Pub Date : 2024-01-10DOI: 10.1088/1361-6501/ad1d2f
Mingyue Liu, R. Tu, Fangxin Li, Qiushi Chen, Qi Li, Junmei Chen, Pengfei Zhang, Xiaochun Lu
With the continuous development and increasing popularity of the fifth generation (5G) of mobile communications technology, its fusion with the Global Navigation Satellite System (GNSS) data can effectively improve problems with service interruptions or poor satellite signal reception. In this study, GPS and 5G data were fused, and the resulting experimental algorithm showed that it effective improves time transfer’s frequency stability and reliability. Moreover, this technique reduced the noise level of time-transmitted clock offset sequences, while suppressing short-term mutations. By simulating different degrees of satellite signal occlusion, it was further verified that the GPS+5G fusion method can provide stable, high-precision, and real-time delivery services under insufficient satellite signal reception. This provides a reference for high-precision time transfer technology in complex environments and further improves the reliability of GNSS high-precision time transfer.
{"title":"GPS + 5G fusion for high-precision time transfer","authors":"Mingyue Liu, R. Tu, Fangxin Li, Qiushi Chen, Qi Li, Junmei Chen, Pengfei Zhang, Xiaochun Lu","doi":"10.1088/1361-6501/ad1d2f","DOIUrl":"https://doi.org/10.1088/1361-6501/ad1d2f","url":null,"abstract":"\u0000 With the continuous development and increasing popularity of the fifth generation (5G) of mobile communications technology, its fusion with the Global Navigation Satellite System (GNSS) data can effectively improve problems with service interruptions or poor satellite signal reception. In this study, GPS and 5G data were fused, and the resulting experimental algorithm showed that it effective improves time transfer’s frequency stability and reliability. Moreover, this technique reduced the noise level of time-transmitted clock offset sequences, while suppressing short-term mutations. By simulating different degrees of satellite signal occlusion, it was further verified that the GPS+5G fusion method can provide stable, high-precision, and real-time delivery services under insufficient satellite signal reception. This provides a reference for high-precision time transfer technology in complex environments and further improves the reliability of GNSS high-precision time transfer.","PeriodicalId":18526,"journal":{"name":"Measurement Science and Technology","volume":"82 6","pages":""},"PeriodicalIF":2.4,"publicationDate":"2024-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139440445","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The aviation community is actively pursuing advanced receiver autonomous integrity monitoring (ARAIM) to enhance the safety of aircraft navigation services. Protection level calculation is a crucial task in the solution separation-based ARAIM as it determines the availability of the ARAIM. Accurately determining the worst-case fault bias (WCFB) is beneficial in improving the bounding tightness of protection level on positioning error. Unfortunately, the WCFB determination is a challenging task that requires a time-consuming searching procedure, especially when dealing with the multi-satellite faults. The traditional ARAIM protection level is achieved by constructing an over-conservative worst-case positioning error bound to avoid the unacceptable time-consumption of the brute-force searching for multi-satellite WCFBs. However, this approach comes at the cost of losing the tightness of the protection level and the availability of the ARAIM. The ARAIM milestone reports have pointed out that the availability of the baseline ARAIM needs to be continuously improved in order to satisfy the worldwide localizer precision vertical 200 (LPV-200) requirements. In response, this paper proposes a novel multi-satellite WCFBs searching method for the ARAIM to improve the tightness of protection level. The method consists of determining the worst-case fault direction and constructing an efficient WCFBs searching procedure. GPS/Galileo dual-constellation simulation result demonstrates that the proposed method not only can improve the availability of the ARAIM up to 9.33% when compared with the traditional method, but also achieves comparable computation efficiency.
{"title":"Improved protection level for the solution-separation ARAIM based on worst-case fault bias searching","authors":"Ruijie Li, Liang Li, jiachang jiang, Fengze Du, Zhibo Na, Xin Xu","doi":"10.1088/1361-6501/ad1d2b","DOIUrl":"https://doi.org/10.1088/1361-6501/ad1d2b","url":null,"abstract":"\u0000 The aviation community is actively pursuing advanced receiver autonomous integrity monitoring (ARAIM) to enhance the safety of aircraft navigation services. Protection level calculation is a crucial task in the solution separation-based ARAIM as it determines the availability of the ARAIM. Accurately determining the worst-case fault bias (WCFB) is beneficial in improving the bounding tightness of protection level on positioning error. Unfortunately, the WCFB determination is a challenging task that requires a time-consuming searching procedure, especially when dealing with the multi-satellite faults. The traditional ARAIM protection level is achieved by constructing an over-conservative worst-case positioning error bound to avoid the unacceptable time-consumption of the brute-force searching for multi-satellite WCFBs. However, this approach comes at the cost of losing the tightness of the protection level and the availability of the ARAIM. The ARAIM milestone reports have pointed out that the availability of the baseline ARAIM needs to be continuously improved in order to satisfy the worldwide localizer precision vertical 200 (LPV-200) requirements. In response, this paper proposes a novel multi-satellite WCFBs searching method for the ARAIM to improve the tightness of protection level. The method consists of determining the worst-case fault direction and constructing an efficient WCFBs searching procedure. GPS/Galileo dual-constellation simulation result demonstrates that the proposed method not only can improve the availability of the ARAIM up to 9.33% when compared with the traditional method, but also achieves comparable computation efficiency.","PeriodicalId":18526,"journal":{"name":"Measurement Science and Technology","volume":"2 12","pages":""},"PeriodicalIF":2.4,"publicationDate":"2024-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139440247","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-10DOI: 10.1088/1361-6501/ad1d2e
Huaian Yi, Xiao Lv, Ai-qin Shu, Hao Wang, Kai Shi
The traditional deep learning method for detecting workpiece surface roughness relies heavily on a large number of training samples. Also, when detecting the surface roughness of workpieces processed by different machining techniques, it requires a large number of samples of that workpiece to rebuild the model. To address these problems, this paper proposes a few-sample visual detection method for the surface roughness of workpieces processed by different techniques. This method first trains a base model using a relatively large amount of samples from one machining technique, then fine-tunes the model using small amounts of samples from workpieces of different techniques. By introducing contrastive proposal encoding into Faster R-CNN, the model's ability to learn surface features from small amounts of workpiece samples is enhanced, thus improving the detection accuracy of surface roughness of workpieces processed by different techniques. Experiments show that this method reduces the model's dependence on training samples and the cost of data preparation. It also demonstrates higher accuracy in surface roughness detection tasks of workpieces processed by different techniques, providing a new approach and insights for few-sample surface roughness detection.
{"title":"Few-Shot Detection of Surface Roughness of Workpieces Processed by Different Machining Techniques","authors":"Huaian Yi, Xiao Lv, Ai-qin Shu, Hao Wang, Kai Shi","doi":"10.1088/1361-6501/ad1d2e","DOIUrl":"https://doi.org/10.1088/1361-6501/ad1d2e","url":null,"abstract":"\u0000 The traditional deep learning method for detecting workpiece surface roughness relies heavily on a large number of training samples. Also, when detecting the surface roughness of workpieces processed by different machining techniques, it requires a large number of samples of that workpiece to rebuild the model. To address these problems, this paper proposes a few-sample visual detection method for the surface roughness of workpieces processed by different techniques. This method first trains a base model using a relatively large amount of samples from one machining technique, then fine-tunes the model using small amounts of samples from workpieces of different techniques. By introducing contrastive proposal encoding into Faster R-CNN, the model's ability to learn surface features from small amounts of workpiece samples is enhanced, thus improving the detection accuracy of surface roughness of workpieces processed by different techniques. Experiments show that this method reduces the model's dependence on training samples and the cost of data preparation. It also demonstrates higher accuracy in surface roughness detection tasks of workpieces processed by different techniques, providing a new approach and insights for few-sample surface roughness detection.","PeriodicalId":18526,"journal":{"name":"Measurement Science and Technology","volume":"4 3","pages":""},"PeriodicalIF":2.4,"publicationDate":"2024-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139440037","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-10DOI: 10.1088/1361-6501/ad1d31
Dajun Cai, Jiantao Yao, Weihua Gao, Hongyu Zhang, Zeyu Li
The high sensitivity six-dimensional force sensor is the core component of intelligent robot force feedback perception, which is related to the structure of the sensor. This paper proposed a new modeling method to chain the near singular eight-branch Stewart mechanism, which indicates the relationship between structural parameters and sensitivity. The new method gains 16 new configurations by innovating the traditional Stewart institution. Configuration with high sensitivity in Fz, Mx and My directions is selected for further study. The theoretical stiffness model and force mapping model show that the sensitivity of the new structure can be amplified by 4 times in specific degrees of freedom. The results show that the nonlinear error and coupling error are 2.77% and 2.26%, respectively. The proposed method can be widely applied in the field of parallel six-dimensional force sensors.
{"title":"Design and analysis of parallel six-dimensional force sensor based on near-singular characteristics","authors":"Dajun Cai, Jiantao Yao, Weihua Gao, Hongyu Zhang, Zeyu Li","doi":"10.1088/1361-6501/ad1d31","DOIUrl":"https://doi.org/10.1088/1361-6501/ad1d31","url":null,"abstract":"\u0000 The high sensitivity six-dimensional force sensor is the core component of intelligent robot force feedback perception, which is related to the structure of the sensor. This paper proposed a new modeling method to chain the near singular eight-branch Stewart mechanism, which indicates the relationship between structural parameters and sensitivity. The new method gains 16 new configurations by innovating the traditional Stewart institution. Configuration with high sensitivity in Fz, Mx and My directions is selected for further study. The theoretical stiffness model and force mapping model show that the sensitivity of the new structure can be amplified by 4 times in specific degrees of freedom. The results show that the nonlinear error and coupling error are 2.77% and 2.26%, respectively. The proposed method can be widely applied in the field of parallel six-dimensional force sensors.","PeriodicalId":18526,"journal":{"name":"Measurement Science and Technology","volume":"76 5","pages":""},"PeriodicalIF":2.4,"publicationDate":"2024-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139440829","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-09DOI: 10.1088/1361-6501/ad1cc7
Gaige Chen, Yudong Yang, Hui Qi, Xianzhi Wang, Hongbo Ma
The monitoring data (current and rotational speed) of the subway sliding plug door transmission system changed positively and negatively with the acceleration and weight of the door. How to perceive the changes is a challenging problem in the health state assessment of transmission system. To address this problem, an enhanced perception health state assessment method was proposed for the transmission systems. In the method, firstly, the equivalent resistance force is calculated by monitoring the current and rotational speed data according to mechanical dynamic knowledge. Secondly, the sensitive features of normal and abnormal states are screened out from the enhanced dataset constructed by current, rotational speed data, and equivalent resistance force data. Finally, the health state of the transmission system is assessed using an integrated learning algorithm. The effectiveness of the method is verified by benchmark experimental data, and the results indicate that the method has a higher accuracy with four classifiers and a broader suitable range with varying door acceleration and weight.
{"title":"An enhanced perception health state assessment method for subway sliding plug door transmission system","authors":"Gaige Chen, Yudong Yang, Hui Qi, Xianzhi Wang, Hongbo Ma","doi":"10.1088/1361-6501/ad1cc7","DOIUrl":"https://doi.org/10.1088/1361-6501/ad1cc7","url":null,"abstract":"\u0000 The monitoring data (current and rotational speed) of the subway sliding plug door transmission system changed positively and negatively with the acceleration and weight of the door. How to perceive the changes is a challenging problem in the health state assessment of transmission system. To address this problem, an enhanced perception health state assessment method was proposed for the transmission systems. In the method, firstly, the equivalent resistance force is calculated by monitoring the current and rotational speed data according to mechanical dynamic knowledge. Secondly, the sensitive features of normal and abnormal states are screened out from the enhanced dataset constructed by current, rotational speed data, and equivalent resistance force data. Finally, the health state of the transmission system is assessed using an integrated learning algorithm. The effectiveness of the method is verified by benchmark experimental data, and the results indicate that the method has a higher accuracy with four classifiers and a broader suitable range with varying door acceleration and weight.","PeriodicalId":18526,"journal":{"name":"Measurement Science and Technology","volume":"11 25","pages":""},"PeriodicalIF":2.4,"publicationDate":"2024-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139443396","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Strong winds can make a bridge’s cable-stayed cables produce violent vibrations, leading to fatigue of the cable-stayed cables and damaging the cable-stayed bridge’s structure. Accurately and effectively obtaining data on the wind loads applied to the cable-stayed cables is important for assessing the cable-stayed cables’ health. Currently, the more widely used sensing elements include diffuse silicon piezoresistive sensors and strain gauges. However, most of them present such disadvantages as rigidity, difficult to fit the curved surface, high cost and low sensitivity. In this paper, a conductive hydrogel flexible pressure sensor based on TA/CB@PDMS was developed, using carbon black (CB) as the main conductive medium, with good electrical conductivity, high sensitivity (0.95 kPa-1) and excellent tensile properties (210% tensile breakage). Meanwhile, a salt permeation method(Soak the sensor in LiBr solution) was used to effectively inhibit the sensor's water from being evaporated and frozen. Its substrate incorporates tannic acid (TA) to increase the sensor’s adhesion so that it adheres well to the diagonal cable’s surface. In this paper, the wind speed variation around the diagonal cable and the force distribution on the surface with considering the fluid-structure coupling effect are analyzed by ANSYS WORKBECH finite element simulation. Wind tunnel experiments simulate the sensor’s force response when the inclined cable is subjected to different wind speeds, and the detection accuracy reaches 96.17%. The results show the sensor developed in this paper can realize accurate wind pressure detection of the inclined cable. This study provides a new method for wind pressure detection and health inspection of diagonal cables.
{"title":"Design of flexible sensor for wind pressure monitoring of stay cables","authors":"Xiaoming Wang, Zhilong Guo, Yifeng Huang, Longbo Xiong, Daojin Yao, Wentao Dong","doi":"10.1088/1361-6501/ad1cc3","DOIUrl":"https://doi.org/10.1088/1361-6501/ad1cc3","url":null,"abstract":"\u0000 Strong winds can make a bridge’s cable-stayed cables produce violent vibrations, leading to fatigue of the cable-stayed cables and damaging the cable-stayed bridge’s structure. Accurately and effectively obtaining data on the wind loads applied to the cable-stayed cables is important for assessing the cable-stayed cables’ health. Currently, the more widely used sensing elements include diffuse silicon piezoresistive sensors and strain gauges. However, most of them present such disadvantages as rigidity, difficult to fit the curved surface, high cost and low sensitivity. In this paper, a conductive hydrogel flexible pressure sensor based on TA/CB@PDMS was developed, using carbon black (CB) as the main conductive medium, with good electrical conductivity, high sensitivity (0.95 kPa-1) and excellent tensile properties (210% tensile breakage). Meanwhile, a salt permeation method(Soak the sensor in LiBr solution) was used to effectively inhibit the sensor's water from being evaporated and frozen. Its substrate incorporates tannic acid (TA) to increase the sensor’s adhesion so that it adheres well to the diagonal cable’s surface. In this paper, the wind speed variation around the diagonal cable and the force distribution on the surface with considering the fluid-structure coupling effect are analyzed by ANSYS WORKBECH finite element simulation. Wind tunnel experiments simulate the sensor’s force response when the inclined cable is subjected to different wind speeds, and the detection accuracy reaches 96.17%. The results show the sensor developed in this paper can realize accurate wind pressure detection of the inclined cable. This study provides a new method for wind pressure detection and health inspection of diagonal cables.","PeriodicalId":18526,"journal":{"name":"Measurement Science and Technology","volume":"110 12","pages":""},"PeriodicalIF":2.4,"publicationDate":"2024-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139444510","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-09DOI: 10.1088/1361-6501/ad1cc6
Jiadong Meng, Lu Liu, Zhigang Zhao, Cheng Su
To avoid the overuse or underutilization of lithium battery in practical applications, the state of health of lithium battery should be assessed in time to ensure safety and maximize utilization. A health indicator is proposed to show the SOH of lithium battery in this paper. Firstly, the degradation mechanism of lithium battery is described based on the working principle and the aging phenomenon of lithium-iron phosphate battery, and the existing problems in the current research on battery aging phenomenon are expounded. Secondly, to deal with the hidden dangers caused by the aging problem of the battery, the variation of capacity variance is selected to construct the health indicator of lithium battery. Finally, the performance degradation state of lithium battery is divided into four stages according to the changing trend of the constructed indicators in the whole lifetime, which are the formation stage of the first solid electrolyte interface film stage, the normal working stage, the new solid electrolyte interface film stage and the internal resistance rise stage. The battery data set from MIT-Stanford-Toyota Research Center is utilized to verify the proposed method. The results show that the performance of lithium batteries will accelerate degradation with temperature rapidly rising in the new solid electrolyte interface film stage; the internal resistance will increase sharply in the internal resistance rise stage, which can easily lead to accidents such as thermal runaway. The proposed health state assessment method compares with other methods, and it is shown that the proposed method helps to ensure safety and maximize the utilization of lithium batteries.
{"title":"Stages assessment of state of health in a lifetime based on the capacity variance of lithium batteries","authors":"Jiadong Meng, Lu Liu, Zhigang Zhao, Cheng Su","doi":"10.1088/1361-6501/ad1cc6","DOIUrl":"https://doi.org/10.1088/1361-6501/ad1cc6","url":null,"abstract":"\u0000 To avoid the overuse or underutilization of lithium battery in practical applications, the state of health of lithium battery should be assessed in time to ensure safety and maximize utilization. A health indicator is proposed to show the SOH of lithium battery in this paper. Firstly, the degradation mechanism of lithium battery is described based on the working principle and the aging phenomenon of lithium-iron phosphate battery, and the existing problems in the current research on battery aging phenomenon are expounded. Secondly, to deal with the hidden dangers caused by the aging problem of the battery, the variation of capacity variance is selected to construct the health indicator of lithium battery. Finally, the performance degradation state of lithium battery is divided into four stages according to the changing trend of the constructed indicators in the whole lifetime, which are the formation stage of the first solid electrolyte interface film stage, the normal working stage, the new solid electrolyte interface film stage and the internal resistance rise stage. The battery data set from MIT-Stanford-Toyota Research Center is utilized to verify the proposed method. The results show that the performance of lithium batteries will accelerate degradation with temperature rapidly rising in the new solid electrolyte interface film stage; the internal resistance will increase sharply in the internal resistance rise stage, which can easily lead to accidents such as thermal runaway. The proposed health state assessment method compares with other methods, and it is shown that the proposed method helps to ensure safety and maximize the utilization of lithium batteries.","PeriodicalId":18526,"journal":{"name":"Measurement Science and Technology","volume":"108 6","pages":""},"PeriodicalIF":2.4,"publicationDate":"2024-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139444619","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-09DOI: 10.1088/1361-6501/ad1cc2
Guanghui Li, Xinxiu Zhou, Lihong Duan, Zhaoyu Wang, Wei Quan
Owing to the presence of noise and the Doppler background, accurate saturated absorption (SA) peak automatic identification technology poses a significant challenge for laser frequency tuning and locking. To address this issue, a novel peak identification algorithm for the SA spectrum is proposed. First, a Gaussian filter based on a wavelet transform (GCWT) is proposed to mitigate the spectral high-frequency noise. Subsequently, a hybrid method combining a first-order Gaussian wavelet transformation (FGCWT) and adaptive threshold judgment was designed for multi-peak boundary segmentation. Finally, we obtained the target peak and its sweeping voltage based on an adaptive nonlinear fitting algorithm, which was almost unaffected by the peak asymmetry caused by the Doppler background.
{"title":"Wavelet-Based Saturated Absorption Line Detection for Laser Frequency Locking","authors":"Guanghui Li, Xinxiu Zhou, Lihong Duan, Zhaoyu Wang, Wei Quan","doi":"10.1088/1361-6501/ad1cc2","DOIUrl":"https://doi.org/10.1088/1361-6501/ad1cc2","url":null,"abstract":"\u0000 Owing to the presence of noise and the Doppler background, accurate saturated absorption (SA) peak automatic identification technology poses a significant challenge for laser frequency tuning and locking. To address this issue, a novel peak identification algorithm for the SA spectrum is proposed. First, a Gaussian filter based on a wavelet transform (GCWT) is proposed to mitigate the spectral high-frequency noise. Subsequently, a hybrid method combining a first-order Gaussian wavelet transformation (FGCWT) and adaptive threshold judgment was designed for multi-peak boundary segmentation. Finally, we obtained the target peak and its sweeping voltage based on an adaptive nonlinear fitting algorithm, which was almost unaffected by the peak asymmetry caused by the Doppler background.","PeriodicalId":18526,"journal":{"name":"Measurement Science and Technology","volume":"9 18","pages":""},"PeriodicalIF":2.4,"publicationDate":"2024-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139443676","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-09DOI: 10.1088/1361-6501/ad1cc4
Fei Zeng, Xiaotong Ren, Qing Wu
Identification of motor vibration signals is one of the important tasks in motor fault diagnosis and predictive maintenance, and wavelet time-frequency diagram is a commonly used signal analysis method to extract the frequency and time characteristics of signals. In this paper, a method based on LSH-Swin Transformer network is proposed for identifying the wavelet time-frequency diagrams of motor vibration signals to analyze the fault types.The traditional Swin Transformer model is difficult to improve the accuracy due to the smoothing of the attention distribution when dealing with data with sparse features, while the method proposed in this paper reduces the smoothing of the computed attention and enables the network to learn the key features better by introducing locally-sensitive hash attention in the network model, dividing the sequences in the input attention into multiple hash buckets, calculating the attention weights of only some of the vectors with a high degree of hash similarity, and by sampling discrete samples with the use of the Gumbel Softmax. The experimental results show that the method proposed in this paper has better recognition accuracy and higher computational efficiency compared with the traditional network when processing wavelet time-frequency maps of motor vibration signals, and its validation accuracy reaches 99.7%, the number of parameters also has a decrease of about 10%, and the training network to reach converged epochs is also faster. The method in this paper can provide an effective solution for the analysis and processing of motor vibration signals, and has certain application value in practical engineering.
{"title":"A Fault Diagnosis Method for Motor Vibration Signals Incorporating Swin Transformer with Locally Sensitive Hash Attention","authors":"Fei Zeng, Xiaotong Ren, Qing Wu","doi":"10.1088/1361-6501/ad1cc4","DOIUrl":"https://doi.org/10.1088/1361-6501/ad1cc4","url":null,"abstract":"\u0000 Identification of motor vibration signals is one of the important tasks in motor fault diagnosis and predictive maintenance, and wavelet time-frequency diagram is a commonly used signal analysis method to extract the frequency and time characteristics of signals. In this paper, a method based on LSH-Swin Transformer network is proposed for identifying the wavelet time-frequency diagrams of motor vibration signals to analyze the fault types.The traditional Swin Transformer model is difficult to improve the accuracy due to the smoothing of the attention distribution when dealing with data with sparse features, while the method proposed in this paper reduces the smoothing of the computed attention and enables the network to learn the key features better by introducing locally-sensitive hash attention in the network model, dividing the sequences in the input attention into multiple hash buckets, calculating the attention weights of only some of the vectors with a high degree of hash similarity, and by sampling discrete samples with the use of the Gumbel Softmax. The experimental results show that the method proposed in this paper has better recognition accuracy and higher computational efficiency compared with the traditional network when processing wavelet time-frequency maps of motor vibration signals, and its validation accuracy reaches 99.7%, the number of parameters also has a decrease of about 10%, and the training network to reach converged epochs is also faster. The method in this paper can provide an effective solution for the analysis and processing of motor vibration signals, and has certain application value in practical engineering.","PeriodicalId":18526,"journal":{"name":"Measurement Science and Technology","volume":"25 3","pages":""},"PeriodicalIF":2.4,"publicationDate":"2024-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139443113","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-09DOI: 10.1088/1361-6501/ad1cc5
Zichao Ding, Aimin Wang, Maosen Gao, Jiazhe Li
A pivotal technology for autonomous robot grasping is efficient and accurate grasp pose detection, which enables robotic arms to grasp objects in cluttered environments without human intervention. However, most existing methods rely on PointNet or CNN as backbones for grasp pose prediction, which may lead to unnecessary computational overhead on invalid grasp points or background information. Consequently, performing efficient grasp pose detection for graspable points in complex scenes becomes a challenge. In this paper, we propose FastGNet, an end-to-end model that combines multiple attention mechanisms with the Transformer architecture to generate 6-DOF grasp poses efficiently. Our approach involves a novel sparse point cloud voxelization technique, preserving the complete mapping between points and voxels while generating positional embeddings for the Transformer network. By integrating unsupervised and supervised attention mechanisms into the grasp model, our method significantly improves the performance of focusing on graspable target points in complex scenes. The effectiveness of FastGNet is validated on the large-scale GraspNet-1Billion dataset. Our approach outperforms previous methods and achieves relatively fast inference times, highlighting its potential to advance autonomous robot grasping capabilities.
{"title":"FastGNet: An Efficient 6-DOF Grasp Detection Method with Multi-Attention Mechanisms and Point Transformer Network","authors":"Zichao Ding, Aimin Wang, Maosen Gao, Jiazhe Li","doi":"10.1088/1361-6501/ad1cc5","DOIUrl":"https://doi.org/10.1088/1361-6501/ad1cc5","url":null,"abstract":"\u0000 A pivotal technology for autonomous robot grasping is efficient and accurate grasp pose detection, which enables robotic arms to grasp objects in cluttered environments without human intervention. However, most existing methods rely on PointNet or CNN as backbones for grasp pose prediction, which may lead to unnecessary computational overhead on invalid grasp points or background information. Consequently, performing efficient grasp pose detection for graspable points in complex scenes becomes a challenge. In this paper, we propose FastGNet, an end-to-end model that combines multiple attention mechanisms with the Transformer architecture to generate 6-DOF grasp poses efficiently. Our approach involves a novel sparse point cloud voxelization technique, preserving the complete mapping between points and voxels while generating positional embeddings for the Transformer network. By integrating unsupervised and supervised attention mechanisms into the grasp model, our method significantly improves the performance of focusing on graspable target points in complex scenes. The effectiveness of FastGNet is validated on the large-scale GraspNet-1Billion dataset. Our approach outperforms previous methods and achieves relatively fast inference times, highlighting its potential to advance autonomous robot grasping capabilities.","PeriodicalId":18526,"journal":{"name":"Measurement Science and Technology","volume":"3 4","pages":""},"PeriodicalIF":2.4,"publicationDate":"2024-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139443606","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：