Modern manufacturing processes have become more reliant on automation because of the accelerated transition from Industry 3.0 to Industry 4.0.Manual inspection of products on assembly lines remains inefficient,prone t...Modern manufacturing processes have become more reliant on automation because of the accelerated transition from Industry 3.0 to Industry 4.0.Manual inspection of products on assembly lines remains inefficient,prone to errors and lacks consistency,emphasizing the need for a reliable and automated inspection system.Leveraging both object detection and image segmentation approaches,this research proposes a vision-based solution for the detection of various kinds of tools in the toolkit using deep learning(DL)models.Two Intel RealSense D455f depth cameras were arranged in a top down configuration to capture both RGB and depth images of the toolkits.After applying multiple constraints and enhancing them through preprocessing and augmentation,a dataset consisting of 3300 annotated RGB-D photos was generated.Several DL models were selected through a comprehensive assessment of mean Average Precision(mAP),precision-recall equilibrium,inference latency(target≥30 FPS),and computational burden,resulting in a preference for YOLO and Region-based Convolutional Neural Networks(R-CNN)variants over ViT-based models due to the latter’s increased latency and resource requirements.YOLOV5,YOLOV8,YOLOV11,Faster R-CNN,and Mask R-CNN were trained on the annotated dataset and evaluated using key performance metrics(Recall,Accuracy,F1-score,and Precision).YOLOV11 demonstrated balanced excellence with 93.0%precision,89.9%recall,and a 90.6%F1-score in object detection,as well as 96.9%precision,95.3%recall,and a 96.5%F1-score in instance segmentation with an average inference time of 25 ms per frame(≈40 FPS),demonstrating real-time performance.Leveraging these results,a YOLOV11-based windows application was successfully deployed in a real-time assembly line environment,where it accurately processed live video streams to detect and segment tools within toolkits,demonstrating its practical effectiveness in industrial automation.The application is capable of precisely measuring socket dimensions by utilising edge detection techniques on YOLOv11 segmentation masks,in addition to detection and segmentation.This makes it possible to do specification-level quality control right on the assembly line,which improves the ability to examine things in real time.The implementation is a big step forward for intelligent manufacturing in the Industry 4.0 paradigm.It provides a scalable,efficient,and accurate way to do automated inspection and dimensional verification activities.展开更多
Despite the gradual transformation of traditional manufacturing by the Human-Robot Collaboration Assembly(HRCA),challenges remain in the robot’s ability to understand and predict human assembly intentions.This study ...Despite the gradual transformation of traditional manufacturing by the Human-Robot Collaboration Assembly(HRCA),challenges remain in the robot’s ability to understand and predict human assembly intentions.This study aims to enhance the robot’s comprehension and prediction capabilities of operator assembly intentions by capturing and analyzing operator behavior and movements.We propose a video feature extraction method based on the Temporal Shift Module Network(TSM-ResNet50)to extract spatiotemporal features from assembly videos and differentiate various assembly actions using feature differences between video frames.Furthermore,we construct an action recognition and segmentation model based on the Refined-Multi-Scale Temporal Convolutional Network(Refined-MS-TCN)to identify assembly action intervals and accurately acquire action categories.Experiments on our self-built reducer assembly action dataset demonstrate that our network can classify assembly actions frame by frame,achieving an accuracy rate of 83%.Additionally,we develop a HiddenMarkovModel(HMM)integrated with assembly task constraints to predict operator assembly intentions based on the probability transition matrix and assembly task constraints.The experimental results show that our method for predicting operator assembly intentions can achieve an accuracy of 90.6%,which is a 13.3%improvement over the HMM without task constraints.展开更多
Segmentally assembled bridges are increasinglyfinding engineering applications in recent years due to their unique advantages,especially as urban viaducts.Vehicle loads are one of the most important variable loads acti...Segmentally assembled bridges are increasinglyfinding engineering applications in recent years due to their unique advantages,especially as urban viaducts.Vehicle loads are one of the most important variable loads acting on bridge structures.Accordingly,the influence of overloaded vehicles on existing assembled bridge structures is an urgent concern at present.This paper establishes thefinite element model of the segmentally assembled bridge based on ABAQUS software and analyzes the influence of vehicle overload on an assembled girder bridge struc-ture.First,afinite element model corresponding to the target bridge is established based on ABAQUS software,and the load is controlled to simulate vehicle movement in each area of the traveling zone at different times.Sec-ond,the key cross-sections of segmental girder bridges are monitored in real time based on the force character-istics of continuous girder bridges,and they are compared with the simulation results.Finally,a material damage ontology model is introduced,and the structural damage caused by different overloading rates is compared and analyzed.Results show that thefinite element modeling method is accurate by comparing with on-site measured data,and it is suitable for the numerical simulation of segmental girder bridges;Dynamic sensors installed at 1/4L,1/2L,and 3/4L of the segmental girder main beams could be used to identify the dynamic response of segmental girder bridges;The bottom plate of the segmental girder bridge is mostly damaged at the position where the length of the precast beam section changes and the midspan position.With the increase in load,damage in the direction of the bridge develops faster than that in the direction of the transverse bridge.Thefindings of this study can guide maintenance departments in the management and maintenance of bridges and vehicles.展开更多
We develop assembled reinforcement structures(ARSs)composed of connection parts,connecting rods,and straight bolts to strengthen segmental joints in the lining of shield tunnels.Through full-scale bending experiments ...We develop assembled reinforcement structures(ARSs)composed of connection parts,connecting rods,and straight bolts to strengthen segmental joints in the lining of shield tunnels.Through full-scale bending experiments and numerical simulations,we investigate the deformation and failure characteristics of segmental joints strengthened by ARSs,and propose a novel optimization method for ARSs.The experimental results show that the ARSs can effectively limit the opening of a segmental joint,but also that separation can occur during loading if the connection between the ARSs and segments is not designed properly.Importantly,this connection can be improved by embedding anchor parts in the concrete.In numerical modeling,we investigate the failure modes of segmental joints strengthened by ARSs for both positive bending and negative bending loading cases.In the case of positive bending loading,first the concrete around the anchor parts cracks,and subsequently the concrete on the external side of the joint is crushed.The joint failure is caused by the crushing of concrete on the external side of the joint.While the un-strengthened segmental joint fails with an opening of 5.884 mm,the strengthened segmental joint only opens by 0.288 mm under the same loading,corresponding to a reduction of 95.1%.In the case of negative bending loading,the concrete around the anchor parts first cracks,and then the amount of joint opening exceeds a limiting value for waterproofing(6 mm),i.e.,the joint’s failure is caused by water leakage.While the opening of the un-strengthened segmental joint is 9.033 mm and experiences waterproofing failure,the opening of the strengthened segmental joint is only 2.793 mm under the same loading,corresponding to a reduction of 69.1%.When constructing a new shield tunnel,anchor parts could be embedded in the concrete segments in tandem with ARSs for improved resistance to joint opening.For existing shield tunnel linings,anchor parts cannot be embedded in the concrete segments;therefore,the connections between the ARSs and concrete need to be optimized to strengthen the segmental joint.展开更多
为解决大型汽车卡车运输船(Pure Car and Truck Carrier,PCTC)折叠式艉门建造难点,以某8600标准车位(Car Equivalent Unit,CEU)PCTC为例,介绍折叠式艉门结构特点与安装要求,分析折叠式艉门建造工艺现状,进行折叠式艉门建造工艺创新。应...为解决大型汽车卡车运输船(Pure Car and Truck Carrier,PCTC)折叠式艉门建造难点,以某8600标准车位(Car Equivalent Unit,CEU)PCTC为例,介绍折叠式艉门结构特点与安装要求,分析折叠式艉门建造工艺现状,进行折叠式艉门建造工艺创新。应用结果表明,工艺创新可提高折叠式艉门建造精度和舾装完整性,有效缩短船坞(船台)周期并减少码头调试风险,具有一定推广价值。展开更多
Automatic segment assembly,which increases boring efficiency and construction safety,is a trend in tunnel boring.However,the current situation still relies on manual operation and experience.Electro-hydraulic rotation...Automatic segment assembly,which increases boring efficiency and construction safety,is a trend in tunnel boring.However,the current situation still relies on manual operation and experience.Electro-hydraulic rotation systems are crucial in segment grasping,transporting,and assembly.This article presents the automation of rotation systems in segment assembly to improve motion smoothness and accuracy.As a result of strong nonlinearity and system complexity,parameter estimation is performed by using a noise reduction method based on multialgorithm fusion and the stochastic gradient deviation correction recursive least squares identification algorithm.Active disturbance rejection control(ADRC)is introduced into sliding mode control(SMC)to compensate for model uncertainty and disturbance.An improved cuckoo algorithm is used to optimize influential parameters in ADRC.Moreover,full-scale bench tests are conducted to verify the proposed system automation.Results indicate that the proposed method has a better displacement tracking performance and lower tracking error than the ADRC,SMC,and proportional-integral-derivative methods.Furthermore,such a procedure facilitates the success rate of complete segment assembly.展开更多
文摘Modern manufacturing processes have become more reliant on automation because of the accelerated transition from Industry 3.0 to Industry 4.0.Manual inspection of products on assembly lines remains inefficient,prone to errors and lacks consistency,emphasizing the need for a reliable and automated inspection system.Leveraging both object detection and image segmentation approaches,this research proposes a vision-based solution for the detection of various kinds of tools in the toolkit using deep learning(DL)models.Two Intel RealSense D455f depth cameras were arranged in a top down configuration to capture both RGB and depth images of the toolkits.After applying multiple constraints and enhancing them through preprocessing and augmentation,a dataset consisting of 3300 annotated RGB-D photos was generated.Several DL models were selected through a comprehensive assessment of mean Average Precision(mAP),precision-recall equilibrium,inference latency(target≥30 FPS),and computational burden,resulting in a preference for YOLO and Region-based Convolutional Neural Networks(R-CNN)variants over ViT-based models due to the latter’s increased latency and resource requirements.YOLOV5,YOLOV8,YOLOV11,Faster R-CNN,and Mask R-CNN were trained on the annotated dataset and evaluated using key performance metrics(Recall,Accuracy,F1-score,and Precision).YOLOV11 demonstrated balanced excellence with 93.0%precision,89.9%recall,and a 90.6%F1-score in object detection,as well as 96.9%precision,95.3%recall,and a 96.5%F1-score in instance segmentation with an average inference time of 25 ms per frame(≈40 FPS),demonstrating real-time performance.Leveraging these results,a YOLOV11-based windows application was successfully deployed in a real-time assembly line environment,where it accurately processed live video streams to detect and segment tools within toolkits,demonstrating its practical effectiveness in industrial automation.The application is capable of precisely measuring socket dimensions by utilising edge detection techniques on YOLOv11 segmentation masks,in addition to detection and segmentation.This makes it possible to do specification-level quality control right on the assembly line,which improves the ability to examine things in real time.The implementation is a big step forward for intelligent manufacturing in the Industry 4.0 paradigm.It provides a scalable,efficient,and accurate way to do automated inspection and dimensional verification activities.
文摘Despite the gradual transformation of traditional manufacturing by the Human-Robot Collaboration Assembly(HRCA),challenges remain in the robot’s ability to understand and predict human assembly intentions.This study aims to enhance the robot’s comprehension and prediction capabilities of operator assembly intentions by capturing and analyzing operator behavior and movements.We propose a video feature extraction method based on the Temporal Shift Module Network(TSM-ResNet50)to extract spatiotemporal features from assembly videos and differentiate various assembly actions using feature differences between video frames.Furthermore,we construct an action recognition and segmentation model based on the Refined-Multi-Scale Temporal Convolutional Network(Refined-MS-TCN)to identify assembly action intervals and accurately acquire action categories.Experiments on our self-built reducer assembly action dataset demonstrate that our network can classify assembly actions frame by frame,achieving an accuracy rate of 83%.Additionally,we develop a HiddenMarkovModel(HMM)integrated with assembly task constraints to predict operator assembly intentions based on the probability transition matrix and assembly task constraints.The experimental results show that our method for predicting operator assembly intentions can achieve an accuracy of 90.6%,which is a 13.3%improvement over the HMM without task constraints.
基金supported in part by the Key Research Projects of Higher Education Institutions in Henan Province(Grant No.24A560021)in part by the Henan Postdoctoral Foundation(Grant No.202102015).
文摘Segmentally assembled bridges are increasinglyfinding engineering applications in recent years due to their unique advantages,especially as urban viaducts.Vehicle loads are one of the most important variable loads acting on bridge structures.Accordingly,the influence of overloaded vehicles on existing assembled bridge structures is an urgent concern at present.This paper establishes thefinite element model of the segmentally assembled bridge based on ABAQUS software and analyzes the influence of vehicle overload on an assembled girder bridge struc-ture.First,afinite element model corresponding to the target bridge is established based on ABAQUS software,and the load is controlled to simulate vehicle movement in each area of the traveling zone at different times.Sec-ond,the key cross-sections of segmental girder bridges are monitored in real time based on the force character-istics of continuous girder bridges,and they are compared with the simulation results.Finally,a material damage ontology model is introduced,and the structural damage caused by different overloading rates is compared and analyzed.Results show that thefinite element modeling method is accurate by comparing with on-site measured data,and it is suitable for the numerical simulation of segmental girder bridges;Dynamic sensors installed at 1/4L,1/2L,and 3/4L of the segmental girder main beams could be used to identify the dynamic response of segmental girder bridges;The bottom plate of the segmental girder bridge is mostly damaged at the position where the length of the precast beam section changes and the midspan position.With the increase in load,damage in the direction of the bridge develops faster than that in the direction of the transverse bridge.Thefindings of this study can guide maintenance departments in the management and maintenance of bridges and vehicles.
基金supported by the National Natural Science Foundation of China(No.52008308)the China Postdoctoral Science Foundation(Nos.BX20200247 and 2021M692447)the Research Project from Jinan Rail Transit Group Co.,Ltd.and China Railway No.5 Engineering Group Co.,Ltd.(No.R2-ZF-2019-039).
文摘We develop assembled reinforcement structures(ARSs)composed of connection parts,connecting rods,and straight bolts to strengthen segmental joints in the lining of shield tunnels.Through full-scale bending experiments and numerical simulations,we investigate the deformation and failure characteristics of segmental joints strengthened by ARSs,and propose a novel optimization method for ARSs.The experimental results show that the ARSs can effectively limit the opening of a segmental joint,but also that separation can occur during loading if the connection between the ARSs and segments is not designed properly.Importantly,this connection can be improved by embedding anchor parts in the concrete.In numerical modeling,we investigate the failure modes of segmental joints strengthened by ARSs for both positive bending and negative bending loading cases.In the case of positive bending loading,first the concrete around the anchor parts cracks,and subsequently the concrete on the external side of the joint is crushed.The joint failure is caused by the crushing of concrete on the external side of the joint.While the un-strengthened segmental joint fails with an opening of 5.884 mm,the strengthened segmental joint only opens by 0.288 mm under the same loading,corresponding to a reduction of 95.1%.In the case of negative bending loading,the concrete around the anchor parts first cracks,and then the amount of joint opening exceeds a limiting value for waterproofing(6 mm),i.e.,the joint’s failure is caused by water leakage.While the opening of the un-strengthened segmental joint is 9.033 mm and experiences waterproofing failure,the opening of the strengthened segmental joint is only 2.793 mm under the same loading,corresponding to a reduction of 69.1%.When constructing a new shield tunnel,anchor parts could be embedded in the concrete segments in tandem with ARSs for improved resistance to joint opening.For existing shield tunnel linings,anchor parts cannot be embedded in the concrete segments;therefore,the connections between the ARSs and concrete need to be optimized to strengthen the segmental joint.
文摘为解决大型汽车卡车运输船(Pure Car and Truck Carrier,PCTC)折叠式艉门建造难点,以某8600标准车位(Car Equivalent Unit,CEU)PCTC为例,介绍折叠式艉门结构特点与安装要求,分析折叠式艉门建造工艺现状,进行折叠式艉门建造工艺创新。应用结果表明,工艺创新可提高折叠式艉门建造精度和舾装完整性,有效缩短船坞(船台)周期并减少码头调试风险,具有一定推广价值。
基金supported by the Natural Science Foundation of Zhejiang Province,China(Grant No.LD22E050003)the National Natural Science Foundation of China(Grant No.52222503).
文摘Automatic segment assembly,which increases boring efficiency and construction safety,is a trend in tunnel boring.However,the current situation still relies on manual operation and experience.Electro-hydraulic rotation systems are crucial in segment grasping,transporting,and assembly.This article presents the automation of rotation systems in segment assembly to improve motion smoothness and accuracy.As a result of strong nonlinearity and system complexity,parameter estimation is performed by using a noise reduction method based on multialgorithm fusion and the stochastic gradient deviation correction recursive least squares identification algorithm.Active disturbance rejection control(ADRC)is introduced into sliding mode control(SMC)to compensate for model uncertainty and disturbance.An improved cuckoo algorithm is used to optimize influential parameters in ADRC.Moreover,full-scale bench tests are conducted to verify the proposed system automation.Results indicate that the proposed method has a better displacement tracking performance and lower tracking error than the ADRC,SMC,and proportional-integral-derivative methods.Furthermore,such a procedure facilitates the success rate of complete segment assembly.
