Articles in press have been peer-reviewed and accepted, which are not yet assigned to volumes /issues, but are citable by Digital Object Identifier (DOI).
Display Method:
2025, 39(3): 257-265.
doi: 10.12299/jsues.24-0050
Abstract:
Aiming at the issues of model parameter size, detection accuracy, and efficiency in rail fastener detection algorithms, a lightweight network model named YOLO-FBS based on YOLOv5n was proposed. The model embeds the Faster Block module to achieve a reduction in network parameters, introduces the SimAM parameter-free feature selection module and the CAFARE lightweight upsampling module to improve detection accuracy, and finally employs LAMP score-based pruning to remove channels with lower weights. The resulting YOLO-FBS model has only 0.28×106 parameters, achieves a detection accuracy of 89.2%, and reaches a detection speed of 190.2 frames/s. This network model offers the advantages of low parameter count and high detection speed while maintaining high detection accuracy.
Aiming at the issues of model parameter size, detection accuracy, and efficiency in rail fastener detection algorithms, a lightweight network model named YOLO-FBS based on YOLOv5n was proposed. The model embeds the Faster Block module to achieve a reduction in network parameters, introduces the SimAM parameter-free feature selection module and the CAFARE lightweight upsampling module to improve detection accuracy, and finally employs LAMP score-based pruning to remove channels with lower weights. The resulting YOLO-FBS model has only 0.28×106 parameters, achieves a detection accuracy of 89.2%, and reaches a detection speed of 190.2 frames/s. This network model offers the advantages of low parameter count and high detection speed while maintaining high detection accuracy.
2025, 39(3): 266-273.
doi: 10.12299/jsues.24-0119
Abstract:
Traditional vehicle lighting control system are characterized by low development efficiency, long iteration cycles, high costs of real-vehicle simulation, and an inability to meet the intelligent demands of modern automobiles. To address these issues, a calculation formula for the deflection of the headlamp's illumination angle was proposed based on the adaptive control logic of the headlamps, and a model of the adaptive front-lighting system (AFS) was constructed. Then, the AFS model was rapidly transformed intoan executable real-time system by utilizing the dSPACE-based rapid control prototyping (RCP) technology. Various complex driving scenarios were simulated to verify the accuracy and validity of the AFS control logic and the deflection of the illumination angle. The application results indicate that this development method shortens the system development cycle and improve development efficiency, and that the proposed AFS significantly enhances driving safety and comfort.
Traditional vehicle lighting control system are characterized by low development efficiency, long iteration cycles, high costs of real-vehicle simulation, and an inability to meet the intelligent demands of modern automobiles. To address these issues, a calculation formula for the deflection of the headlamp's illumination angle was proposed based on the adaptive control logic of the headlamps, and a model of the adaptive front-lighting system (AFS) was constructed. Then, the AFS model was rapidly transformed intoan executable real-time system by utilizing the dSPACE-based rapid control prototyping (RCP) technology. Various complex driving scenarios were simulated to verify the accuracy and validity of the AFS control logic and the deflection of the illumination angle. The application results indicate that this development method shortens the system development cycle and improve development efficiency, and that the proposed AFS significantly enhances driving safety and comfort.
2025, 39(3): 274-279.
doi: 10.12299/jsues.24-0116
Abstract:
To address the problem of under-maintenance and over-maintenance in the preventive maintenance of electric locomotive pantograph airbag cylinder, a method for optimizing inspection and maintenance parameters using failure data of the airbag cylinders was proposed. Firstly, based on the traditional preventive maintenance strategy, the graph parameter method was used to analyze whether the failure data of the airbag cylinders followed an exponential distribution. Subsequently, the Bartlett test was employed to verify the validity of this distribution assumption. Secondly, based on the characteristic constant failure rate of the airbag cylinders, a reliability model was established. Key reliability indicators, such as the instantaneous availability and steady-state availability, then derived, providing a basis for formulating the optimal preventive maintenance strategy. Thirdly, a minimum cost model for the preventive maintenance and replacement of the airbag cylinder was established, and the optimal preventive maintenance interval and maintenance cost were obtained. The optimized inspection and maintenance parameters for the airbag cylinders can provide scientific basis for maintenance personnel to improve quality and reduce maintenance costs.
To address the problem of under-maintenance and over-maintenance in the preventive maintenance of electric locomotive pantograph airbag cylinder, a method for optimizing inspection and maintenance parameters using failure data of the airbag cylinders was proposed. Firstly, based on the traditional preventive maintenance strategy, the graph parameter method was used to analyze whether the failure data of the airbag cylinders followed an exponential distribution. Subsequently, the Bartlett test was employed to verify the validity of this distribution assumption. Secondly, based on the characteristic constant failure rate of the airbag cylinders, a reliability model was established. Key reliability indicators, such as the instantaneous availability and steady-state availability, then derived, providing a basis for formulating the optimal preventive maintenance strategy. Thirdly, a minimum cost model for the preventive maintenance and replacement of the airbag cylinder was established, and the optimal preventive maintenance interval and maintenance cost were obtained. The optimized inspection and maintenance parameters for the airbag cylinders can provide scientific basis for maintenance personnel to improve quality and reduce maintenance costs.
2025, 39(3): 280-285, 291.
doi: 10.12299/jsues.24-0046
Abstract:
To investigate the strategy choice between airlines and online travel agency (OTA) platforms within a competitive-cooperative context, an evolutionary game model was constructed. The replication dynamic equation and Jacobian matrix applied to explore the evolutionary trajectory of their strategies. Taking China Eastern Airlines and the OTA platform Ctrip as a case study, the impact of varying parameters on the strategies of both sides was examined through Matlab based evolutionary game simulation method. Research shows that the stable equilibrium result of evolutionary game is that China Eastern Airlines will adopt the strategy of "increasing direct sales and reducing agency sales", while the OTA platform will adopt the strategy of "retreating to the hotel and leasing industry".
To investigate the strategy choice between airlines and online travel agency (OTA) platforms within a competitive-cooperative context, an evolutionary game model was constructed. The replication dynamic equation and Jacobian matrix applied to explore the evolutionary trajectory of their strategies. Taking China Eastern Airlines and the OTA platform Ctrip as a case study, the impact of varying parameters on the strategies of both sides was examined through Matlab based evolutionary game simulation method. Research shows that the stable equilibrium result of evolutionary game is that China Eastern Airlines will adopt the strategy of "increasing direct sales and reducing agency sales", while the OTA platform will adopt the strategy of "retreating to the hotel and leasing industry".
2025, 39(3): 286-291.
doi: 10.12299/jsues.24-0114
Abstract:
The impact of shift system on the emergency response performance and alertness of subway drivers was investigated through simulation experiments. Emergency response performance was measured by the average response time and accuracy rate, while electroencephalogram (EEG) indicators included an alertness index calculated from SMR waves and the average EEG power ratio. Given the special nature of emergency response scenarios, partial gamma (γ) rhythms were incorporated into the calculation of both the alertness index and the average power ratio. The findings indicate that incorporating gamma-band rhythms can enhance the evaluation of changes in subway drivers' alertness levels. Drivers on the day shift exhibited higher alertness and better emergency response performance than those on the night shift. As task complexity increased, emergency response performance declined. Meanwhile alertness showed an trend, initially rising before subsequently falling.
The impact of shift system on the emergency response performance and alertness of subway drivers was investigated through simulation experiments. Emergency response performance was measured by the average response time and accuracy rate, while electroencephalogram (EEG) indicators included an alertness index calculated from SMR waves and the average EEG power ratio. Given the special nature of emergency response scenarios, partial gamma (γ) rhythms were incorporated into the calculation of both the alertness index and the average power ratio. The findings indicate that incorporating gamma-band rhythms can enhance the evaluation of changes in subway drivers' alertness levels. Drivers on the day shift exhibited higher alertness and better emergency response performance than those on the night shift. As task complexity increased, emergency response performance declined. Meanwhile alertness showed an trend, initially rising before subsequently falling.
2025, 39(3): 292-299.
doi: 10.12299/jsues.24-0158
Abstract:
The optimization of the collector/evaporator structure in a direct-expansion solar heat pump system was investigated. A distributed parameter numerical model for a roll-bond collector/evaporator was developed and validated against experimental data. The effects of the path tube number arrangement and tube spacing on the collector/evaporator collector efficiency and the coefficient of performance (COP) of the heat pump system were simulated and analyzed. The results indicate that, for a given total number of tubes, a larger tube number ratio between the first and second paths can lead to a lower total refrigerant pressure drop and a higher evaporation temperature. The optimal heat collection efficiency and COP are achieved when the tube number ratio is 1∶1. Furthermore, as the tube spacing increases, both the heat collection efficiency and COP first increase and then decrease, reaching their maximum values at a tube spacing of 0.04 m.
The optimization of the collector/evaporator structure in a direct-expansion solar heat pump system was investigated. A distributed parameter numerical model for a roll-bond collector/evaporator was developed and validated against experimental data. The effects of the path tube number arrangement and tube spacing on the collector/evaporator collector efficiency and the coefficient of performance (COP) of the heat pump system were simulated and analyzed. The results indicate that, for a given total number of tubes, a larger tube number ratio between the first and second paths can lead to a lower total refrigerant pressure drop and a higher evaporation temperature. The optimal heat collection efficiency and COP are achieved when the tube number ratio is 1∶1. Furthermore, as the tube spacing increases, both the heat collection efficiency and COP first increase and then decrease, reaching their maximum values at a tube spacing of 0.04 m.
2025, 39(3): 300-306.
doi: 10.12299/jsues.24-0129
Abstract:
A three-dimensional mathematical model of the auxiliary gas flow field in laser cutting of medium-thickness plates was developed based on the Gaussian heat source equation, heat transfer equation, Reynolds-averaged N-S equation and SST k-ω turbulence model. The validity of the simulation model was verified by experiments. Based on the established model, the distribution of pressure, temperature, velocity and shear stress in the auxiliary gas flow field were analyzed, and the effect of nozzle inlet pressure on the dynamic performance of the flow field was investigated. The results show that a reasonable nozzle inlet pressure can improve the dynamic characteristics of the auxiliary gas and cutting quality.
A three-dimensional mathematical model of the auxiliary gas flow field in laser cutting of medium-thickness plates was developed based on the Gaussian heat source equation, heat transfer equation, Reynolds-averaged N-S equation and SST k-ω turbulence model. The validity of the simulation model was verified by experiments. Based on the established model, the distribution of pressure, temperature, velocity and shear stress in the auxiliary gas flow field were analyzed, and the effect of nozzle inlet pressure on the dynamic performance of the flow field was investigated. The results show that a reasonable nozzle inlet pressure can improve the dynamic characteristics of the auxiliary gas and cutting quality.
2025, 39(3): 307-313.
doi: 10.12299/jsues.24-0093
Abstract:
Focusing on the end-effector of minimally invasive surgical robots, proposing a novel rotation mechanism and conducting numerical simulation and experimental research on its transmission backlash characteristics. A computational model for transmission backlash was established through theoretical derivation, with analysis of transfer deviation's impact on output torque under different influencing factors. Numerical results indicate that when output torque ranges from 0 to 0.3 N·m, the transfer deviation stays within 0° to 0.04°. Transmission backlash shows significant sensitivity to preload force and steel wire cross-sectional area, while being less affected by free wire segment length. A physical prototype and experimental platform were constructed to perform tracking experiments, measuring the end-effector's rotation angle threshold at ±74.29°. Data and error analyses of transmission backlash characteristics were conducted through simulations mimicking frequent surgical direction changes during operations.
Focusing on the end-effector of minimally invasive surgical robots, proposing a novel rotation mechanism and conducting numerical simulation and experimental research on its transmission backlash characteristics. A computational model for transmission backlash was established through theoretical derivation, with analysis of transfer deviation's impact on output torque under different influencing factors. Numerical results indicate that when output torque ranges from 0 to 0.3 N·m, the transfer deviation stays within 0° to 0.04°. Transmission backlash shows significant sensitivity to preload force and steel wire cross-sectional area, while being less affected by free wire segment length. A physical prototype and experimental platform were constructed to perform tracking experiments, measuring the end-effector's rotation angle threshold at ±74.29°. Data and error analyses of transmission backlash characteristics were conducted through simulations mimicking frequent surgical direction changes during operations.
2025, 39(3): 314-319.
doi: 10.12299/jsues.24-0165
Abstract:
To address the errors caused by the non-coincidence of the hole surface with the calibration plane in machine vision-based aperture measurement, as well as the edge degradation encountered during edge extraction, an aperture diameter compensation method based on perspective projection error was proposed. Edge detection was performed using an adaptive median filtering algorithm, and the aperture diameter was determined by employing quadratic curve invariant. Perspective projection error was then introduced to obtain the corrected hole diameter. A visual inspection experimental platform was constructed, and the test results demonstrate that the proposed compensation method achieves a measurement error within ± 0.002 mm, and a measurement accuracy of 0.001 mm. The method demonstrates a better compensation effect and can meet the enterprise's accuracy requirements.
To address the errors caused by the non-coincidence of the hole surface with the calibration plane in machine vision-based aperture measurement, as well as the edge degradation encountered during edge extraction, an aperture diameter compensation method based on perspective projection error was proposed. Edge detection was performed using an adaptive median filtering algorithm, and the aperture diameter was determined by employing quadratic curve invariant. Perspective projection error was then introduced to obtain the corrected hole diameter. A visual inspection experimental platform was constructed, and the test results demonstrate that the proposed compensation method achieves a measurement error within ± 0.002 mm, and a measurement accuracy of 0.001 mm. The method demonstrates a better compensation effect and can meet the enterprise's accuracy requirements.
2025, 39(3): 320-325.
doi: 10.12299/jsues.24-0133
Abstract:
Multimodal machine learning, which combines image and text data, is able to improve the accuracy of industrial equipment identification. To address the limitation that existing industrial equipment identification algorithms rely solely on image data and fail to leverage the widely available textual data in industrial settings, a multimodal industrial equipment identification model based on image and text was proposed. This model consists of a novel image-text dataset and a multimodal identification algorithm. The dataset images were collected from real scenarios, and the text was automatically annotated by a multimodal large model. The identification algorithm can adaptively adjust the weight of the modalities, integrate the information of the two modalities, and assist the model in decision-making. Experimental results show that the proposed algorithm can achieve an improvement of 17.62% in identification accuracy compared to single-modal image data, and outperforms other multimodal recognition algorithms by 5.3%, demonstrating its effectiveness.
Multimodal machine learning, which combines image and text data, is able to improve the accuracy of industrial equipment identification. To address the limitation that existing industrial equipment identification algorithms rely solely on image data and fail to leverage the widely available textual data in industrial settings, a multimodal industrial equipment identification model based on image and text was proposed. This model consists of a novel image-text dataset and a multimodal identification algorithm. The dataset images were collected from real scenarios, and the text was automatically annotated by a multimodal large model. The identification algorithm can adaptively adjust the weight of the modalities, integrate the information of the two modalities, and assist the model in decision-making. Experimental results show that the proposed algorithm can achieve an improvement of 17.62% in identification accuracy compared to single-modal image data, and outperforms other multimodal recognition algorithms by 5.3%, demonstrating its effectiveness.
2025, 39(3): 326-332.
doi: 10.12299/jsues.24-0199
Abstract:
Bardoxolone, also known as 2−cyano−3,12−dioxolanoleana−1,9(11)−dien−28−oic' acid (CDDO), is a pentacyclic triterpenoid synthesized from natural oleanolic acid through a series of chemical modifications. CDDO and its derivatives exhibit a broad pharmacological activities and have shown great potential in studies. In this review, recent advancements in the synthesis of CDDO and its derivatives were summarized. Their pharmacological effects and underlying mechanisms, including antitumor activity, anti-inflammatory properties, neuroprotection, kidney protection, antiviral activity, among others, were also dicussed. This review was conducted to provide valuable insights for the future development and application of these compounds.
Bardoxolone, also known as 2−cyano−3,12−dioxolanoleana−1,9(11)−dien−28−oic' acid (CDDO), is a pentacyclic triterpenoid synthesized from natural oleanolic acid through a series of chemical modifications. CDDO and its derivatives exhibit a broad pharmacological activities and have shown great potential in studies. In this review, recent advancements in the synthesis of CDDO and its derivatives were summarized. Their pharmacological effects and underlying mechanisms, including antitumor activity, anti-inflammatory properties, neuroprotection, kidney protection, antiviral activity, among others, were also dicussed. This review was conducted to provide valuable insights for the future development and application of these compounds.
2025, 39(3): 333-340.
doi: 10.12299/jsues.24-0127
Abstract:
Catalytic combustion is presently considered as one of the most effective methods to control soot particles from diesel exhaust, while its key issue is how to construct efficient catalysts. Herein, a series of MnOx/Co3O4 composites within nanosheets were prepared using the hydrothermal integrating with wet-impregnation approach, chiefly aiming to catalyze soot purification. Experimental results indicate that, 20MnCo nanosheet composites with Mn/Co molar ratio of 20% show the optimal soot purification performance with Tm of 354 ℃ and 100% CO2 selectivity, mainly attributing to the three aspects: 1) porous nanosheets with high surface area and rich pores not merely improved the contact area of composites and soot, but also benefited the decreased mass transfer resistance of gas reactants; 2) loading MnOx led to the increased ratio of Mn3+ and Co3+ on composite’s surface, which favored the activating the adsorbed oxygen species on composite’s surface to generate more reactive oxygen species; 3) loading MnOx resulted in the improved redox ability that facilitated NO oxidation to produce NO2. Meanwhile, 20MnCo nanosheet composites demonstrated the good recycling stability, showing a good application prospect.
Catalytic combustion is presently considered as one of the most effective methods to control soot particles from diesel exhaust, while its key issue is how to construct efficient catalysts. Herein, a series of MnOx/Co3O4 composites within nanosheets were prepared using the hydrothermal integrating with wet-impregnation approach, chiefly aiming to catalyze soot purification. Experimental results indicate that, 20MnCo nanosheet composites with Mn/Co molar ratio of 20% show the optimal soot purification performance with Tm of 354 ℃ and 100% CO2 selectivity, mainly attributing to the three aspects: 1) porous nanosheets with high surface area and rich pores not merely improved the contact area of composites and soot, but also benefited the decreased mass transfer resistance of gas reactants; 2) loading MnOx led to the increased ratio of Mn3+ and Co3+ on composite’s surface, which favored the activating the adsorbed oxygen species on composite’s surface to generate more reactive oxygen species; 3) loading MnOx resulted in the improved redox ability that facilitated NO oxidation to produce NO2. Meanwhile, 20MnCo nanosheet composites demonstrated the good recycling stability, showing a good application prospect.
2025, 39(3): 341-346.
doi: 10.12299/jsues.24-0098
Abstract:
To optimize the virtual fabric drape effect in VStitcher, drape experiments on 38 selected fabric samples were conducted. A comparative analysis was performed by measuring 8 drape indicators for both virtual and physical fabrics to quantify their discrepancies. Subsequently, the physical properties of the virtual fabrics were systematically adjusted to align their drape behavior with that of their real-world counterparts. Using the least squares, a fitting analysis was conducted on the physical properties before and after adjustment, leading to the establishment of an optimization model for virtual fabric physical properties. To validate the model, three additional fabric samples were randomly selected for testing. The results demonstrate that the relative error between the drape indicators of the optimized virtual fabrics and the physical fabrics is within 5%. This confirms the effectiveness and accuracy of the proposed model in enhancing the realism of virtual fabric drape simulation.
To optimize the virtual fabric drape effect in VStitcher, drape experiments on 38 selected fabric samples were conducted. A comparative analysis was performed by measuring 8 drape indicators for both virtual and physical fabrics to quantify their discrepancies. Subsequently, the physical properties of the virtual fabrics were systematically adjusted to align their drape behavior with that of their real-world counterparts. Using the least squares, a fitting analysis was conducted on the physical properties before and after adjustment, leading to the establishment of an optimization model for virtual fabric physical properties. To validate the model, three additional fabric samples were randomly selected for testing. The results demonstrate that the relative error between the drape indicators of the optimized virtual fabrics and the physical fabrics is within 5%. This confirms the effectiveness and accuracy of the proposed model in enhancing the realism of virtual fabric drape simulation.
2025, 39(3): 347-353, 374.
doi: 10.12299/jsues.24-0163
Abstract:
As an artificial composite structure, an acoustic metamaterial plate (AMP) can block the propagation of the elastic wave within specific frequency ranges. Due to the combined effects of multiple mechanisms in the composite structure, it is difficult to derive and express the physical properties using theoretical formulas. An AMP structure based on local resonance mechanism was proposed, and the bandgap characteristics were analyzed using finite element method, to obtain different cell structure parameters and corresponding bandgap characteristics. A sample set of the relationship between the structure parameters and corresponding bandgap characteristics was established for the AMP. A forward neural network model was designed to obtain the mapping relationship between structural parameters and bandgap range, and the forward prediction of the corresponding bandgap range by inputting structural parameters can be achieved. On this basis, a reverse neural network was proposed, and the reverse design can be carried out to obtain the structural parameters of AMP aiming the required bandgap range. The simulation results show that complex and cumbersome theoretical derivation and calculation can be avoided with the forward-reverse design method for AMP, which is helpful to promote further development of acoustic metamaterials.
As an artificial composite structure, an acoustic metamaterial plate (AMP) can block the propagation of the elastic wave within specific frequency ranges. Due to the combined effects of multiple mechanisms in the composite structure, it is difficult to derive and express the physical properties using theoretical formulas. An AMP structure based on local resonance mechanism was proposed, and the bandgap characteristics were analyzed using finite element method, to obtain different cell structure parameters and corresponding bandgap characteristics. A sample set of the relationship between the structure parameters and corresponding bandgap characteristics was established for the AMP. A forward neural network model was designed to obtain the mapping relationship between structural parameters and bandgap range, and the forward prediction of the corresponding bandgap range by inputting structural parameters can be achieved. On this basis, a reverse neural network was proposed, and the reverse design can be carried out to obtain the structural parameters of AMP aiming the required bandgap range. The simulation results show that complex and cumbersome theoretical derivation and calculation can be avoided with the forward-reverse design method for AMP, which is helpful to promote further development of acoustic metamaterials.
2025, 39(3): 354-359.
doi: 10.12299/jsues.24-0143
Abstract:
To address the poor classification performance of traditional machine learning models in the drug-target prediction, a problem caused by their shallow structure and complex data features, a novel prediction model DMS-DF was proposed. The model was based on the deep forest algorithm, the model incorporated a dynamic adaptive multi-granularity scanning mechanism. Furthermore, CatBoost and XGBoost were selected as cascade forest-based classifiers. It demonstrates that that the DMS-DF model outperforms the other four models in terms of drug-target prediction on the same dataset, providing a novel approach for drug discovery.
To address the poor classification performance of traditional machine learning models in the drug-target prediction, a problem caused by their shallow structure and complex data features, a novel prediction model DMS-DF was proposed. The model was based on the deep forest algorithm, the model incorporated a dynamic adaptive multi-granularity scanning mechanism. Furthermore, CatBoost and XGBoost were selected as cascade forest-based classifiers. It demonstrates that that the DMS-DF model outperforms the other four models in terms of drug-target prediction on the same dataset, providing a novel approach for drug discovery.
2025, 39(3): 360-365.
doi: 10.12299/jsues.24-0147
Abstract:
The early diagnosis of breast cancer can significantly improve the possibility of cure. In recent years, the boom of big data and artificial intelligence technology provides technical support for early diagnosis of many diseases, including breast cancer. In order to improve the accuracy of breast cancer diagnosis, an improved Stacking integration model based on area under curve (AUC) was constructed. Firstly, an AdaBoost ensemble model based on$v$ -SVM is constructed and used as a meta learner for Stacking. Secondly, the overall AUC values of each base learner were used to weight the training results of each base learner, and the weighted results were used as the training set of the meta learner for training. Finally, empirical analysis was conducted on the WDBC and WBC datasets. The experimental results show that the Stacking ensemble model based on AUC improvement can achieve high accuracy on two datasets, provide doctors with more refined and personalized diagnostic criteria, thereby achieving the goal of earlier intervention and more efficient treatment.
The early diagnosis of breast cancer can significantly improve the possibility of cure. In recent years, the boom of big data and artificial intelligence technology provides technical support for early diagnosis of many diseases, including breast cancer. In order to improve the accuracy of breast cancer diagnosis, an improved Stacking integration model based on area under curve (AUC) was constructed. Firstly, an AdaBoost ensemble model based on
2025, 39(3): 366-374.
doi: 10.12299/jsues.24-0125
Abstract:
To achieve the accurate positioning of automated welding of aviation electrical connector (AEC), a method for the detection and segmentation of welding cup profiles was proposed based on machine learning. The effectiveness of feature extraction and prediction accuracy of the original network model were enhanced by incorporating a small target detection layer, the CBAM mechanism, and the GhostNet network. Concurrently, the number of parameters and space size of the improved model were reduced. The experimental results show that the improved YOLOv5s-Seg model achieves mean average precision of 84.2% and 44.6%. Compared with the original YOLOv5s model, this represents improved by 5.5% and 1.3%, respectively. The detection-segmentation method proposed effectively balances precision and speed, facilitating practical application and equipment deployment, and provides a theoretical basis for advancing the automated welding of AEC based on machine vision.
To achieve the accurate positioning of automated welding of aviation electrical connector (AEC), a method for the detection and segmentation of welding cup profiles was proposed based on machine learning. The effectiveness of feature extraction and prediction accuracy of the original network model were enhanced by incorporating a small target detection layer, the CBAM mechanism, and the GhostNet network. Concurrently, the number of parameters and space size of the improved model were reduced. The experimental results show that the improved YOLOv5s-Seg model achieves mean average precision of 84.2% and 44.6%. Compared with the original YOLOv5s model, this represents improved by 5.5% and 1.3%, respectively. The detection-segmentation method proposed effectively balances precision and speed, facilitating practical application and equipment deployment, and provides a theoretical basis for advancing the automated welding of AEC based on machine vision.
2017, 31(2): 174-177,182.
doi: 10.3969/j.issn.1009-444X.2017.02.016
摘要：
随着纺织印染行业的不断发展,纺织印染废水成分也日益复杂,大大增加了废水处理难度.针对纺织印染废水的来源和水质特点,综述了国内外纺织印染废水处理技术的研究进展,包括物理处理法、化学处理法和生物处理法,分析了各类处理方法的机制、处理效果及其优缺点.对纺织印染废水处理技术的发展前景进行了展望,指出优化组合废水处理工艺可提高纺织印染废水处理效果及综合回用率.
随着纺织印染行业的不断发展,纺织印染废水成分也日益复杂,大大增加了废水处理难度.针对纺织印染废水的来源和水质特点,综述了国内外纺织印染废水处理技术的研究进展,包括物理处理法、化学处理法和生物处理法,分析了各类处理方法的机制、处理效果及其优缺点.对纺织印染废水处理技术的发展前景进行了展望,指出优化组合废水处理工艺可提高纺织印染废水处理效果及综合回用率.
2018, 32(1): 64-67.
doi: 10.3969/j.issn.1009-444X.2018.01.014
摘要：
交通数据质量是影响交通控制方法有效实施的关键因素之一.为进一步提升交通数据的准确性和时效性,对比分析交通预处理方法,采用标准差公式对拉依达准则进行优化,设计一种适用于交通异常数据检测和修复的优化方法,并结合上海城市快速路实际数据对模型有效性和时效性进行检验.结果表明,基于拉依达准则的数据处理优化方法能够有效、实时地检测交通异常数据、改善数据质量,为道路交通状况的监测和预警提供数据支持.
交通数据质量是影响交通控制方法有效实施的关键因素之一.为进一步提升交通数据的准确性和时效性,对比分析交通预处理方法,采用标准差公式对拉依达准则进行优化,设计一种适用于交通异常数据检测和修复的优化方法,并结合上海城市快速路实际数据对模型有效性和时效性进行检验.结果表明,基于拉依达准则的数据处理优化方法能够有效、实时地检测交通异常数据、改善数据质量,为道路交通状况的监测和预警提供数据支持.
2020, 34(3): 238-246.
doi: 10.3969/j.issn.1009-444X.2020.03.006
摘要：
针对机械臂关节空间轨迹规划的时间优化问题,结合机械臂运动约束,提出基于非线性动态改变惯性权重的粒子群优化(NPSO)算法.根据传统3-5-3多项式插值方法,采用改进粒子群算法寻求最短关节运动时间,研究线性递减改变惯性权重(LPSO)算法和NPSO算法的性能,选用NPSO算法完成关节运动时间最优求解.研究结果显示,经时间优化后的3-5-3插值曲线连续光滑且具备更好的运动特性,整体运动时间缩短约26%,证实提出的方法具有可行性.
针对机械臂关节空间轨迹规划的时间优化问题,结合机械臂运动约束,提出基于非线性动态改变惯性权重的粒子群优化(NPSO)算法.根据传统3-5-3多项式插值方法,采用改进粒子群算法寻求最短关节运动时间,研究线性递减改变惯性权重(LPSO)算法和NPSO算法的性能,选用NPSO算法完成关节运动时间最优求解.研究结果显示,经时间优化后的3-5-3插值曲线连续光滑且具备更好的运动特性,整体运动时间缩短约26%,证实提出的方法具有可行性.
2017, 31(4): 371-375.
doi: 10.3969/j.issn.1009-444X.2017.04.017
摘要：
老年人参与各项社会活动,能够延缓衰老,促进身体健康、提高生活质量.通过开展调研探究了老年人的社会参与意愿及其影响因素.调查发现,城市老年人对经济、政治、社区、文化等各项活动的参与意愿都较高,但是多数的参与停留在被动层面;年龄、性别、文化程度、健康状况等因素会影响参与意愿,且不同活动受影响的程度不同;参与意愿会受到现实条件的约束,如照看小孩,料理家务等.
老年人参与各项社会活动,能够延缓衰老,促进身体健康、提高生活质量.通过开展调研探究了老年人的社会参与意愿及其影响因素.调查发现,城市老年人对经济、政治、社区、文化等各项活动的参与意愿都较高,但是多数的参与停留在被动层面;年龄、性别、文化程度、健康状况等因素会影响参与意愿,且不同活动受影响的程度不同;参与意愿会受到现实条件的约束,如照看小孩,料理家务等.
2021, 35(3): 237-242.
摘要：
针对传统光流法处理视频序列时存在运行效率低及跟踪偏移问题，结合粒子滤波模型提出一种改进光流法的视频目标跟踪技术. 该技术首先通过遍历法搜索运动点，采用质心定位方式捕获目标质心坐标，然后将得到的视频序列进行光流处理，最后经粒子滤波求解质心运动信息，以实现对视频中目标的精确检测与追踪. 在不同场景下对不同数量、不同类型目标进行仿真跟踪试验，并与光流法、ViBe算法及YOLO算法进行比较. 仿真结果表明，该跟踪技术可使目标跟踪的精准率有效提高5.2%，跟踪效率提高13.7%，同时表现出较好的鲁棒性.
针对传统光流法处理视频序列时存在运行效率低及跟踪偏移问题，结合粒子滤波模型提出一种改进光流法的视频目标跟踪技术. 该技术首先通过遍历法搜索运动点，采用质心定位方式捕获目标质心坐标，然后将得到的视频序列进行光流处理，最后经粒子滤波求解质心运动信息，以实现对视频中目标的精确检测与追踪. 在不同场景下对不同数量、不同类型目标进行仿真跟踪试验，并与光流法、ViBe算法及YOLO算法进行比较. 仿真结果表明，该跟踪技术可使目标跟踪的精准率有效提高5.2%，跟踪效率提高13.7%，同时表现出较好的鲁棒性.
Numerical Simulation Analysis on Air Distribution of Inter-Column Air Conditioning in a Machine Room
2018, 32(3): 244-249.
doi: 10.3969/j.issn.1009-444X.2018.03.009
摘要：
以上海某数据中心机房为研究对象,针对机房服务器散热量大的特点及传统封闭冷通道存在的问题,提出适合该机房的封闭热通道方案.采用计算流体动力学(CFD)模拟的方法,分析机房内封闭热通道与传统封闭冷通道的温度场和速度场分布特性.通过模拟分析得出:大散热量机房列间空调采用封闭热通道形式效果更佳.
以上海某数据中心机房为研究对象,针对机房服务器散热量大的特点及传统封闭冷通道存在的问题,提出适合该机房的封闭热通道方案.采用计算流体动力学(CFD)模拟的方法,分析机房内封闭热通道与传统封闭冷通道的温度场和速度场分布特性.通过模拟分析得出:大散热量机房列间空调采用封闭热通道形式效果更佳.
2019, 33(3): 215-218,277.
doi: 10.3969/j.issn.1009-444X.2019.03.005
摘要：
综述了二氧化硅(SiO2)薄膜的制备方法和研究进展,介绍了磁控溅射、溶胶凝胶法和真空镀等SiO2薄膜制备方法及其优缺点,论述SiO2薄膜在光学、电学和光电等性能研究方面的新进展,最后对SiO2薄膜的应用和发展进行了展望.
综述了二氧化硅(SiO2)薄膜的制备方法和研究进展,介绍了磁控溅射、溶胶凝胶法和真空镀等SiO2薄膜制备方法及其优缺点,论述SiO2薄膜在光学、电学和光电等性能研究方面的新进展,最后对SiO2薄膜的应用和发展进行了展望.
2018, 32(3): 214-220.
doi: 10.3969/j.issn.1009-444X.2018.03.004
摘要：
介绍功能纺织品的分类,分析功能纺织品的开发途径,主要阐述国内现代功能纺织品研究及开发进展,包括纳米光触媒多功能纺织品、防电磁辐射纺织品、超疏水多功能纺织品、阻燃纺织品、防蚊虫纺织品,并指出采用高新技术开发生态安全高附加值多功能纺织品是功能性纺织品的发展趋势.
介绍功能纺织品的分类,分析功能纺织品的开发途径,主要阐述国内现代功能纺织品研究及开发进展,包括纳米光触媒多功能纺织品、防电磁辐射纺织品、超疏水多功能纺织品、阻燃纺织品、防蚊虫纺织品,并指出采用高新技术开发生态安全高附加值多功能纺织品是功能性纺织品的发展趋势.
2017, 31(1): 90-94.
doi: 10.3969/j.issn.1009-444X.2017.01.019
摘要：
针对我国黄金期货价格预测问题,对影响我国黄金期货价格的5项指标进行灰色关联度分析,得出我国黄金期货价格与美国黄金期货价格之间的关联度最高.建立反向传播(BP)神经网络模型对我国黄金期货价格预测,并与GM(1,1)方法和ARIMA(0,2,1)模型下的预测结果进行对比.结果显示:与后两个模型相比,BP神经网络模型在黄金期货价格预测方面的精确性更高,具有较好的实用价值.
针对我国黄金期货价格预测问题,对影响我国黄金期货价格的5项指标进行灰色关联度分析,得出我国黄金期货价格与美国黄金期货价格之间的关联度最高.建立反向传播(BP)神经网络模型对我国黄金期货价格预测,并与GM(1,1)方法和ARIMA(0,2,1)模型下的预测结果进行对比.结果显示:与后两个模型相比,BP神经网络模型在黄金期货价格预测方面的精确性更高,具有较好的实用价值.
2018, 32(4): 324-328.
doi: 10.3969/j.issn.1009-444X.2018.04.007
摘要：
综述人工神经网络在轧制参数预报中的应用,包括轧制力能参数、组织性能和温度的预报;并介绍了混合人工神经网络模型与改进的人工神经网络模型,其中混合对象包括传统数学模型、有限元和其他智能算法.研究表明,人工神经网络在轧钢参数预报中具备很大的优势,能显著提高轧钢相关参数预报精度,提升轧材的品质.
综述人工神经网络在轧制参数预报中的应用,包括轧制力能参数、组织性能和温度的预报;并介绍了混合人工神经网络模型与改进的人工神经网络模型,其中混合对象包括传统数学模型、有限元和其他智能算法.研究表明,人工神经网络在轧钢参数预报中具备很大的优势,能显著提高轧钢相关参数预报精度,提升轧材的品质.
2003, 17(2): 83-86.
doi: 10.3969/j.issn.1009-444X.2003.02.001
Abstract:
2019, 33(3): 283-284,288.
doi: 10.3969/j.issn.1009-444X.2019.03.017
Abstract:
About Journal
More>
Chinese Quarterly
Supervisor: Shanghai Municipal Education Commission
Sponsors by: Shanghai University of Engineering Science
Publisher: Editorial Department of Journal of Shanghai University of Engineering Science
Editor-in-Chief: WANG Yansong
Email: xuebao@sues.edu.cn
ISSN1009-444X
CN31-1598/T
Link
More>
Email Alert
RSS
