In the food industry, confusion stemming from expiration and date labels contributes to unnecessary food waste, underscoring the growing need for innovative food freshness sensors. This study presents a novel, cost-effective, and environmentally friendly dual-mode ammonia sensor tailored for real-time quality monitoring of protein-rich food products. Utilizing naturally occurring anthocyanin extracted from Clitoria ternatea (CT) and reinforced with polyvinyl alcohol (PVA) in a paper-based colorimetric system, the sensor demonstrates heightened sensitivity to ammonia gas, a key indicator of spoilage in protein-rich foods. Integration of a graphene nanoplatelets (GNPs) layer enables additional resistive gas sensing capabilities. The practicality and versatility of the fabricated sensor are enhanced by integrating near-field communication (NFC) technology, which facilitates batteryless and wireless sensing response transmission. The fabrication process of the sensor involves a straightforward, low-temperature solution route utilizing dip-coating and brush-coating methods. The incorporation of PVA significantly amplifies the colorimetric response, evidenced by a 44% increase in total color change compared to non-PVA reinforced sensors. This augmentation results in a more pronounced color change, which is readily discernible to the naked eye. The developed dual-mode sensor, equipped with NFC, is successfully applied to monitor shrimp freshness, demonstrating distinct color changes and NFC tag readability in response to ammonia release during spoilage. With its attributes of cost-effectiveness, environmental friendliness, simplicity, and wireless capabilities, this sensor offers a promising solution for widespread adoption in the food industry. This work contributes to advancing sensor technology, providing a versatile tool to ensure the quality and safety of perishable goods.
{"title":"Dual-Mode Batteryless Ammonia Sensor Using Polyvinyl Alcohol-Reinforced Clitoria ternatea Anthocyanin With Graphene Nanoplatelets for Enhanced Food Quality Monitoring","authors":"Thiresamary Kurian;Chun-Hui Tan;Pei-Song Chee;Vinod Ganesan","doi":"10.1109/JSEN.2024.3392954","DOIUrl":"10.1109/JSEN.2024.3392954","url":null,"abstract":"In the food industry, confusion stemming from expiration and date labels contributes to unnecessary food waste, underscoring the growing need for innovative food freshness sensors. This study presents a novel, cost-effective, and environmentally friendly dual-mode ammonia sensor tailored for real-time quality monitoring of protein-rich food products. Utilizing naturally occurring anthocyanin extracted from Clitoria ternatea (CT) and reinforced with polyvinyl alcohol (PVA) in a paper-based colorimetric system, the sensor demonstrates heightened sensitivity to ammonia gas, a key indicator of spoilage in protein-rich foods. Integration of a graphene nanoplatelets (GNPs) layer enables additional resistive gas sensing capabilities. The practicality and versatility of the fabricated sensor are enhanced by integrating near-field communication (NFC) technology, which facilitates batteryless and wireless sensing response transmission. The fabrication process of the sensor involves a straightforward, low-temperature solution route utilizing dip-coating and brush-coating methods. The incorporation of PVA significantly amplifies the colorimetric response, evidenced by a 44% increase in total color change compared to non-PVA reinforced sensors. This augmentation results in a more pronounced color change, which is readily discernible to the naked eye. The developed dual-mode sensor, equipped with NFC, is successfully applied to monitor shrimp freshness, demonstrating distinct color changes and NFC tag readability in response to ammonia release during spoilage. With its attributes of cost-effectiveness, environmental friendliness, simplicity, and wireless capabilities, this sensor offers a promising solution for widespread adoption in the food industry. This work contributes to advancing sensor technology, providing a versatile tool to ensure the quality and safety of perishable goods.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140827715","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
To enhance the accuracy of object detection with event-based neuromorphic vision sensors, a novel event-based detector named spatiotemporal aggregation transformer (STAT) is proposed. First, in order to collect sufficient event information to estimate the problem considered, STAT uses a density-based adaptive sampling (DAS) module to sample continuous event stream into multiple groups adaptively. This module can determine the sampling termination condition by quantifying the velocity and size of objects. Second, STAT integrates a sparse event tensor (SET) to establish compatibility between event stream and traditional vision algorithms. SET maps events to a dense representation by end-to-end fitting the optimal mapping function, mitigating the loss of spatiotemporal information within the event stream. Finally, in order to enhance the features of slowly moving objects, a lightweight and efficient triaxial vision transformer (TVT) is designed for modeling global features and integrating historical motion information. Experimental evaluations on two benchmark datasets show that the performance of STAT achieves a mean average precision (mAP) of 68.2% and 49.9% on the Neuromorphic-Caltech101 (N-Caltech101) dataset and the Gen1 dataset, respectively. These results demonstrate that the detection accuracy of STAT outperforms the state-of-the-art methods by 2.0% on the Gen1 dataset. The code of this project is available at https://github.com/TJU-guozhaoxuan/STAT .
{"title":"Spatiotemporal Aggregation Transformer for Object Detection With Neuromorphic Vision Sensors","authors":"Zhaoxuan Guo;Jiandong Gao;Guangyuan Ma;Jiangtao Xu","doi":"10.1109/JSEN.2024.3392973","DOIUrl":"10.1109/JSEN.2024.3392973","url":null,"abstract":"To enhance the accuracy of object detection with event-based neuromorphic vision sensors, a novel event-based detector named spatiotemporal aggregation transformer (STAT) is proposed. First, in order to collect sufficient event information to estimate the problem considered, STAT uses a density-based adaptive sampling (DAS) module to sample continuous event stream into multiple groups adaptively. This module can determine the sampling termination condition by quantifying the velocity and size of objects. Second, STAT integrates a sparse event tensor (SET) to establish compatibility between event stream and traditional vision algorithms. SET maps events to a dense representation by end-to-end fitting the optimal mapping function, mitigating the loss of spatiotemporal information within the event stream. Finally, in order to enhance the features of slowly moving objects, a lightweight and efficient triaxial vision transformer (TVT) is designed for modeling global features and integrating historical motion information. Experimental evaluations on two benchmark datasets show that the performance of STAT achieves a mean average precision (mAP) of 68.2% and 49.9% on the Neuromorphic-Caltech101 (N-Caltech101) dataset and the Gen1 dataset, respectively. These results demonstrate that the detection accuracy of STAT outperforms the state-of-the-art methods by 2.0% on the Gen1 dataset. The code of this project is available at \u0000<uri>https://github.com/TJU-guozhaoxuan/STAT</uri>\u0000.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140837451","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-29DOI: 10.1109/JSEN.2024.3390231
Lin Zhang;Menglong Wu;Xichang Cai;Yundong Li;Wenkai Liu
Acoustic scene classification (ASC) and sound event detection (SED) are two research directions in the field of acoustics, and they are closely related. Previous works have adopted a joint analysis method for acoustic scenes and events based on multitask learning (MTL). However, the traditional MTL models are often sensitive to the proportion of dataset partitioning, and multitask analysis is not as effective as single-task analysis. In addition, the performance of traditional MTL models is highly dependent on the weights of the loss function, and manually adjusting weights is costly. In response to these issues, we suggest improvements in both the model and loss function formulation, to utilize additional sound event information to assist in improving the performance of ASC. First, the multigate mixture-of-experts (MMoEs) model is introduced into the field of acoustics. Experimental results obtained using TUT Sound Events 2016/2017 and TUT Acoustic Scenes 2016 datasets indicate that the mixture-of-experts model achieves an optimal performance of 98.74% in terms of $F1$ -score, which is 1.43% higher than traditional MTL models; second, we improve the mixture-of-experts model and propose the Cross_MMoE model, which increases the information interaction between different task branches, and the $F1$ -score is further improved to 99.04%; finally, to address the issue of imbalanced sample categories in the dataset, we evaluate the class balanced loss formulation to replace the traditional multitask loss function. The performance of the traditional multitask model, MMoE model, and Cross_MMoE model has been improved, and more specifically, the $F1$ -score of the Cross_MMoE model has increased to 99.31%.
{"title":"Joint Analysis of Acoustic Scenes and Sound Events in Multitask Learning Based on Cross_MMoE Model and Class-Balanced Loss","authors":"Lin Zhang;Menglong Wu;Xichang Cai;Yundong Li;Wenkai Liu","doi":"10.1109/JSEN.2024.3390231","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3390231","url":null,"abstract":"Acoustic scene classification (ASC) and sound event detection (SED) are two research directions in the field of acoustics, and they are closely related. Previous works have adopted a joint analysis method for acoustic scenes and events based on multitask learning (MTL). However, the traditional MTL models are often sensitive to the proportion of dataset partitioning, and multitask analysis is not as effective as single-task analysis. In addition, the performance of traditional MTL models is highly dependent on the weights of the loss function, and manually adjusting weights is costly. In response to these issues, we suggest improvements in both the model and loss function formulation, to utilize additional sound event information to assist in improving the performance of ASC. First, the multigate mixture-of-experts (MMoEs) model is introduced into the field of acoustics. Experimental results obtained using TUT Sound Events 2016/2017 and TUT Acoustic Scenes 2016 datasets indicate that the mixture-of-experts model achieves an optimal performance of 98.74% in terms of \u0000<inline-formula> <tex-math>$F1$ </tex-math></inline-formula>\u0000-score, which is 1.43% higher than traditional MTL models; second, we improve the mixture-of-experts model and propose the Cross_MMoE model, which increases the information interaction between different task branches, and the \u0000<inline-formula> <tex-math>$F1$ </tex-math></inline-formula>\u0000-score is further improved to 99.04%; finally, to address the issue of imbalanced sample categories in the dataset, we evaluate the class balanced loss formulation to replace the traditional multitask loss function. The performance of the traditional multitask model, MMoE model, and Cross_MMoE model has been improved, and more specifically, the \u0000<inline-formula> <tex-math>$F1$ </tex-math></inline-formula>\u0000-score of the Cross_MMoE model has increased to 99.31%.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141245126","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-29DOI: 10.1109/JSEN.2024.3392304
Chengcheng Luo;Lin Chen
A spoof surface plasmon polaritons (SSPPs) sensor for enhanced terahertz fingerprint detection is numerically and experimentally demonstrated. As lactose thin film is deposited from a prism surface to spoof plasmon surface, the physical principle of sensing changes from total internal reflection (TIR) to absorption-induced transparency (AIT), which is induced by the coupling between a narrowband molecule vibrational resonance of lactose and broadband spoof plasmonic resonance. The coupling strength of the coupled plasmon-lactose system increases with an increasing thickness of lactose (within the enhanced electrical field decay range). The sensitivity based on the AIT effect of fingerprint detection is four times higher than the TIR effect, where the differential reflectance $Delta {R}$ is obtained from 22.9% to 45.9% compared to that from 15.3% to 17.3%. $8~mu $ m lactose film was successfully detected and displayed a clear mode splitting in the experiment. It is noted that the resonant frequency can be tuned from 0.47 to 0.59 THz by easily adjusting the coupling air gap, which can compensate for random fabrication errors. The coupling between spoof plasmonic resonance and molecular vibrational modes provides a new way for studying light-matter interactions and flexible fingerprint detection of different molecules with high sensitivity in the terahertz region.
一种用于增强太赫兹指纹检测的欺骗性表面等离子体极化子（SSPPs）传感器得到了数值和实验验证。当乳糖薄膜从棱镜表面沉积到欺骗性质子表面时，传感的物理原理从全内反射（TIR）转变为吸收诱导透明（AIT），这是由乳糖的窄带分子振动共振和宽带欺骗性质子共振之间的耦合引起的。耦合质子-乳糖系统的耦合强度随着乳糖厚度的增加而增加（在增强电场衰减范围内）。基于 AIT 效应的指纹检测灵敏度比 TIR 效应高四倍，其中差分反射率 $Delta {R}$ 从 22.9% 到 45.9%，而 TIR 效应从 15.3% 到 17.3%。 8~mu $ m 乳糖薄膜被成功检测到，并在实验中显示出明显的模式分裂。实验表明，通过调节耦合气隙，共振频率可在 0.47 至 0.59 THz 之间调节，从而弥补了随机制造误差。欺骗性等离子体共振与分子振动模式之间的耦合为研究光-物质相互作用提供了一种新方法，并能在太赫兹区域以高灵敏度灵活地检测不同分子的指纹。
{"title":"Highly Sensitive and Tunable Absorption-Induced Transparency for Terahertz Fingerprint Sensing With Spoof Surface Plasmon Polaritons","authors":"Chengcheng Luo;Lin Chen","doi":"10.1109/JSEN.2024.3392304","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3392304","url":null,"abstract":"A spoof surface plasmon polaritons (SSPPs) sensor for enhanced terahertz fingerprint detection is numerically and experimentally demonstrated. As lactose thin film is deposited from a prism surface to spoof plasmon surface, the physical principle of sensing changes from total internal reflection (TIR) to absorption-induced transparency (AIT), which is induced by the coupling between a narrowband molecule vibrational resonance of lactose and broadband spoof plasmonic resonance. The coupling strength of the coupled plasmon-lactose system increases with an increasing thickness of lactose (within the enhanced electrical field decay range). The sensitivity based on the AIT effect of fingerprint detection is four times higher than the TIR effect, where the differential reflectance \u0000<inline-formula> <tex-math>$Delta {R}$ </tex-math></inline-formula>\u0000 is obtained from 22.9% to 45.9% compared to that from 15.3% to 17.3%. \u0000<inline-formula> <tex-math>$8~mu $ </tex-math></inline-formula>\u0000m lactose film was successfully detected and displayed a clear mode splitting in the experiment. It is noted that the resonant frequency can be tuned from 0.47 to 0.59 THz by easily adjusting the coupling air gap, which can compensate for random fabrication errors. The coupling between spoof plasmonic resonance and molecular vibrational modes provides a new way for studying light-matter interactions and flexible fingerprint detection of different molecules with high sensitivity in the terahertz region.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141245128","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-29DOI: 10.1109/JSEN.2024.3392837
Zihan Wang;Kai-Chong Lei;Huaze Tang;Yang Luo;Hongfa Zhao;Peisheng He;Wenbo Ding;Liwei Lin
Vibration perception can help robots recognize their dynamic states to explore the surrounding environment. However, the intrinsic stretchability of soft robots poses challenges to integrating vibration sensors. This study introduces an innovative stretchable electronic skin (e-skin) that facilitates vibration proprioception in soft robots. Constructed with a thickness of approximately 0.1 mm, this ultrathin e-skin is produced using a screen-printing technique with liquid metal particles (LMPs), incorporating a kirigami design for seamless integration. The e-skin works by the triboelectric nanogenerator-based sensing mechanism, which transduces mechanical vibration into an electrical signal without an external power source. By analyzing the vibration signals generated by the dynamic motions of soft robots, the e-skin shows a wide range of applications. From the vibration signal of the soft robotic finger’s sliding motion, 17 different textures can be distinguished with 99% accuracy. Furthermore, analysis of the vibration signal from a soft robotic gripper’s swinging motion enables the estimation of both the type and weight of grains inside the container it grips, achieving accuracies of 97.7% and 95.3%, respectively. As such, this work presents a new approach to realizing the vibration proprioception of soft robots, thereby broadening the applications of dynamic proprioception in soft robotics.
{"title":"Stretchable Liquid Metal E-Skin for Soft Robot Proprioceptive Vibration Sensing","authors":"Zihan Wang;Kai-Chong Lei;Huaze Tang;Yang Luo;Hongfa Zhao;Peisheng He;Wenbo Ding;Liwei Lin","doi":"10.1109/JSEN.2024.3392837","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3392837","url":null,"abstract":"Vibration perception can help robots recognize their dynamic states to explore the surrounding environment. However, the intrinsic stretchability of soft robots poses challenges to integrating vibration sensors. This study introduces an innovative stretchable electronic skin (e-skin) that facilitates vibration proprioception in soft robots. Constructed with a thickness of approximately 0.1 mm, this ultrathin e-skin is produced using a screen-printing technique with liquid metal particles (LMPs), incorporating a kirigami design for seamless integration. The e-skin works by the triboelectric nanogenerator-based sensing mechanism, which transduces mechanical vibration into an electrical signal without an external power source. By analyzing the vibration signals generated by the dynamic motions of soft robots, the e-skin shows a wide range of applications. From the vibration signal of the soft robotic finger’s sliding motion, 17 different textures can be distinguished with 99% accuracy. Furthermore, analysis of the vibration signal from a soft robotic gripper’s swinging motion enables the estimation of both the type and weight of grains inside the container it grips, achieving accuracies of 97.7% and 95.3%, respectively. As such, this work presents a new approach to realizing the vibration proprioception of soft robots, thereby broadening the applications of dynamic proprioception in soft robotics.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141245065","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In order to deal with the issue of low sensitivity to biomolecules concentrations in traditional optical fiber surface plasmon resonance (SPR) biosensors, a fiber SPR deoxyribonucleic acid (DNA) biosensor was developed, which is based on an amino-functionalized Zr-based metal–organic framework (MOF) (UIO-66-NH2)-modified heterogeneous core structure. Porous material UIO-66-NH2 facilitates electron transfer and amplifies the SPR effect, resulting in a heightened sensitivity for the detection of DNA hybridization. This can be attributed to the material’s advantageous characteristics, including a substantial specific surface area, an excellent water stability, and a pronounced affinity for probe-DNA (pDNA). The DNA biosensor presented in this study demonstrates enhanced sensitivity as a result of the covalent attachment of MOFs to pDNA, hence increasing the number of binding sites available for target DNA (tDNA). The optical fiber SPR biosensor, which has been modified with UIO-66-NH2, exhibits remarkable capabilities in detecting DNA concentrations ranging from 1 pM to $1 ~mu text{M}$ . It demonstrates a sensitivity of −7.72 nm/log ( $mu text{M}$ ) and achieves a detection limit of 3 pM. In addition, the DNA biosensor has exceptional performance in terms of specificity and durability, rendering it a highly favorable option for utilization in several fields, such as drug delivery, protein identification, and environmental surveillance.
{"title":"Metal–Organic Frameworks Modified Optical Fiber SPR Biosensor for DNA Detection","authors":"Lingling Li;Shen Yu;Chonglu Jing;Lihua Chen;Ai Zhou","doi":"10.1109/JSEN.2024.3350676","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3350676","url":null,"abstract":"In order to deal with the issue of low sensitivity to biomolecules concentrations in traditional optical fiber surface plasmon resonance (SPR) biosensors, a fiber SPR deoxyribonucleic acid (DNA) biosensor was developed, which is based on an amino-functionalized Zr-based metal–organic framework (MOF) (UIO-66-NH2)-modified heterogeneous core structure. Porous material UIO-66-NH2 facilitates electron transfer and amplifies the SPR effect, resulting in a heightened sensitivity for the detection of DNA hybridization. This can be attributed to the material’s advantageous characteristics, including a substantial specific surface area, an excellent water stability, and a pronounced affinity for probe-DNA (pDNA). The DNA biosensor presented in this study demonstrates enhanced sensitivity as a result of the covalent attachment of MOFs to pDNA, hence increasing the number of binding sites available for target DNA (tDNA). The optical fiber SPR biosensor, which has been modified with UIO-66-NH2, exhibits remarkable capabilities in detecting DNA concentrations ranging from 1 pM to \u0000<inline-formula> <tex-math>$1 ~mu text{M}$ </tex-math></inline-formula>\u0000. It demonstrates a sensitivity of −7.72 nm/log (\u0000<inline-formula> <tex-math>$mu text{M}$ </tex-math></inline-formula>\u0000) and achieves a detection limit of 3 pM. In addition, the DNA biosensor has exceptional performance in terms of specificity and durability, rendering it a highly favorable option for utilization in several fields, such as drug delivery, protein identification, and environmental surveillance.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141245110","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-29DOI: 10.1109/JSEN.2024.3392755
Juncai Song;Long Qian;Xianhong Wu;Jing Wu;Siliang Lu;Xiaoxian Wang
To detect the eccentricity fault of dual-sided permanent magnet synchronous linear motor (DPMSLM) and ensure the stable operation of the equipment, a new method based on multiview of mystery curve transformation (MCT) and deep feature extraction is proposed to detect complex eccentricity faults of DPMSLM in this work. First, finite element analysis (FEA) calculation models of DPMSLM under different static eccentricity and dynamic eccentricity fault conditions are established to extract the external magnetic leakage signal (EMLS) as the efficient fault diagnostic signal. Second, an MCT signal processing method is proposed to convert a 1-D EMLS into a 3-D curve and obtain 2-D projections of multiple views (top, front, and side). This method achieves eccentricity fault signal visual display in a 2-D multiview fusion image and realizes complementary enhancement of fault characteristics. Thereafter, a novel classification deep learning framework, named SA-ConvNeXt, is proposed to conduct deep fault feature extraction and realize eccentricity faults’ accurate classification in fault types and severity levels. The diagnostic accuracy of SA-ConvNeXt is as high as 99.5%, which is better than those of comparison models, such as CNN, ResNet-34, ShuffleNet, and ConvNeXt. Finally, tunnel magnetoresistance (TMR) sensor circuit hardware is integrated designation with a motor mover module to realize EMLS data noninvasive online measurement, and the DPMSLM experimental platform under several eccentricity faults is built to verify the superiority and robustness of the proposed method.
{"title":"DPMSLM Eccentricity Fault Detection Based on Multiview of Mystery Curve Transformation and Deep Feature Extraction","authors":"Juncai Song;Long Qian;Xianhong Wu;Jing Wu;Siliang Lu;Xiaoxian Wang","doi":"10.1109/JSEN.2024.3392755","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3392755","url":null,"abstract":"To detect the eccentricity fault of dual-sided permanent magnet synchronous linear motor (DPMSLM) and ensure the stable operation of the equipment, a new method based on multiview of mystery curve transformation (MCT) and deep feature extraction is proposed to detect complex eccentricity faults of DPMSLM in this work. First, finite element analysis (FEA) calculation models of DPMSLM under different static eccentricity and dynamic eccentricity fault conditions are established to extract the external magnetic leakage signal (EMLS) as the efficient fault diagnostic signal. Second, an MCT signal processing method is proposed to convert a 1-D EMLS into a 3-D curve and obtain 2-D projections of multiple views (top, front, and side). This method achieves eccentricity fault signal visual display in a 2-D multiview fusion image and realizes complementary enhancement of fault characteristics. Thereafter, a novel classification deep learning framework, named SA-ConvNeXt, is proposed to conduct deep fault feature extraction and realize eccentricity faults’ accurate classification in fault types and severity levels. The diagnostic accuracy of SA-ConvNeXt is as high as 99.5%, which is better than those of comparison models, such as CNN, ResNet-34, ShuffleNet, and ConvNeXt. Finally, tunnel magnetoresistance (TMR) sensor circuit hardware is integrated designation with a motor mover module to realize EMLS data noninvasive online measurement, and the DPMSLM experimental platform under several eccentricity faults is built to verify the superiority and robustness of the proposed method.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141245232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-29DOI: 10.1109/JSEN.2024.3390676
Shuang Zhao;Yuanxi Yang;Zhenjie Wang;Shuqiang Xue
Global navigation satellite system-acoustic (GNSS-A) combined underwater positioning technique is widely applied in seafloor displacement monitoring and offshore exploration. The conventional GNSS-A positioning strategy is under the assumption of equal-precision sea-surface transducer’s positions determined by GNSS positioning, which weakens the positioning accuracy of single seafloor transponder-equipped station. In this article, the extended joint adjustment (JA) of the measurements of the sea-surface transducers to seafloor transponders and ranging measurements of the transponder to transponder is proposed. First, we refine the transducer-to-transponder timing observation equation system by acoustic ray-tracing strategy to reduce the sound-speed-related errors. Second, we establish the mathematical model for extended JA with interstation ranging measurements for seafloor geodetic network (SGN) positioning. Finally, the efficacy of the proposed method is demonstrated both in simulations and in real measurement datasets. The experimental results show that the proposed method outperforms the traditional positioning methods, especially in the horizontal components.
全球导航卫星系统-声学（GNSS-A）组合水下定位技术广泛应用于海底位移监测和近海勘探。传统的 GNSS-A 定位策略是在 GNSS 定位确定等精度海面换能器位置的假设下进行的，这削弱了单个海底换能器站的定位精度。本文提出了海面换能器对海底转发器的测量和转发器对换能器的测距测量的扩展联合调整（JA）。首先，我们通过声学射线追踪策略完善了换能器到转发器的定时观测方程系统，以减少与声速相关的误差。其次，我们建立了海底大地测量网络（SGN）定位的站间测距扩展 JA 数学模型。最后，在模拟和实际测量数据集中证明了所提方法的有效性。实验结果表明，所提出的方法优于传统的定位方法，尤其是在水平分量方面。
{"title":"Extended Joint Adjustment of the Transducer and Seafloor Transponder for GNSS-Acoustic Seafloor Geodetic Network Positioning With Interstation Ranging Information","authors":"Shuang Zhao;Yuanxi Yang;Zhenjie Wang;Shuqiang Xue","doi":"10.1109/JSEN.2024.3390676","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3390676","url":null,"abstract":"Global navigation satellite system-acoustic (GNSS-A) combined underwater positioning technique is widely applied in seafloor displacement monitoring and offshore exploration. The conventional GNSS-A positioning strategy is under the assumption of equal-precision sea-surface transducer’s positions determined by GNSS positioning, which weakens the positioning accuracy of single seafloor transponder-equipped station. In this article, the extended joint adjustment (JA) of the measurements of the sea-surface transducers to seafloor transponders and ranging measurements of the transponder to transponder is proposed. First, we refine the transducer-to-transponder timing observation equation system by acoustic ray-tracing strategy to reduce the sound-speed-related errors. Second, we establish the mathematical model for extended JA with interstation ranging measurements for seafloor geodetic network (SGN) positioning. Finally, the efficacy of the proposed method is demonstrated both in simulations and in real measurement datasets. The experimental results show that the proposed method outperforms the traditional positioning methods, especially in the horizontal components.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141326345","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-29DOI: 10.1109/JSEN.2024.3392588
Shuting Tang;Chaoqun Ma;Debo Wang
In order to achieve highly efficient energy harvesting in low-frequency vibration environment, a combined cantilever piezoelectric energy harvester (CCPEH) is studied in this work. The design uses a circular cantilever beam to reduce the resonant frequency and achieve multidirectional energy harvesting. The trapezoidal cantilever beam and arc spiral cantilever beam are coupled to each other to improve the energy harvester efficiency per unit volume. The relationship of the arc with a radial pitch of circular cantilever beam is studied. The piezoelectric energy harvesting system with different combined cantilever beam structures is fabricated, and the output performance of those energy harvesting systems is measured and compared. The measured results show that the CCPEH with an arc of $4pi $ and a radial pitch of 8 mm can achieve multidirectional harvesting and improve the energy harvesting efficiency per unit volume with a resonant frequency of 39 Hz, an output voltage of 46.4 V, and an output power of $2337 , mu $ W. This structure of piezoelectric energy harvester (PEH) can be effectively applied in wireless sensors and microelectronic devices.
为了在低频振动环境中实现高效能量收集，本研究对一种组合式悬臂压电能量收集器（CCPEH）进行了研究。该设计使用圆形悬臂梁来降低谐振频率，实现多向能量采集。梯形悬臂梁和弧形螺旋悬臂梁相互耦合，以提高单位体积能量收集器的效率。研究了圆弧与圆形悬臂梁径向间距的关系。制作了具有不同组合悬臂梁结构的压电能量收集系统，并对这些能量收集系统的输出性能进行了测量和比较。测量结果表明，弧度为 $4pi $、径向间距为 8 mm 的 CCPEH 可以实现多向能量收集，提高单位体积的能量收集效率，谐振频率为 39 Hz，输出电压为 46.4 V，输出功率为 2337 , mu $ W。
{"title":"A Low-Frequency High-Performance Harvesting System With Combined Cantilever Beam","authors":"Shuting Tang;Chaoqun Ma;Debo Wang","doi":"10.1109/JSEN.2024.3392588","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3392588","url":null,"abstract":"In order to achieve highly efficient energy harvesting in low-frequency vibration environment, a combined cantilever piezoelectric energy harvester (CCPEH) is studied in this work. The design uses a circular cantilever beam to reduce the resonant frequency and achieve multidirectional energy harvesting. The trapezoidal cantilever beam and arc spiral cantilever beam are coupled to each other to improve the energy harvester efficiency per unit volume. The relationship of the arc with a radial pitch of circular cantilever beam is studied. The piezoelectric energy harvesting system with different combined cantilever beam structures is fabricated, and the output performance of those energy harvesting systems is measured and compared. The measured results show that the CCPEH with an arc of \u0000<inline-formula> <tex-math>$4pi $ </tex-math></inline-formula>\u0000 and a radial pitch of 8 mm can achieve multidirectional harvesting and improve the energy harvesting efficiency per unit volume with a resonant frequency of 39 Hz, an output voltage of 46.4 V, and an output power of \u0000<inline-formula> <tex-math>$2337 , mu $ </tex-math></inline-formula>\u0000W. This structure of piezoelectric energy harvester (PEH) can be effectively applied in wireless sensors and microelectronic devices.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141326297","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rotating machinery is an important component of modern electromechanical systems and its failure can result in significant economic losses. However, existing deep learning methods only consider the features within each sample, not the neighborhood relationships among samples; this results in poor performance when few labeled samples are available. To overcome this problem, we developed a meta-adaptive graph convolutional network (MAGCNet) to uncover the neighborhood relationships among samples and construct better features for the fault diagnosis of rotating machines when labeled samples are scarce. The wavelet-packet coefficient matrices of raw vibration data are extracted and defined as node features in a graph. To enhance the correlation properties of the few samples, an adjacency matrix is constructed by measuring the Euclidean distance between time- and frequency-domain characteristics and adding prior knowledge. The graph is divided into a series of subgraphs that are trained to optimize the initialization parameters of the adaptive graph convolution layers. The effectiveness of the proposed method was verified using datasets from the drivetrain diagnostics simulator (DDS) test rig and wind-turbine gearboxes.
{"title":"Meta-Adaptive Graph Convolutional Networks With Few Samples for the Fault Diagnosis of Rotating Machinery","authors":"Xiaoxia Yu;Zhigang Zhang;Baoping Tang;Minghang Zhao","doi":"10.1109/JSEN.2024.3392372","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3392372","url":null,"abstract":"Rotating machinery is an important component of modern electromechanical systems and its failure can result in significant economic losses. However, existing deep learning methods only consider the features within each sample, not the neighborhood relationships among samples; this results in poor performance when few labeled samples are available. To overcome this problem, we developed a meta-adaptive graph convolutional network (MAGCNet) to uncover the neighborhood relationships among samples and construct better features for the fault diagnosis of rotating machines when labeled samples are scarce. The wavelet-packet coefficient matrices of raw vibration data are extracted and defined as node features in a graph. To enhance the correlation properties of the few samples, an adjacency matrix is constructed by measuring the Euclidean distance between time- and frequency-domain characteristics and adding prior knowledge. The graph is divided into a series of subgraphs that are trained to optimize the initialization parameters of the adaptive graph convolution layers. The effectiveness of the proposed method was verified using datasets from the drivetrain diagnostics simulator (DDS) test rig and wind-turbine gearboxes.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141328273","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","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：