Pub Date : 2024-10-04DOI: 10.1109/JSEN.2024.3470250
Jianping Yu;Shengjie Yao;Xiaoliang Jiang;Zhehe Yao
Soft capacitive sensors, mainly attributing to their structural simplicity, fast response, and high spatial resolution, have drawn great attention for possible use in many kinds of human–machine interactions. Nevertheless, mechanical coupling and pressure-induced continuous deformation between the contacted areas and other adjacent units would bring unexpected crosstalk and thus vague spatial resolution during distributed pressure recognition. Herein, a stretchable $16times 16$ capacitive tactile sensor array of minimum proximate crosstalk for distributed pressure recognition is proposed. Benefiting from the introduction of serpentine island bridge structure, the sensor array has displayed excellent stretchability (over 30%) as well as low crosstalk between adjacent units (8.53%) in a wide measuring range (265 kPa) and still maintaining high sensitivity up to 5.40 kPa $^{-{1}}$ , low limit of detection (2 Pa), and fast response time (44 ms) as well as long-term stable working durability for over 1000 cycles. An improved bilinear convolutional neural network (BCNN) integrated with deep residual shrinkage network (DRSN) is proposed to actually heighten the feature extraction capability and thus precise distributed pressure recognition. Cataloged pressure images of capital letter shapes from A to Z in different letter patterns, random angles, and uncertain positions are collected to validate the proposed models. The test results reveal that the recognition accuracy is up to 97.70% in this work and thus provide a more detailed pressure distribution in activated sensing areas.
软电容传感器因其结构简单、响应速度快、空间分辨率高等优点，在多种人机交互中的应用备受关注。然而，在分布式压力识别过程中，接触区域和其他相邻单元之间的机械耦合和压力引起的持续变形会带来意想不到的串扰，从而导致空间分辨率模糊不清。在此，我们提出了一种用于分布式压力识别的、可拉伸的、近距离串扰最小的 $16times 16$ 电容式触觉传感器阵列。得益于蛇形岛桥结构的引入，该传感器阵列在较宽的测量范围（265 kPa）内表现出了卓越的可拉伸性（超过 30%）和相邻单元间的低串扰（8.53%），并且仍然保持了高达 5.40 kPa $^{-{1}}$的高灵敏度、低检测限（2 Pa）、快速响应时间（44 ms）以及超过 1000 次循环的长期稳定工作耐久性。改进的双线性卷积神经网络（BCNN）与深度残差收缩网络（DRSN）相结合，切实提高了特征提取能力，从而实现了精确的分布式压力识别。为了验证所提出的模型，收集了从 A 到 Z 不同字母形状、随机角度和不确定位置的大写字母压力图像。测试结果表明，这项工作的识别准确率高达 97.70%，从而为活化传感区域提供了更详细的压力分布。
{"title":"Stretchable Capacitive Tactile Sensor Array for Accurate Distributed Pressure Recognition","authors":"Jianping Yu;Shengjie Yao;Xiaoliang Jiang;Zhehe Yao","doi":"10.1109/JSEN.2024.3470250","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3470250","url":null,"abstract":"Soft capacitive sensors, mainly attributing to their structural simplicity, fast response, and high spatial resolution, have drawn great attention for possible use in many kinds of human–machine interactions. Nevertheless, mechanical coupling and pressure-induced continuous deformation between the contacted areas and other adjacent units would bring unexpected crosstalk and thus vague spatial resolution during distributed pressure recognition. Herein, a stretchable \u0000<inline-formula> <tex-math>$16times 16$ </tex-math></inline-formula>\u0000 capacitive tactile sensor array of minimum proximate crosstalk for distributed pressure recognition is proposed. Benefiting from the introduction of serpentine island bridge structure, the sensor array has displayed excellent stretchability (over 30%) as well as low crosstalk between adjacent units (8.53%) in a wide measuring range (265 kPa) and still maintaining high sensitivity up to 5.40 kPa\u0000<inline-formula> <tex-math>$^{-{1}}$ </tex-math></inline-formula>\u0000, low limit of detection (2 Pa), and fast response time (44 ms) as well as long-term stable working durability for over 1000 cycles. An improved bilinear convolutional neural network (BCNN) integrated with deep residual shrinkage network (DRSN) is proposed to actually heighten the feature extraction capability and thus precise distributed pressure recognition. Cataloged pressure images of capital letter shapes from A to Z in different letter patterns, random angles, and uncertain positions are collected to validate the proposed models. The test results reveal that the recognition accuracy is up to 97.70% in this work and thus provide a more detailed pressure distribution in activated sensing areas.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"24 22","pages":"37836-37845"},"PeriodicalIF":4.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142636561","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 : 2024-10-04DOI: 10.1109/JSEN.2024.3470231
Junyu Qi;Zhuyun Chen;Yuchen Song;Jingyan Xia;Weihua Li
Condition monitoring (CM) has garnered extensive attention in the era of Industry 4.0 and digital manufacturing. It is crucial to monitor the health status of machinery to ensure reliability, safety, production quality, and effectiveness. Advanced predictive maintenance strategies increase equipment availability and effectiveness by accurately predicting failures, thus facilitating maintenance engineering decisions and preventing unplanned machinery breakdowns. In this research, a novel predictive maintenance strategy is proposed by integrating anomaly detection and fault prognostics, two significant challenges in smart maintenance, into one CM system. For anomaly detection, we developed an intelligent methodology based on the skip convolution generative adversarial network (SCGAN). This network combines a convolutional autoencoder (CAE), generative adversarial network (GAN), and skip connections, forming a robust system to construct health indicators (HIs), effectively and efficiently tracking the degradation status of rolling element bearings with fault identified using the $3sigma $ criterion. Validation on real experimental datasets demonstrates that the developed HIs show a stable trend during the healthy stage and a marked increase when deterioration is detected. Subsequently, we employ an advanced graph isomorphic network (GIN) for remaining useful life (RUL) prediction. GIN utilizes graph data and graph convolutions (GCs) to map complex relationships between degradation evolution and RUL. This approach outperforms existing deep learning models, such as convolutional neural networks (CNNs), long short-term memory (LSTM) networks, and CNN-LSTM, providing more accurate RUL prediction.
{"title":"Remaining Useful Life Prediction Combining Advanced Anomaly Detection and Graph Isomorphic Network","authors":"Junyu Qi;Zhuyun Chen;Yuchen Song;Jingyan Xia;Weihua Li","doi":"10.1109/JSEN.2024.3470231","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3470231","url":null,"abstract":"Condition monitoring (CM) has garnered extensive attention in the era of Industry 4.0 and digital manufacturing. It is crucial to monitor the health status of machinery to ensure reliability, safety, production quality, and effectiveness. Advanced predictive maintenance strategies increase equipment availability and effectiveness by accurately predicting failures, thus facilitating maintenance engineering decisions and preventing unplanned machinery breakdowns. In this research, a novel predictive maintenance strategy is proposed by integrating anomaly detection and fault prognostics, two significant challenges in smart maintenance, into one CM system. For anomaly detection, we developed an intelligent methodology based on the skip convolution generative adversarial network (SCGAN). This network combines a convolutional autoencoder (CAE), generative adversarial network (GAN), and skip connections, forming a robust system to construct health indicators (HIs), effectively and efficiently tracking the degradation status of rolling element bearings with fault identified using the \u0000<inline-formula> <tex-math>$3sigma $ </tex-math></inline-formula>\u0000 criterion. Validation on real experimental datasets demonstrates that the developed HIs show a stable trend during the healthy stage and a marked increase when deterioration is detected. Subsequently, we employ an advanced graph isomorphic network (GIN) for remaining useful life (RUL) prediction. GIN utilizes graph data and graph convolutions (GCs) to map complex relationships between degradation evolution and RUL. This approach outperforms existing deep learning models, such as convolutional neural networks (CNNs), long short-term memory (LSTM) networks, and CNN-LSTM, providing more accurate RUL prediction.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"24 22","pages":"38365-38376"},"PeriodicalIF":4.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142645389","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 : 2024-10-04DOI: 10.1109/JSEN.2024.3464635
Philip Krukenfellner;Elmar Rueckert;Helmut Flachberger
Vibrating screens are crucial in the waste and mineral processing industries. However, they often lack comprehensive digital monitoring, which necessitates subjective condition assessments. This study introduces a system developed in cooperation with IFE Aufbereitungstechnik GmbH that provides an objective machine state evaluation using permanently installed acceleration sensors, developed by eSensial Data Science GmbH. Unlike previous research, data for this project were collected from a linear vibrating screen, which is operating in a waste processing plant, introducing uncertainties and occasionally missing data due to sensor damage to the analysis. The study focuses on applying supervised machine learning algorithms to predict the machine’s operating condition. In particular, decision trees, multilayer perceptron (MLP) networks, and long-short-term memory (LSTM) networks that were evaluated using classical performance metrics such the MSE and the R2-Score. The models were also tested with respect to missing input data. The MLP network achieved a prediction accuracy of over 90%. Further, it displayed the ability to determine previously unlabeled intermediate states. Additionally, the main cause of prediction errors was identified and a method of handling missing input data was developed.
{"title":"Predicting Condition States, Based on Displacement Data, Generated by Acceleration Sensors on Industrial Linear Vibrating Screens Through Neural Networks","authors":"Philip Krukenfellner;Elmar Rueckert;Helmut Flachberger","doi":"10.1109/JSEN.2024.3464635","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3464635","url":null,"abstract":"Vibrating screens are crucial in the waste and mineral processing industries. However, they often lack comprehensive digital monitoring, which necessitates subjective condition assessments. This study introduces a system developed in cooperation with IFE Aufbereitungstechnik GmbH that provides an objective machine state evaluation using permanently installed acceleration sensors, developed by eSensial Data Science GmbH. Unlike previous research, data for this project were collected from a linear vibrating screen, which is operating in a waste processing plant, introducing uncertainties and occasionally missing data due to sensor damage to the analysis. The study focuses on applying supervised machine learning algorithms to predict the machine’s operating condition. In particular, decision trees, multilayer perceptron (MLP) networks, and long-short-term memory (LSTM) networks that were evaluated using classical performance metrics such the MSE and the R2-Score. The models were also tested with respect to missing input data. The MLP network achieved a prediction accuracy of over 90%. Further, it displayed the ability to determine previously unlabeled intermediate states. Additionally, the main cause of prediction errors was identified and a method of handling missing input data was developed.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"24 22","pages":"38232-38243"},"PeriodicalIF":4.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10705945","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142645435","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-04DOI: 10.1109/JSEN.2024.3459097
Kang Jiang;Songtao Xue;Liyu Xie;Guochun Wan;Zhuoran Yi;Zeyu Li
Conventional deformation sensors are inconvenient to deploy in large areas due to the employment of cables and power supplies. This article proposes a wireless passive sensor based on U-shaped resonators for bidirectional deformation sensing. The proposed sensor consists of two ultra-wideband (UWB) antennas for receiving the inquiry signal and retransmitting the coded signal, and a sensing element for encoding bidirectional deformation information in the signal. The two UWB antennas are set to cross-polarization to avoid interference from ambient reflected signals. The sensing element consists of a microstrip line for signal transmission, a U-shaped resonator for horizontal deformation monitoring at low frequency band, and a U-shaped resonator for vertical deformation monitoring at high frequency band. The deformation alters the overlapped length between the U-shaped resonator and its corresponding movable sub-patch, thereby shifting its resonant frequency. The absence of force between the U-shaped resonator and its corresponding movable sub-patch serves to prevent sensor damage from excessive force and to ensure a complete strain transfer ratio. The equivalent circuit model of the sensor is established to reveal its sensing mechanism, and the relationships between horizontal deformation and resonance frequency of low frequency band, and vertical deformation and resonance frequency of high frequency band, are studied by simulation in CST Microwave Studio. To verify the performance, the sensor is fabricated and installed on the concrete surface, and a series of wireless experiments are conducted. The experimental results demonstrate the feasibility of the sensor in simultaneously monitoring both horizontal and vertical deformation.
由于需要使用电缆和电源，传统的形变传感器不便在大面积区域部署。本文提出了一种基于 U 形谐振器的无线无源传感器，用于双向形变感应。该传感器由两个超宽带（UWB）天线和一个传感元件组成，前者用于接收询问信号和转发编码信号，后者用于在信号中编码双向形变信息。两个 UWB 天线设置为交叉极化，以避免环境反射信号的干扰。传感元件由一条用于信号传输的微带线、一个用于在低频段监测水平形变的 U 型谐振器和一个用于在高频段监测垂直形变的 U 型谐振器组成。形变会改变 U 形谐振器与其相应的可移动子贴片之间的重叠长度，从而移动其谐振频率。U 形谐振器和相应的可移动子贴片之间不受力，可防止传感器因受力过大而损坏，并确保完整的应变传递比。建立了传感器的等效电路模型以揭示其传感机制，并在 CST Microwave Studio 中模拟研究了水平变形与低频段谐振频率、垂直变形与高频段谐振频率之间的关系。为了验证其性能，制作了传感器并将其安装在混凝土表面，并进行了一系列无线实验。实验结果证明了传感器同时监测水平和垂直变形的可行性。
{"title":"A Wireless Passive Sensor Based on U-Shaped Resonators for Bidirectional Deformation Sensing","authors":"Kang Jiang;Songtao Xue;Liyu Xie;Guochun Wan;Zhuoran Yi;Zeyu Li","doi":"10.1109/JSEN.2024.3459097","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3459097","url":null,"abstract":"Conventional deformation sensors are inconvenient to deploy in large areas due to the employment of cables and power supplies. This article proposes a wireless passive sensor based on U-shaped resonators for bidirectional deformation sensing. The proposed sensor consists of two ultra-wideband (UWB) antennas for receiving the inquiry signal and retransmitting the coded signal, and a sensing element for encoding bidirectional deformation information in the signal. The two UWB antennas are set to cross-polarization to avoid interference from ambient reflected signals. The sensing element consists of a microstrip line for signal transmission, a U-shaped resonator for horizontal deformation monitoring at low frequency band, and a U-shaped resonator for vertical deformation monitoring at high frequency band. The deformation alters the overlapped length between the U-shaped resonator and its corresponding movable sub-patch, thereby shifting its resonant frequency. The absence of force between the U-shaped resonator and its corresponding movable sub-patch serves to prevent sensor damage from excessive force and to ensure a complete strain transfer ratio. The equivalent circuit model of the sensor is established to reveal its sensing mechanism, and the relationships between horizontal deformation and resonance frequency of low frequency band, and vertical deformation and resonance frequency of high frequency band, are studied by simulation in CST Microwave Studio. To verify the performance, the sensor is fabricated and installed on the concrete surface, and a series of wireless experiments are conducted. The experimental results demonstrate the feasibility of the sensor in simultaneously monitoring both horizontal and vertical deformation.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"24 22","pages":"36467-36476"},"PeriodicalIF":4.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142636506","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 : 2024-10-04DOI: 10.1109/JSEN.2024.3470515
Jun Wang;Ziwei Xu;Fuzhou Niu;Jinzhao Liu;Zhongkui Zhu
It is a great challenging task to diagnose compound faults of rolling bearings because of the complex coupling characteristics of the single faults. Compound fault samples are generally requisite to establish traditional compound fault diagnosis models. However, the infrequency of compound faults in bearings within industrial scenarios may result in a lack of available corresponding data for training the models. To address the above issue, this article proposes a new model named adaptively weighted semantic autoencoder (AWSAE) for bearing compound fault diagnosis based on zero-shot learning (ZSL). Specifically, the proposed AWSAE constructs compound fault semantics by weighted superposition of the semantics of the accessible single faults, in which the weights are adaptively determined by an attention mechanism. A projection matrix is established by the semantic autoencoder (SAE) that can effectively project the features of the tested compound fault samples to the corresponding semantics. Euclidean distances are then calculated in the semantic space to diagnose the types of the compound faults. In addition, to realize generalization zero-shot diagnosis, a prejudgment strategy is designed by integrating the idea of decoupling learning. The architecture of the proposed AWSAE model is simple because the feature extractor is used repeatedly in the prejudgment, the obtaining of adaptive weights, as well as the classification of all the health states. The proposed method is verified on two bearing datasets acquired from a rotor-bearing system and a wheel-rail system. The results show that the proposed AWSAE model performs well in identifying the bearing compound faults and is superior to the most advanced ZSL models.
{"title":"Adaptively Weighted Semantic Autoencoder for Zero-Shot Compound Fault Diagnosis","authors":"Jun Wang;Ziwei Xu;Fuzhou Niu;Jinzhao Liu;Zhongkui Zhu","doi":"10.1109/JSEN.2024.3470515","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3470515","url":null,"abstract":"It is a great challenging task to diagnose compound faults of rolling bearings because of the complex coupling characteristics of the single faults. Compound fault samples are generally requisite to establish traditional compound fault diagnosis models. However, the infrequency of compound faults in bearings within industrial scenarios may result in a lack of available corresponding data for training the models. To address the above issue, this article proposes a new model named adaptively weighted semantic autoencoder (AWSAE) for bearing compound fault diagnosis based on zero-shot learning (ZSL). Specifically, the proposed AWSAE constructs compound fault semantics by weighted superposition of the semantics of the accessible single faults, in which the weights are adaptively determined by an attention mechanism. A projection matrix is established by the semantic autoencoder (SAE) that can effectively project the features of the tested compound fault samples to the corresponding semantics. Euclidean distances are then calculated in the semantic space to diagnose the types of the compound faults. In addition, to realize generalization zero-shot diagnosis, a prejudgment strategy is designed by integrating the idea of decoupling learning. The architecture of the proposed AWSAE model is simple because the feature extractor is used repeatedly in the prejudgment, the obtaining of adaptive weights, as well as the classification of all the health states. The proposed method is verified on two bearing datasets acquired from a rotor-bearing system and a wheel-rail system. The results show that the proposed AWSAE model performs well in identifying the bearing compound faults and is superior to the most advanced ZSL models.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"24 22","pages":"37472-37481"},"PeriodicalIF":4.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142645492","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}
Dimensionality reduction is an important task for Wi-Fi-based indoor localization (IL). Most such techniques do not take into account realistic data collection issues such as the presence of outliers or inconsistent fingerprint instances. These fingerprints either represent a class boundary or an outlier. Instance hardness is a measure that better characterizes such instances. Accordingly, in this work, our contribution is to propose a convolutional autoencoder-based dimensionality reduction approach that works on the basis of feature transformation and instance hardness. The encoding process of the data input involves a two-channel representation of a fingerprint dataset that holds the normalized RSS and an instance hardness measure, that is, a k-disagreeing score. The inclusion of the k-disagreeing score into the training pipeline is made with the objective of injecting instance importance for training using 1-D CNN architectures for classification. The experimentations were performed on three benchmark datasets and a collected dataset. The proposed pipeline is found to yield an accuracy of more than 97% with error deviation ranging from 2.2– $2.37{m}$ which is quite acceptable for any localization system.
{"title":"Dimensionality Reduction Through Multiple Convolutional Channels for RSS-Based Indoor Localization","authors":"Ayan Kumar Panja;Snehan Biswas;Sarmistha Neogy;Chandreyee Chowdhury","doi":"10.1109/JSEN.2024.3470549","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3470549","url":null,"abstract":"Dimensionality reduction is an important task for Wi-Fi-based indoor localization (IL). Most such techniques do not take into account realistic data collection issues such as the presence of outliers or inconsistent fingerprint instances. These fingerprints either represent a class boundary or an outlier. Instance hardness is a measure that better characterizes such instances. Accordingly, in this work, our contribution is to propose a convolutional autoencoder-based dimensionality reduction approach that works on the basis of feature transformation and instance hardness. The encoding process of the data input involves a two-channel representation of a fingerprint dataset that holds the normalized RSS and an instance hardness measure, that is, a k-disagreeing score. The inclusion of the k-disagreeing score into the training pipeline is made with the objective of injecting instance importance for training using 1-D CNN architectures for classification. The experimentations were performed on three benchmark datasets and a collected dataset. The proposed pipeline is found to yield an accuracy of more than 97% with error deviation ranging from 2.2–\u0000<inline-formula> <tex-math>$2.37{m}$ </tex-math></inline-formula>\u0000 which is quite acceptable for any localization system.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"24 22","pages":"37482-37491"},"PeriodicalIF":4.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142645433","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 : 2024-10-04DOI: 10.1109/JSEN.2024.3469991
Jie Li;Peidong Xue;Zhengwu Zhu;Huiming Wu;Junguo Xu
Monitoring milli-thermal variations and achieving high dynamic response are crucial aspects of temperature sensor technology. Achieving high dynamic response requires a delicate structure design of temperature sensor, which can minimize its thermal capacity to facilitate quick reactions while keeping enough mechanical robustness. In this study, we introduce an innovative nano-platinum-thin-film sensor with an exceptionally low thermal capacity. With the dimensions of $10~mu $ m $times 2~mu $ m $times 30$ nm ( $0.6~mu $ m3), this sensor is meticulously engineered to offer a high dynamic resolution of 0.016° C, a response time of 21 ms in water drop experiments, and a high-frequency dynamic response capability up to 20 kHz. Its time constant, approximately $25.5~mu $ s for nominal temperature change, highlights its ability to capture transient thermal events effectively. The sensor’s high dynamic characteristics render it uniquely suited for applications demanding the detection of rapid thermal phenomena, such as object impacts and high-frequency thermal fluctuations, which are crucial in a variety of scientific and industrial fields.
监测微热变化和实现高动态响应是温度传感器技术的关键环节。要实现高动态响应，就必须对温度传感器进行精细的结构设计，在保持足够机械坚固性的同时，最大限度地降低其热容量，以促进快速反应。在这项研究中，我们介绍了一种热容量极低的创新型纳米铂薄膜传感器。这种传感器的尺寸为 10~mu $ m $/times 2~mu $ m $/times 30$ nm ( $0.6~mu $ m3)，经过精心设计，具有 0.016° C 的高动态分辨率、21 ms 的水滴实验响应时间以及高达 20 kHz 的高频动态响应能力。其时间常数（标称温度变化时约为 25.5 美元/秒）可有效捕捉瞬态热事件。该传感器的高动态特性使其非常适合需要检测快速热现象的应用，如物体撞击和高频热波动，这些在各种科学和工业领域都至关重要。
{"title":"A Nano-Platinum-Thin-Film Sensor With Milli-Thermal Resolution and Microsec Response","authors":"Jie Li;Peidong Xue;Zhengwu Zhu;Huiming Wu;Junguo Xu","doi":"10.1109/JSEN.2024.3469991","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3469991","url":null,"abstract":"Monitoring milli-thermal variations and achieving high dynamic response are crucial aspects of temperature sensor technology. Achieving high dynamic response requires a delicate structure design of temperature sensor, which can minimize its thermal capacity to facilitate quick reactions while keeping enough mechanical robustness. In this study, we introduce an innovative nano-platinum-thin-film sensor with an exceptionally low thermal capacity. With the dimensions of \u0000<inline-formula> <tex-math>$10~mu $ </tex-math></inline-formula>\u0000 m \u0000<inline-formula> <tex-math>$times 2~mu $ </tex-math></inline-formula>\u0000 m \u0000<inline-formula> <tex-math>$times 30$ </tex-math></inline-formula>\u0000 nm (\u0000<inline-formula> <tex-math>$0.6~mu $ </tex-math></inline-formula>\u0000 m3), this sensor is meticulously engineered to offer a high dynamic resolution of 0.016° C, a response time of 21 ms in water drop experiments, and a high-frequency dynamic response capability up to 20 kHz. Its time constant, approximately \u0000<inline-formula> <tex-math>$25.5~mu $ </tex-math></inline-formula>\u0000 s for nominal temperature change, highlights its ability to capture transient thermal events effectively. The sensor’s high dynamic characteristics render it uniquely suited for applications demanding the detection of rapid thermal phenomena, such as object impacts and high-frequency thermal fluctuations, which are crucial in a variety of scientific and industrial fields.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"24 22","pages":"36419-36425"},"PeriodicalIF":4.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142636564","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 : 2024-10-04DOI: 10.1109/JSEN.2024.3469376
Yang Zhang;Yongzhao Sun;Yue Zhou;Wenjie Hao;Tingting Lin
Magnetic resonance sounding (MRS) is a noninvasive geophysical method, which may detect the underground water directly. However, the weak MRS signals oscillating at Larmor frequency always suffer from low signal-to-noise ratios (SNRs) due to the multisource noise, especially the random noise. To solve this problem, a novel method for random noise suppression based on cross correlation is proposed in this manuscript. According to the characteristics of the MRS signal, a sinusoidal signal is constructed as a reference signal, which has the same frequency as Larmor frequency. It shows a strong correlation with the MRS signal, while demonstrating minimal correlation with the random noise. In terms of this property, the cross correlation is used to recover the MRS signal from random noise interference. By convolving the noisy signal with the reference signal and deconvolving the processed convolution waveform, the desired MRS signal is acquired. In order to validate the efficiency of the denoising strategy, numerical simulations on the synthetic signals embedded in different noise levels are performed, and the uncertainties of the estimated signal parameters are compared. In addition, the cross correlation method is applied following a standard processing scheme in field data, also resulting in improved SNRs. The cross correlation algorithm may achieve better denoising results than the commonly used denoising method with fewer filtering parameters and less human labor.
{"title":"Random Noise Suppression of Magnetic Resonance Sounding Oscillating Signal Based on Cross Correlation","authors":"Yang Zhang;Yongzhao Sun;Yue Zhou;Wenjie Hao;Tingting Lin","doi":"10.1109/JSEN.2024.3469376","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3469376","url":null,"abstract":"Magnetic resonance sounding (MRS) is a noninvasive geophysical method, which may detect the underground water directly. However, the weak MRS signals oscillating at Larmor frequency always suffer from low signal-to-noise ratios (SNRs) due to the multisource noise, especially the random noise. To solve this problem, a novel method for random noise suppression based on cross correlation is proposed in this manuscript. According to the characteristics of the MRS signal, a sinusoidal signal is constructed as a reference signal, which has the same frequency as Larmor frequency. It shows a strong correlation with the MRS signal, while demonstrating minimal correlation with the random noise. In terms of this property, the cross correlation is used to recover the MRS signal from random noise interference. By convolving the noisy signal with the reference signal and deconvolving the processed convolution waveform, the desired MRS signal is acquired. In order to validate the efficiency of the denoising strategy, numerical simulations on the synthetic signals embedded in different noise levels are performed, and the uncertainties of the estimated signal parameters are compared. In addition, the cross correlation method is applied following a standard processing scheme in field data, also resulting in improved SNRs. The cross correlation algorithm may achieve better denoising results than the commonly used denoising method with fewer filtering parameters and less human labor.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"24 22","pages":"37463-37471"},"PeriodicalIF":4.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142645532","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 : 2024-10-04DOI: 10.1109/JSEN.2024.3470069
Sunil Kumar;Naresh Kedam;Evgeny A. Maksimovskiy;Arcady V. Ishchenko;Tatyana V. Larina;Yuriy A. Chesalov;Alexander G. Bannov
In order to manage the environment and perform noninvasive disease diagnostics, it is necessary to continuously identify harmful and highly toxic gases, such as nitrogen dioxide (NO2). This study demonstrates how to design nanocomposites and build a cost-effective NO2 gas sensor based on exfoliated tungsten disulphide and functionalized multiwalled carbon nanotubes (f-MWCNTs) as a highly efficient sensing material operating at room temperature (RT) in humid conditions. The composite sensor’s response under various humidity levels, ranging from 2% to 65%, as well as at different temperatures ( $25~^{circ }$ C– $80~^{circ }$ C), was studied. Scanning electron microscopy (SEM), Raman spectroscopy, transmission electron microscopy (TEM), and energy-dispersive X-ray spectroscopy (EDX) were used to analyze the sensing material. The composite-based sensor showed an improved response $Delta {R}/{R}_{{0}}$ of 52% at RT for 50-ppm NO2 with good selectivity to other gases (e.g., ammonia, methane, benzene, isobutene, and hydrogen). The composite sensor exhibited a low detection limit of 1.39 ppm for NO2 at RT. Furthering this advancement, we delve into the integration of machine learning, specifically the CatBoost regression model, with the NO2 sensor. This integration elevates the sensor from a conventional passive detector to an advanced analytical system, significantly boosting its predictive accuracy and adaptability for real-time environmental monitoring and nuanced data interpretation, thereby opening new frontiers in sensor technology and applications in environmental monitoring and health diagnostics.
{"title":"Integration of Exfoliated WS2/Functionalized MWCNT Nanocomposites for NO2 Sensing Using Machine Learning for Response Prediction","authors":"Sunil Kumar;Naresh Kedam;Evgeny A. Maksimovskiy;Arcady V. Ishchenko;Tatyana V. Larina;Yuriy A. Chesalov;Alexander G. Bannov","doi":"10.1109/JSEN.2024.3470069","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3470069","url":null,"abstract":"In order to manage the environment and perform noninvasive disease diagnostics, it is necessary to continuously identify harmful and highly toxic gases, such as nitrogen dioxide (NO2). This study demonstrates how to design nanocomposites and build a cost-effective NO2 gas sensor based on exfoliated tungsten disulphide and functionalized multiwalled carbon nanotubes (f-MWCNTs) as a highly efficient sensing material operating at room temperature (RT) in humid conditions. The composite sensor’s response under various humidity levels, ranging from 2% to 65%, as well as at different temperatures (\u0000<inline-formula> <tex-math>$25~^{circ }$ </tex-math></inline-formula>\u0000C–\u0000<inline-formula> <tex-math>$80~^{circ }$ </tex-math></inline-formula>\u0000C), was studied. Scanning electron microscopy (SEM), Raman spectroscopy, transmission electron microscopy (TEM), and energy-dispersive X-ray spectroscopy (EDX) were used to analyze the sensing material. The composite-based sensor showed an improved response \u0000<inline-formula> <tex-math>$Delta {R}/{R}_{{0}}$ </tex-math></inline-formula>\u0000 of 52% at RT for 50-ppm NO2 with good selectivity to other gases (e.g., ammonia, methane, benzene, isobutene, and hydrogen). The composite sensor exhibited a low detection limit of 1.39 ppm for NO2 at RT. Furthering this advancement, we delve into the integration of machine learning, specifically the CatBoost regression model, with the NO2 sensor. This integration elevates the sensor from a conventional passive detector to an advanced analytical system, significantly boosting its predictive accuracy and adaptability for real-time environmental monitoring and nuanced data interpretation, thereby opening new frontiers in sensor technology and applications in environmental monitoring and health diagnostics.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"24 22","pages":"36366-36376"},"PeriodicalIF":4.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142636551","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 : 2024-10-04DOI: 10.1109/JSEN.2024.3470591
José Carvalho;Luís Cunha;Sandro Pinto;Tiago Gomes
The automotive industry keeps moving fast toward the development of smarter, safer, and more sustainable autonomous vehicles. Today, these come equipped with advanced driver assistance systems (ADAS), which include sophisticated perception technologies to safely navigate the environment. One of the key sensors present in the perception system is light detection and ranging (LiDAR). It can accurately measure distances to objects and create detailed real-time 3-D maps of the surrounding environment, including obstacles and road boundaries. Extracting the road information to identify the drivable area is one of the most important steps applied to a LiDAR output; however, due to the amount of data a high-resolution sensor generates, this task becomes quite challenging. This article proposes FESTA, a ground segmentation technique accelerated in field-programmable gate arrays (FPGAs) using the ALFA framework, that can execute the ground segmentation step applied to the sensor output in real time. The performed evaluation shows that FESTA requires, on average, 8.92 ms for processing a point cloud frame from a VLP-16 sensor, 14.41 ms for an HDL-32, 40.87 ms for an HDL-64, and 70.59 ms for a VLS-128, while outperforming state-of-the-art algorithms in other performance metrics.
{"title":"FESTA: FPGA-Enabled Ground Segmentation Technique for Automotive LiDAR","authors":"José Carvalho;Luís Cunha;Sandro Pinto;Tiago Gomes","doi":"10.1109/JSEN.2024.3470591","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3470591","url":null,"abstract":"The automotive industry keeps moving fast toward the development of smarter, safer, and more sustainable autonomous vehicles. Today, these come equipped with advanced driver assistance systems (ADAS), which include sophisticated perception technologies to safely navigate the environment. One of the key sensors present in the perception system is light detection and ranging (LiDAR). It can accurately measure distances to objects and create detailed real-time 3-D maps of the surrounding environment, including obstacles and road boundaries. Extracting the road information to identify the drivable area is one of the most important steps applied to a LiDAR output; however, due to the amount of data a high-resolution sensor generates, this task becomes quite challenging. This article proposes FESTA, a ground segmentation technique accelerated in field-programmable gate arrays (FPGAs) using the ALFA framework, that can execute the ground segmentation step applied to the sensor output in real time. The performed evaluation shows that FESTA requires, on average, 8.92 ms for processing a point cloud frame from a VLP-16 sensor, 14.41 ms for an HDL-32, 40.87 ms for an HDL-64, and 70.59 ms for a VLS-128, while outperforming state-of-the-art algorithms in other performance metrics.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"24 22","pages":"38005-38014"},"PeriodicalIF":4.3,"publicationDate":"2024-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142636389","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：