上海工程技术大学学报

2025, 39(2): 1-2.

Abstract:

[Abstract](45) [FullText HTML] (23) [PDF 379KB](6)

A multi-objective optimization method for active control of loudness and sharpness of wiper windshield systems

SUN Yijia, WANG Yansong, LIU Ningning, YUAN Tao, XIE Xiaolong, GUO Hui

2025, 39(2): 119-124, 222. doi: 10.12299/jsues.24-0080

Abstract:
The reduction of noise in the power system of electric vehicles has accentuated the contradictory issues between loudness and sharpness caused by wiper reversal. Multi-objective optimization methods can help guide multi-band collaborative active control and improve the sound quality of the wiper windshield system. Noise data during wiper operation were collected through actual vehicle test. A multi-objective optimization formulation was constructed based on the direct mapping relationships between the amplitude of characteristic frequency bands and loudness and sharpness. Sensitivity analysis identified a set of sensitive frequency bands that most significantly affect changes in loudness and sharpness. The multi-objective genetic algorithm NSGA-Ⅱ was employed to obtain the amplitude values of these sensitive bands yielding the best compromise between loudness and sharpness. Active control of loudness and sharpness was simulated using an active noise equalization (ANE) algorithm based on a filtered-error least mean square algorithm (FELMS) adaptive filter, verifying the feasibility of the optimal solution. The results show that the loudness is reduced by 15.4%, sharpness by 18%, achieving 94% of the optimal loudness reduction and 87% of the optimal sharpness reduction, demonstrating improved sound quality.

[Abstract](495) [FullText HTML] (29) [PDF 2725KB](13)

Research on vibration control of building over rail transit depot by inerter system

HAN Dongyu, WANG Anbin

2025, 39(2): 125-130. doi: 10.12299/jsues.24-0048

Abstract:
Focusing on inertance systems and double-deck rail transit vehicle depot superstructure buildings, a finite element model was established and validated, the effectiveness of different types and inertance-to-mass ratios of inertance systems in controlling the vibration of these superstructure buildings was investigated. The results indicate that the three types of inertance systems viscous mass damper (VMD), tuned inerter damper (TID) and tuned viscous mass damper (TVMD) can all effectively reduce the Z-weighted vibration level on each floor of the upper building, and the damping effect becomes more pronounced as the inertance-to-mass ratio increases. Inertance systems TID and TVMD with different inertance-to-mass ratios exhibit larger reductions in the Z-weighted vibration level, reaching 1 dB~3 dB; whereas VMD shows relatively weaker damping performance, with a reduction in the Z-weighted vibration level of approximately 1 dB.

[Abstract](40) [FullText HTML] (19) [PDF 1433KB](4)

Energy management strategy for piezoelectric passive sensing nodes in rail vehicle vibration

SUI Huaishan, JIN Mengke, PENG Lele, ZHENG Shubin, ZHU Ting

2025, 39(2): 131-140, 173. doi: 10.12299/jsues.24-0053

Abstract:
In response to the system operational anomalies caused by the mismatch between the vibration energy source and power supply requirements in piezoelectric passive sensing nodes for rail vehicle vibration, a node-level energy management strategy was proposed. By analyzing the structural composition, relevant parameters, and operational modes, an energy management strategy model was constructed. The particle swarm optimization (PSO) algorithm was then utilized to optimize and match the design parameters and operational modes, resulting in the design of a system solution and the development of a prototype. The results demonstrate that the node achieves a minimum standby power consumption of 8 mW, accurately collects traction motor vibration data, and extracts bearing fault characteristic frequencies from the data, validating the effectiveness of the proposed energy management strategy.

[Abstract](42) [FullText HTML] (25) [PDF 4888KB](8)

Fault protection measures for low-power brushless DC motor controllers in vehicles

WANG Lei, GONG Yuanming

2025, 39(2): 141-147. doi: 10.12299/jsues.23-0244

Abstract:
Low-power brushless DC motor (BLDC) in vehicles is prone to failures under complex operating conditions and harsh environments, while traditional detection measures relying on a single threshold setting often lead to issues such as false alarms, missed detections, and over-protection. A fault protection measure for vehicle motors was proposed, which was monitored by setting multi-threshold and sensitive value at the software level, and optimized circuit design at the hardware level. By monitoring the key voltage and current parameters of the test vehicle motor, combined with real-time data, experience and model simulation, multiple thresholds and sensitive values were set to judge the running state of the vehicle motor. According to the triggering conditions of different thresholds, corresponding protection measures were taken, and fault protection test bench was set up to analyze the signals received by the upper computer. The results show that the measure is feasible and correct.

[Abstract](58) [FullText HTML] (26) [PDF 1573KB](4)

Combining improved YOLOX and improved second for road vehicle fusion detection

LIU Kai, LUO Suyun, WEI Dan

2025, 39(2): 148-156, 180. doi: 10.12299/jsues.24-0068

Abstract:
Common detection algorithms often struggled with missed detections, false positives, and large deviations in predicted orientation angles in road vehicle detection. A fusion detection algorithm combining improved YOLOX and Second was designed. By leveraging images and point clouds, two sub-networks were employed for vehicle detection. For image detection, convolutional block attention module, focal loss, and efficient intersection over union loss function were used to improve the detection performance of existing YOLOX. For point cloud detection, a residual sparse convolutional middle layer was designed to enhance the feature expression and context information association of Second algorithm, effectively reducing the missed detection rate of vehicles. The predictive directional angles was optimized by constructing a multi-bins strategy. Experimental conducted on KITTI dataset show that the algorithm surpassed the original, with improvements of 1.00%, 1.38%, and 2.66% in 3D average precision for easy, moderate, and hard targets, respectively. The accuracy of detecting target rotation angles is also greatly improved.

[Abstract](42) [FullText HTML] (33) [PDF 2880KB](6)

Microstructure evolution during hot deformation and dynamic transformation of Ti-5Al-3.3Sn-3.7Zr-0.6Ta-0.5W alloy in α + β two-phase region

LIU Yufei, ZHANG Ying, CHEN Zhiying, WANG Xiang

2025, 39(2): 157-165. doi: 10.12299/jsues.24-0097

Abstract:
The rheological behavior and microstructure evolution of the new near-α titanium alloy under different deformation conditions were investigated by thermal compression tests and microstructure characterization experiments. The results show that the real stress-strain curve displays characteristics of stress decrease and tendency to stabilize after the peak. At high strain rates, the peak stress is high and decreases rapidly. The transformation of the hard-oriented α phase to the soft-oriented β phase leads to a rapid decrease in the flow stress, with dynamic recrystallization (DRX) refining grains and dynamic recovery (DRV) rearranging dislocations within subgrains. Based on these findings, a dynamic recrystallization volume fraction model was established.

[Abstract](42) [FullText HTML] (21) [PDF 7322KB](2)

Experimental study on shear strength of geopolymer stabilized soft soil

XU Chubo, HU Guo

2025, 39(2): 166-173. doi: 10.12299/jsues.24-0043

Abstract:
Using solid sodium hydroxide as the alkaline exciter and slag and fly ash as the precursors, the fly ash-slag based geopolymer was prepared by "one-step" method. Through the unconsolidated and undrained (UU) triaxial shear test, the mechanical properties of soft soil stabilized with fly ash-slag based geopolymer were tested, and the influence mechanism of such as fly ash content, water to biner ratio, and curing time on the shear strength of the stabilized soft soil were investigated. The cohesion and internal friction angle of the stabilized soft soil were calculated, and the mechanisms of geopolymer stabilizing soft soil were also analyzed through microscopic tests. The results show that the increase of fly ash doping leads to the decrease of shear strength and internal friction angle of geopolymer stabilized soft soil; the increase of water to binder ratio reduces the shear strength of stabilized soft soil; the increase of cofining pressure makes the strain softening gradually weakened, and the strength gradually improved; the increase of curing time increases the shear strength, and the cohesion and the internal friction angle show a rising trend at the same time.

[Abstract](48) [FullText HTML] (33) [PDF 3027KB](3)

Optimization of preparation process of clove volatile oil inclusion complex by Box-Behnken response surface methodology

MA Ying, LIU Yurou, LIU Xiaohui

2025, 39(2): 174-180. doi: 10.12299/jsues.24-0085

Abstract:
To prepare the hydroxypropyl-β-cyclodextrin inclusion complex from clove essential oil using saturated aqueous solution method. Based on the single factor experiments, the ratio of hydroxypropyl-β-cyclodextrin to volatile oil, encapsulation temperature, and encapsulation time were used as influencing factors, and the comprehensive score was used as the response value. The encapsulation process was optimized by Box-Behnken response surface methodology. The optimal condition was obtained as the ratio of hydroxypropyl-β-cyclodextrin to volatile oil was 14.5∶1, with an inclusion temperature of 59 ℃ and an inclusion time of 2.4 h. The comprehensive score of the three validation experiments is 92.21%±0.50%, which is close to the predicted value, indicating that the prediction of the model is accurate and reliable. The inclusion complex was observed under microscope, analyzed by thin layer chromatography (TLC), and identified by infrared spectroscopy, proving the formation of a stable inclusion complex. These results can provide data support for further development and utilization of clove volatile oil.

[Abstract](40) [FullText HTML] (37) [PDF 1596KB](3)

Research on extrusion forming process of 25CrMo₄ automotive spline shaft part based on Deform

JI Wenjun, GONG Hongying, LAN Yi, MA Ji, LIAO Zehuan

2025, 39(2): 181-186. doi: 10.12299/jsues.24-0272

Abstract:
A systematic study on the extrusion process of automotive shaft parts with complex external splines was carried out. Firstly, the characterisation of the parts were analyzed. Three process parameters of extrusion speed, friction coefficient and billet taper angle were selected using Deform software, and orthogonal simulation tests were performed. The optimal process parameters determined using the range analysis method were as follows: punch extrusion speed of 20 mm/s, friction coefficient of 0.12, and billet taper angle of 15°. The results indicate that the maximum extrusion load and the minimum damage value obtained under these optimized conditions. Compared with the pre-optimisation process, the extrusion load was reduced by 26.3%, and the damage value was reduced by 81.4%.

[Abstract](44) [FullText HTML] (19) [PDF 1997KB](3)

Design and control of a complementary dual voice coil motor push-pull two-dimensional translational motion parallel precision motion platform

MU Rui, LAI Leijie

2025, 39(2): 187-193. doi: 10.12299/jsues.24-0081

Abstract:
A voice coil motor-driven two-degree-of-freedom translational large-stroke decoupled micro-motion stage was proposed, which adopts a parallel dual-push-pull structure coupling straight flexible reed beams with a compound double parallelogram mechanism. Based on the push-pull driving principle of the voice coil motor and the configuration design of the dual-degree-of-freedom flexible mechanism, the static and dynamic characteristics of the platform were modeled using the force and deformation characteristics of the spring beam and the Lagrange equation, and an analytical model for calculating the static stiffness and natural frequency of the platform was obtained. Finite element analysis was conducted to validate the rationality and accuracy of the proposed models, and a control experimental system for the platform was constructed to compare trajectory tracking performance between single-push-pull and dual-push-pull modes. The experimental results show that the dual-push-pull mode has better tracking performance, higher precision, and better stability compared to the single-push-pull mode. The proposed complementary voice coil motor-driven large stroke flexible parallel precision motion platform features large stroke, high precision, and high stability characteristics, making it particularly suitable for large-stroke high-speed precision motion scenarios.

[Abstract](45) [FullText HTML] (21) [PDF 2885KB](3)

Study on influence of anodic channel groove structure on performance of PEM electrolyzer

FAN Lei, XIA Peng, DIAO Yongfa

2025, 39(2): 194-200, 208. doi: 10.12299/jsues.24-0049

Abstract:
A steady-state two-phase heat and mass transfer model was developed for a single proton exchange membrane (PEM) electrolyzer to investigate the effects of different groove structures (triangular, tooth-shaped and goose-shaped) on its performance. The influence of groove depth, number and spacing was examined. The results show that incorporating grooves in the anode flow channel can improve electrochemical performance, enhance the heat and mass transfer, and increase the efficiency of the electrolyzer, with the best optimization effect of the goose-shaped groove. Within the range of 0.4~1.6 mm in depth, the electrolyzer corresponding to 0.8 mm has the best performance. Compared to a conventional flow channel, the optimized electrolyzer has a lower potential to achieve the same current density, has a better proportion of water saturation and oxygen content in the anode region, and has a lower proton exchange membrane temperature, thereby improving the performance of the electrolyzer.

[Abstract](56) [FullText HTML] (38) [PDF 1660KB](2)

Response analysis of tethered UAV in dynamic wind environment

JIANG Hongbo, LIU Chaofeng, TANG Zhiren, HOU Feiyu

2025, 39(2): 201-208. doi: 10.12299/jsues.24-0072

Abstract:
In order to investigate the aerodynamic performance of the tethered unmanned aerial vehicle (UAV) and the deformation of the tether, the dynamic response analysis of the tethered UAV was conducted. A tethered UAV model was developed, and numerical simulations were performed using using a fluid-structure interaction (FSI) analysis method, standard k-ε turbulence model, and slip-grid technology. The dynamic performance of the tethered UAV system under crosswinds at different wind speeds was investigated. The results shows that the vortex strength increases significantly with crosswind speed, particularly on the windward side. The tether deformed under the combined effect of wind and its gravity, and the maximum deformation occurs at a specific location and varies with wind speed. The maximum deformation was 448 mm, and the overall average deformation was 145 mm in 10 s. It can be used as a reference for the design and utilization conditions of tethered UAVs.

[Abstract](44) [FullText HTML] (23) [PDF 2440KB](3)

Simulation study of cutting Inconel 718 alloy with micro textured tools

ZHU Dawei, LIU Sheng, ZHANG Yekai

2025, 39(2): 209-215. doi: 10.12299/jsues.24-0070

Abstract:
Inconel 718 alloy is considered a typical difficult-to-machine material. The cutting simulation and experimental research of Inconel 718 alloy were conducted through the design of three different shapes of micro-weave on the front face of the tool: square, round, and triangular. The influence of the shape of the micro-weave on the machining performance of Inconel 718 was explored, and the material removal process, stress distribution, chip morphology, and cutting force during the cutting process were analyzed. The change rule of these aspects was passively observed and documented. The results show that compared with ordinary tools, microtextured tools can effectively reduce the cutting force, reduce the tool-chip contact length and improve the friction of the tool. In addition, the phenomenon of secondary cutting in the cutting process was discussed, the mechanism of secondary cutting was analyzed, a method to inhibit secondary cutting through force analysis was propose, and the reliability of the method was verified by using experiments and finite element simulation.

[Abstract](407) [FullText HTML] (37) [PDF 1814KB](9)

Adaptive path planning of surgical robots based on environmental attributes

FANG Cheng, HU Zhi

2025, 39(2): 216-222. doi: 10.12299/jsues.24-0063

Abstract:
In a dual-arm surgical robot environment, it is essential for the robotic arms to reach target positions safely and efficiently. An adaptive APF-Informed-RRT* path planning algorithm was proposed to overcome limitation of traditional APF-Informed-RRT* algorithms, such as lack of adaptability and insufficient path stability, thereby improving the quality of path planning for robotic arms. The coefficients of the attraction term, repulsion term, and random guidance term were dynamically adjusted based on the Euclidean distance from the robot's current position to the target, as well as the density of obstacles in the nearby area. This adaptability helps to reduce the path planning cost by adjusting to the environment. Simulation results show that compared to the APF-Informed-RRT* algorithm, the proposed method performs better in terms of path cost and search range in a 3D oral simulation environment. This adaptability endows the algorithm with effective stability in path planning tasks.

[Abstract](36) [FullText HTML] (32) [PDF 2341KB](2)

Design of a differential negative resistor VCO based on GaAs_BJT

DU Xinwei, XIAO Manlin, XIAO Shuai, ZHANG Wenyu

2025, 39(2): 223-228. doi: 10.12299/jsues.24-0079

Abstract:
A low noise LC-VCO with a working frequency of 3.8~4.0 GHz was designed using the 0.25 μm GaAs HBT process of San'an Optoelectronics. The design adopted a bipolar junction transistor to form a differential negative resistance structure, and used a source resistance bias circuit to provide the working current. Finally, the tail inductance and large-capacity filtering technology were used to improve the noise factor, and the microstrip line was used instead of inductance in the layout design to achieve a lower noise factor. The post-chip simulation results show that the LC-VCO designed has a frequency range of 3.75~4.05 GHz, an amplitude of 1.6 V, a phase noise of −118.8 dBc at 1 MHz offset, and a voltage control sensitivity of 43 MHz/V.

[Abstract](52) [FullText HTML] (40) [PDF 1166KB](3)

Influence of annular portion of wrapped beam on propagation of central filament

XU Hu, XU Guang, WANG Zhaoqiang, CHEN Xi, LU Chenhui

2025, 39(2): 229-235. doi: 10.12299/jsues.24-0075

Abstract:
Light filaments are pivotal in ultrafast laser applications, and enhancing the long-distance propagation of light filaments substantially broadens their application prospects.Wrapped beams exhibits characteristics that significantly enhance the long-distance propagation of light filaments, with their annular portions playing a crucial role during propagation. Based on the split-step Fourier transform simulation method, the propagation characteristics of wrapped beams endowed with varied annular segments were systematically investigated through comparative simulation. The results indicate that the input energy, beam waist radius and central distance of the annular portion of the wrapped beam all influence the filament's propagation. Fine-tuning the annular parameters can optimize the propagation efficiency of laser filamentation, expanding the applications domains of ultrafast lasers.

[Abstract](34) [FullText HTML] (21) [PDF 2080KB](4)

Double-layer path planning based on evolutionary ant colony algorithm and conflict prediction resolution

YU Kaiying, XU Bin

2025, 39(2): 236-242. doi: 10.12299/jsues.24-0087

Abstract:
Aiming at the problems of low search efficiency and difficulty in resolving path conflicts when the ant colony algorithm was used to the cooperative path planning problem of multiple mobile robots, a dual-layer cooperative path planning method was proposed. In the static layer, the crossover mechanism of genetic algorithm was integrated into ant colony algorithm, and the evolutionary path crossover strategy was established to improve the quality of planning path.Concurrently, the calculation method of pheromone increment differentiation was adopted to accelerate the convergence speed of ant colony algorithm. In the dynamic layer, collision prediction was conducted based on a three-dimensional space-time map, and a priority conflict-free strategy was employed to effectively resolve path conflict. Simulation results show that the proposed dual-layer dynamic cooperative path planning method can improve the comprehensive performance of the algorithm and solve the cooperative path planning problem, which verifying the feasibility and effectiveness of the proposed algorithm.

[Abstract](41) [FullText HTML] (30) [PDF 2016KB](2)

An accuracy improvement method for automatic fiber placement defect recognition based on attention data enhancement

CAO Jieqiang, ZHANG Liqiang, LI Junli, MAO Jian, LIU Gang

2025, 39(2): 243-250. doi: 10.12299/jsues.24-0074

Abstract:
The scarcity of infrared defect samples and the difficulty in acquiring them during the automated tape laying process result in low classification accuracy and ineffective defect identification by classification model algorithm. To address this issue, a data augmentation method based on an attention mechanism Wasserstein generative adversarial network (WGAN) was proposed. A CB-Attention module was introduced into the generator to enhance its capability to capture image feature, expand the receptive field, and improve the quality of generated images without adding extra parameters. Batch channel normalization was employed to adaptively integrate information from both channel and batch dimensions, thereby imcreasing the model's training speed and generalization ability. Experimental results demonstrate that the samples generated by the attention-based data augmentation method are diverse and of high quality. Incorporating these samples into the small dataset of automated tape laying defects significantly improves defect recongnition accuracy, which validates the ffectiveness of the proposed algorithm and lays a data foundation for automated tape laying defect classification algorithms.

[Abstract](42) [FullText HTML] (30) [PDF 1289KB](1)

Effect of far-infrared basketball arm guard on muscle fatigue under different pressures

LI Zhenzhen, LI Yanmei

2025, 39(2): 251-256. doi: 10.12299/jsues.24-0073

Abstract:
Existing researches on sports arm sleeves predominantly focus on the singular effect of compression in alleviating muscle fatigue, with limited integration of far-infrared radiation (FIR) technology investigated. In this study, sports arm sleeves were knitted using graphene-nylon composite fiber and nylon yarn, with far-infrared fiber contents (mass fraction, the same hereafter) of 0%, 25%, 50%, and 75%. Their effects on reducing muscle fatigue in the arm were evaluated under three pressure levels. Muscle fatigue was assessed using the root mean square (RMS) and median frequency (MF) of electromyographic signals, as well as subjective fatigue scores. The results indicate that FIR-integrated sleeves effectively reduce muscle fatigue after exercise, with the effect increasing as FIR fiber content rises, peaking at 75% content. At the same FIR contents, medium- and high-compression sleeves are more effective in alleviating muscle fatigue, particularly in the brachioradialis and flexor carpi radialis muscles. This may be attributed to the sequence of muscle activation and involvement degree during exercise.

[Abstract](60) [FullText HTML] (37) [PDF 1015KB](2)

Current Issue 2025 Vol. 39, No. 2

Current Issue
2025 Vol. 39, No. 2