Abstract ( 3 )   HTML ( 0 )   PDF (1485KB) ( 1 )  

to address the problem of non-linear organization arrangement characteristics such as texture and unevenness on the surface of bridge steel structures, which increase the difficulty of corrosion point detection, a corrosion point detection algorithm based on synthetic aperture ultrasonic imaging technology was designed. Firstly, the multi element synthetic aperture focusing technology was adopted to obtain ultrasonic images of bridge steel structures. By focusing ultrasonic energy, the resolution and clarity of the ultrasonic images were improved, making it easier to identify and locate corrosion points. Secondly, using gray level co-occurrence matrix technology to capture the spatial relationships between pixels in the image, the arrangement features of the steel structure organization on the surface of the bridge were extracted. Finally, the Fisher discriminant criterion was used to remove features with low or redundant contribution to the detection, and the filtered features were input into the neural network to accurately detect corrosion points using the non-linear mapping ability of the neural network. The experimental results show that after applying the algorithm, the position and diameter distance of the corrosion points can be clearly observed. The diameter, inclination angle, and depth of the corrosion points detected by the algorithm are basically consistent with the actual values, indicating the effectiveness of the algorithm.

Figures and Tables | References | Related Articles | Metrics
Graph node classification algorithm based on similarity random walk aggregation
Xiang-jiu CHE,Yu-peng SUN
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2069-2075.  DOI: 10.13229/j.cnki.jdxbgxb20231043
Abstract ( 2 )   HTML ( 0 )   PDF (591KB) ( 1 )  

In addressing the issue of relatively low accuracy in node classification tasks on heterophily graphs using methods such as MLP and GCN, a Graph Neural Network based on Similarity Random Walk Aggregation (SRW-GNN) was proposed. To address the impact of heterophily on node embeddings, SRW-GNN employs the similarity between nodes as probabilities for conducting random walks. The sampled paths serve as the neighborhood, enabling the model to gather more homophily-based information. To address the issue of insensitivity to node order in most existing graph neural network (GNN) aggregators, a path aggregator based on recurrent neural networks (RNNs) was introduced to simultaneously extract features and order information of each node in the path. Furthermore, nodes exhibit varying preferences for different paths. To adaptively learn the importance of different paths in node encoding, an attention mechanism was employed to dynamically adjust the contributions of each path to the final embedding. Experimental results on several commonly used heterophily graph datasets demonstrate that the proposed SRW-GNN method achieves significantly higher accuracy compared to the methods such as MLP, GCN, H2GCN, HOG-GCN, validating its effectiveness in heterophily graph node classification tasks.

Figures and Tables | References | Related Articles | Metrics
Multi spectral image fusion algorithm for unmanned aerial vehicles based on gradient consistency constraint
Yi TANG,Bing-chuan LU,Hong-chen YI,Cheng YU,Bin NAN
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2076-2081.  DOI: 10.13229/j.cnki.jdxbgxb.20240525
Abstract ( 2 )   HTML ( 0 )   PDF (3338KB) ( 1 )  

A UAV multispectral image fusion algorithm based on gradient consistency constraint was proposed to address the issues of multiple types of ground objects, difficulty in extracting features from UAV spectral images, and poor fusion performance caused by uncertain spectral information. By using the beam adjustment algorithm to ensure the consistency between the coordinates of ground points and the projected coordinates of images, and using wavelet transform to extract image features, combined with gradient consistency algorithm to construct the target fusion function, high-quality image fusion is achieved through iterative operation. Experimental results have shown that the fused image of this method has clear details, high resolution, and good information integrity, providing strong technical support for the subsequent application of unmanned aerial vehicle multispectral images, especially in the fields of environmental monitoring, agricultural evaluation, etc, it has shown broad application prospects.

Figures and Tables | References | Related Articles | Metrics
Semisupervised monocular depth estimation framework based on data augmentation
Hong-wei ZHAO,Wei-min ZHOU
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2082-2088.  DOI: 10.13229/j.cnki.jdxbgxb.20230964
Abstract ( 2 )   HTML ( 0 )   PDF (602KB) ( 1 )  

To address the problem of requiring a large amount of labeled data for supervised learning in monocular depth estimation, a semi-supervised depth estimation framework AugDepth was proposed based on a teacher-student model. It operates by perturbing the data and training the model to learn depth consistency before and after the perturbation. Firstly, the smooth random intensity enhancement method was used to sample the intensity from the continuous domain. Multiple operations were randomly selected to increase the randomness of the data, and the output was enhanced by mixing the strength and weakness to prevent excessive disturbance. Then, considering the varying training difficulties of different unlabeled samples, while improving the model's inference of global information through Cutout, the Cutout strategy is adaptively adjusted based on the confidence level of unlabeled samples to enhance the model's generalization and learning abilities. The experimental results on the KITTI and NYU Deeph datasets show that AugDepth can significantly improve the accuracy of semi supervised depth estimation and exhibit good robustness in situations where labeled data is scarce.Key words:computer application; semi-supervised learning; data agumentation; monocular image; depth estimation

Figures and Tables | References | Related Articles | Metrics
Feature representation algorithm for imbalanced classification of multi⁃omics cancer data
Feng-feng ZHOU,Zhe GUO,Yu-si FAN
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2089-2096.  DOI: 10.13229/j.cnki.jdxbgxb.20231196
Abstract ( 2 )   HTML ( 0 )   PDF (1235KB) ( 1 )  

Aiming at a series of problems such as complex data structure, difficult prediction, data imbalance, and patient privacy protection in cancer diseases, a feature representation algorithm ImFeatures was proposed to solve the problem of imbalanced cancer data and enrich the sample structure. By combining two types of cancer transcriptome and methylation data as real samples, negative samples obtained after feature selection by logistic regression and random forest were randomly divided and combined with equal numbers of positive samples. The feature representation model proposed was used to generate sample representations that learn key feature information, thereby improving the predictive ability of the model. The experimental results show that on 11 common cancer datasets after feature characterization, the accuracy (Acc) of the algorithm combining feature selection and feature representation proposed in this paper exceeds 80.00% in all cases, and five cancer types even receive accuracies over 95.00%, which can effectively improve the prediction accuracies of cancer diseases.

Figures and Tables | References | Related Articles | Metrics
Multi perspective facial expression recognition algorithm based on spatiotemporal attention
Rui-shan DU,Zi-shan WANG
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2097-2102.  DOI: 10.13229/j.cnki.jdxbgxb.20240582
Abstract ( 1 )   HTML ( 0 )   PDF (2150KB) ( 1 )  

Firstly, skin color segmentation technology was used to locate facial regions in student images, and the located facial regions were input into the spatiotemporal attention module to obtain key information from multiple perspectives of the face. Secondly, the parameters in the convolutional neural network were optimized using an adaptive gradient descent algorithm with weighted decay, and key facial information was input into the optimized network to determine the types of facial expressions of students and complete multi view facial expression recognition. The experimental results show that the proposed algorithm can accurately extract key information of the face, and the accuracy of facial expression recognition is 100%. Therefore, the proposed algorithm can effectively recognize faces and improve the accuracy of facial expression recognition.

Figures and Tables | References | Related Articles | Metrics
Multi⁃depth adaptive fusion dehazing generation network
Bin WEN,Shun PENG,Chao YANG,Yan-jun SHEN,Hui LI
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2103-2113.  DOI: 10.13229/j.cnki.jdxbgxb.20230813
Abstract ( 2 )   HTML ( 0 )   PDF (2563KB) ( 1 )  

To address issues such as incomplete dehazing and distortion in image dehazing, a multi-depth adaptive fusion dehazing generation network is proposed using generative adversarial training. Firitly, the network utilizes U-Net++ architecture and haze perception units to learn the haze features. Secondly, a vector mixed attention module is proposed to expand the bottom-layer information and supplement details in the dehazed image. Thirdly, adaptive weights are constructed from partial and global dimensions to select features at different depths and improve the utilization of effective information in the network. Finally, a mixed loss is employed to ensure the quality of the generated image, and Wasserstein distance is introduced into the adversarial loss. To validate the effectiveness of the proposed algorithm, objective quantitative comparisons are conducted with 10 popular dehazing algorithms on the RESIDE and Haze4k datasets, followed by subjective evaluations on real images. The experimental results demonstrate that the proposed algorithm achieves a PSNR of 36.20 dB and SSIM of 0.988 4 on the SOTS Outdoor validation set, showing superior dehazing performance.

Figures and Tables | References | Related Articles | Metrics
Multi⁃scale context⁃aware and boundary⁃guided image manipulation detection method
Hai-peng CHEN,Shi-bo ZHANG,Ying-da LYU
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2114-2121.  DOI: 10.13229/j.cnki.jdxbgxb.20231027
Abstract ( 2 )   HTML ( 0 )   PDF (1486KB) ( 1 )  

Aiming at the problems of traditional image manipulation detection methods, such as fuzzy boundaries, single scale of extracted features, and ignoring background information, this paper proposes an image manipulation detection method based on multi-scale context-aware and boundary-guided. First, spatial details and base features of manipulated images are extracted using an improved pyramid vision transformer. Second, information related to the edge of the falsified region is explored by an edge context-aware module to generate an edge prediction map. Again, the edge guidance module is utilized to highlight the key channels in the extracted features and reduce the interference of redundant channels. Then, the rich contextual information of the manipulated region is learned from multiple sensory fields through the multi-scale context-aware module. Finally, the feature fusion module is utilized to accurately segment the manipulated region by focusing alternately on the foreground and background of the manipulated images. Comparing this paper's method quantitatively and qualitatively on five commonly used public image manipulation detection datasets, the experimental results show that this paper's method can effectively detect manipulated regions and outperforms other methods.

Figures and Tables | References | Related Articles | Metrics
Unbalanced image classification algorithm based on fine⁃grained analysis
Ping-ping LIU,Wen-li SHANG,Xiao-yu XIE,Xiao-kang YANG
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2122-2130.  DOI: 10.13229/j.cnki.jdxbgxb.20230991
Abstract ( 1 )   HTML ( 0 )   PDF (1235KB) ( 1 )  

Aiming at the complexity and diversity of fine-grained images, where traditional image classification methods exhibit limitations in focusing on fine-grained attributes and perform poorly when handling imbalanced datasets, a threshold-based fine-grained image classification algorithm utilizing deep metric learning was proposed. The focus on fine-grained attributes of images was enhanced by introducing a metric learning approach. Additionally, the classification accuracy was enhanced and the model convergence was expedited by incorporating pairwise loss and agent loss mechanisms. To address the issue of data imbalance, a classifier was devised grounded in threshold analysis techniques. This innovative classifier harnesses threshold analysis to facilitate multi-level classification of fine-grained images, thereby ameliorating the issue of low classification accuracy for certain categories within an imbalanced dataset. The results of these experiments unequivocally demonstrate that the proposed threshold classification algorithm for fine-grained images, based on deep metric learning, outperforms alternative methods in terms of classification accuracy.

Figures and Tables | References | Related Articles | Metrics
Semantic segmentation algorithm for multi temporal high⁃resolution satellite remote sensing images
Ying YU,Chun-ping WANG,Ren-ke KOU,Bo-xiong YANG,Lei WANG,Fu-jun ZHAO,Qiang FU
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2131-2137.  DOI: 10.13229/j.cnki.jdxbgxb.20240503
Abstract ( 2 )   HTML ( 0 )   PDF (1367KB) ( 1 )  

In order to solve the problem of inaccurate recognition of various semantic objects in remote sensing images using a single temporal low resolution image, a multi temporal high-resolution satellite remote sensing image semantic segmentation algorithm was proposed. By solving the multi temporal resolution of image information, remote sensing targets are partitioned, and various semantic objects and features of remote sensing images are accurately identified and extracted. Based on the definition of scale function, the segmentation weight is calculated to realize accurate recognition and segmentation of semantic objects in remote sensing images. The experimental results show that the recognition accuracy of semantic objects is significantly improved by the proposed method, and the maximum numerical difference between the experimental value and the real value of each semantic object content in the segmented ground object information does not exceed 0.2%, which provides strong support for the application of remote sensing images.

Figures and Tables | References | Related Articles | Metrics
Dynamic allocation algorithm of spectrum resources in high⁃dimensional wireless multi⁃user communication networks
Yun GAO,Jian-hui ZHOU,Yan-ping GUO
Journal of Jilin University(Engineering and Technology Edition). 2025, 55 (6):  2138-2144.  DOI: 10.13229/j.cnki.jdxbgxb.20240563
Abstract ( 1 )   HTML ( 0 )   PDF (1401KB) ( 1 )  

In high-dimensional wireless multi-user communication networks, there is a problem of unreasonable spectrum resource allocation, which leads to uneven network load and unstable communication quality. Traditional spectrum resource allocation methods can not respond to dynamic changes in user needs in real time, resulting in low utilization efficiency of spectrum resources. Therefore, a dynamic allocation algorithm for spectrum resources in high-dimensional wireless multi-user communication networks was proposed. By determining the dynamic allocation program of spectrum resources, while considering the goals of communication network load balancing and communication quality, a dynamic allocation objective function of spectrum resources was constructed, and its constraint conditions were explained. An improved genetic algorithm was introduced to solve the objective function and the optimal solution for dynamic allocation of spectrum resources was obtained. The experimental results show that the maximum spectrum resource utilization rate obtained by the proposed algorithm reaches 95%, and the spectrum resource allocation results are consistent with the actual results, fully confirming that the proposed algorithm has better spectrum resource allocation performance.

Figures and Tables | References | Related Articles | Metrics