Electroencephalogram (EEG) signal classification is of utmost importance in brain-computer interface (BCI) systems. However, the inherent complex properties of EEG signals pose a challenge in their analysis and modeling. This letter proposes a novel approach of integrating wavelet scattering transform (WST) with convolutional neural network (CNN) for classifying motor imagery (MI) via EEG signals (referred as WST-CNN), capable of extracting distinctive characteristics in signals even when the data is limited. In this architecture, the first layer is nontrainable WST features with fixed initializations in WST-CNN. Furthermore, WSTs are robust to local perturbations in data, especially in the form of translation invariance, and resilient to deformations, thereby enhancing the network's reliability. The performance of the proposed idea is evaluated on the DBCIE dataset for three different scenarios: left-arm (LA) movement, right-arm (RA) movement, and simultaneous movement of both arms (BA). The BCI Competition IV-2a dataset was also employed to validate the proposed concept across four distinct MI tasks, such as movements in: left-hand (LH), right-hand (RH), feet (FT), and tongue (T). The classifications' performance was evaluated in terms of accuracy ( $eta$ ), sensitivity ( $S_{e}$ ), specificity ( $S_{p}$ ), and weighted F1-score, which reached up to 92.72%, 92.72%, 97.57%, and 92.75% for classifying LH, RH, FT, and T on the BCI Competition IV-2a dataset and 89.19%, 89.19%, 94.60%, and 89.33% for classifying LA, RA, and BA, on the DBCIE dataset, respectively.
{"title":"Classification of Motor Imagery Tasks Using EEG Based on Wavelet Scattering Transform and Convolutional Neural Network","authors":"Rantu Buragohain;Jejariya Ajaybhai;Karan Nathwani;Vinayak Abrol","doi":"10.1109/LSENS.2024.3488356","DOIUrl":"https://doi.org/10.1109/LSENS.2024.3488356","url":null,"abstract":"Electroencephalogram (EEG) signal classification is of utmost importance in brain-computer interface (BCI) systems. However, the inherent complex properties of EEG signals pose a challenge in their analysis and modeling. This letter proposes a novel approach of integrating wavelet scattering transform (WST) with convolutional neural network (CNN) for classifying motor imagery (MI) via EEG signals (referred as WST-CNN), capable of extracting distinctive characteristics in signals even when the data is limited. In this architecture, the first layer is nontrainable WST features with fixed initializations in WST-CNN. Furthermore, WSTs are robust to local perturbations in data, especially in the form of translation invariance, and resilient to deformations, thereby enhancing the network's reliability. The performance of the proposed idea is evaluated on the DBCIE dataset for three different scenarios: left-arm (LA) movement, right-arm (RA) movement, and simultaneous movement of both arms (BA). The BCI Competition IV-2a dataset was also employed to validate the proposed concept across four distinct MI tasks, such as movements in: left-hand (LH), right-hand (RH), feet (FT), and tongue (T). The classifications' performance was evaluated in terms of accuracy (\u0000<inline-formula><tex-math>$eta$</tex-math></inline-formula>\u0000), sensitivity (\u0000<inline-formula><tex-math>$S_{e}$</tex-math></inline-formula>\u0000), specificity (\u0000<inline-formula><tex-math>$S_{p}$</tex-math></inline-formula>\u0000), and weighted F1-score, which reached up to 92.72%, 92.72%, 97.57%, and 92.75% for classifying LH, RH, FT, and T on the BCI Competition IV-2a dataset and 89.19%, 89.19%, 94.60%, and 89.33% for classifying LA, RA, and BA, on the DBCIE dataset, respectively.","PeriodicalId":13014,"journal":{"name":"IEEE Sensors Letters","volume":"8 12","pages":"1-4"},"PeriodicalIF":2.2,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142636262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-06DOI: 10.1109/LSENS.2024.3492373
Winston Doss Marveldoss;Bandaru Joshika;Bijo Sebastian
Accurate perception of the environment, including the detection and tracking of humans, is essential for safe navigation of mobile robots in human-centric environments. Existing State-of-the-Art techniques rely on high-performance sensors. This leads to expensive robotic systems, which limits the large-scale deployment of autonomous mobile robots in social spaces. In this letter, we propose and validate a novel human tracking and estimation approach that relies on a low-cost 2-D LiDAR and a monocular camera. The proposed approach leverages the capabilities of each sensor by relying on the camera for human detection and the LiDAR for human pose estimation. Precise calibration and registration of the sensor frames allow for data association in the presence of multiple human targets. Human detection and pose estimation data from the sensor suite are used as measurement by an extended Kalman filter, which allows for effective tracking over multiple frames, even in the presence of occlusion. The overall approach addresses the limitations of each individual sensor without increasing the overall cost of the sensor suite. Tracking and estimation performance for the proposed approach was evaluated on experimental trails in real-world conditions with artificial markers as ground truth for each human target. The results demonstrate satisfactory performance for the proposed approach to be used in human-aware autonomous navigation in real-world settings.
{"title":"Tracking and Estimation Approach for Human-Aware Mobile Robot Navigation","authors":"Winston Doss Marveldoss;Bandaru Joshika;Bijo Sebastian","doi":"10.1109/LSENS.2024.3492373","DOIUrl":"https://doi.org/10.1109/LSENS.2024.3492373","url":null,"abstract":"Accurate perception of the environment, including the detection and tracking of humans, is essential for safe navigation of mobile robots in human-centric environments. Existing State-of-the-Art techniques rely on high-performance sensors. This leads to expensive robotic systems, which limits the large-scale deployment of autonomous mobile robots in social spaces. In this letter, we propose and validate a novel human tracking and estimation approach that relies on a low-cost 2-D LiDAR and a monocular camera. The proposed approach leverages the capabilities of each sensor by relying on the camera for human detection and the LiDAR for human pose estimation. Precise calibration and registration of the sensor frames allow for data association in the presence of multiple human targets. Human detection and pose estimation data from the sensor suite are used as measurement by an extended Kalman filter, which allows for effective tracking over multiple frames, even in the presence of occlusion. The overall approach addresses the limitations of each individual sensor without increasing the overall cost of the sensor suite. Tracking and estimation performance for the proposed approach was evaluated on experimental trails in real-world conditions with artificial markers as ground truth for each human target. The results demonstrate satisfactory performance for the proposed approach to be used in human-aware autonomous navigation in real-world settings.","PeriodicalId":13014,"journal":{"name":"IEEE Sensors Letters","volume":"8 12","pages":"1-4"},"PeriodicalIF":2.2,"publicationDate":"2024-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142679324","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-04DOI: 10.1109/LSENS.2024.3490983
Wei Xu;Wenlin Xiao;Ke Xiao
This letter presents the modeling and characterization of a pulse-excited micro thermal flow sensor based on electrochemical impedance sensing. The proposed transient model reveals that the sensor output, measured as the impedance slope under pulse excitation, is almost one order of magnitude stronger at the downstream electrodes, as compared to the upstream pair. Consequently, the micro-electromechanical systems (MEMS) flow sensor is designed with an 8-μm-thick flexible structure and a 1.4 mm distance between the microheater and downstream electrodes. Testing results show that the fabricated impedance-type micro flow sensor achieves a maximum sensitivity of 8.9 (mΩ/s)/(μm/s) for the 1X PBS flow, while consuming less than 15.8 mW of heating power with a fluid flow up to 750 μm/s. Furthermore, the proposed theoretical model closely aligns with experimental results, confirming its potential as a valuable tool for optimizing impedance-type flow sensors that utilize pulse heating strategies to detect extremely low fluid flow in the future.
{"title":"Modeling and Characterizing an Impedance-Type Micro Flow Sensor With Pulse Excitation","authors":"Wei Xu;Wenlin Xiao;Ke Xiao","doi":"10.1109/LSENS.2024.3490983","DOIUrl":"https://doi.org/10.1109/LSENS.2024.3490983","url":null,"abstract":"This letter presents the modeling and characterization of a pulse-excited micro thermal flow sensor based on electrochemical impedance sensing. The proposed transient model reveals that the sensor output, measured as the impedance slope under pulse excitation, is almost one order of magnitude stronger at the downstream electrodes, as compared to the upstream pair. Consequently, the micro-electromechanical systems (MEMS) flow sensor is designed with an 8-μm-thick flexible structure and a 1.4 mm distance between the microheater and downstream electrodes. Testing results show that the fabricated impedance-type micro flow sensor achieves a maximum sensitivity of 8.9 (mΩ/s)/(μm/s) for the 1X PBS flow, while consuming less than 15.8 mW of heating power with a fluid flow up to 750 μm/s. Furthermore, the proposed theoretical model closely aligns with experimental results, confirming its potential as a valuable tool for optimizing impedance-type flow sensors that utilize pulse heating strategies to detect extremely low fluid flow in the future.","PeriodicalId":13014,"journal":{"name":"IEEE Sensors Letters","volume":"8 12","pages":"1-4"},"PeriodicalIF":2.2,"publicationDate":"2024-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142672107","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-04DOI: 10.1109/LSENS.2024.3490837
Saraswati Kulkarni;Ruma Ghosh
Analysis and understanding of the mixture of volatile organic compound (VOC) sensing are crucial for the development of sensors in conditions closer to real-life applications, such as health care, air quality monitoring, industrial safety, etc. In this study, we investigated the response dynamics of CuO-nanomaterial-based resistive sensors to 25–75 ppm of individual, binary, ternary, and quaternary mixtures of five VOCs—acetone, acetonitrile, isopropanol, methanol, and toluene at 300 °C. The CuO exhibited responses equal to the sum of its steady-state responses to individual VOCs for all the possible combinations of binary and ternary mixtures with 25 ppm of the constituent gases. A systematic study based on the recovery cycle was conducted by retracting the VOCs sequentially from the proximity of CuO surface after recording response cycle. Interestingly, the recovery time constant τ rec was found to follow the order—isopropanol (96.93 – 435.45) ≥ methanol (111.82 – 313.21) > toluene (9.9 – 220.49) > acetonitrile (85.96 – 332.32) > acetone (could not be found) in all binary, ternary, and quaternary mixtures of VOCs, irrespective of the sequence of retraction of the VOCs from the mixture. Also, it was found that VOCs with –OH groups have higher adsorption capacity on the sensing layer as compared to –NH 2 , – C = O, and aromatic VOCs.
{"title":"Investigation of Surface Affinity and Desorption Kinetics of Mixture of Volatile Organic Compounds on CuO-Based Resistive Gas Sensors","authors":"Saraswati Kulkarni;Ruma Ghosh","doi":"10.1109/LSENS.2024.3490837","DOIUrl":"https://doi.org/10.1109/LSENS.2024.3490837","url":null,"abstract":"Analysis and understanding of the mixture of volatile organic compound (VOC) sensing are crucial for the development of sensors in conditions closer to real-life applications, such as health care, air quality monitoring, industrial safety, etc. In this study, we investigated the response dynamics of CuO-nanomaterial-based resistive sensors to 25–75 ppm of individual, binary, ternary, and quaternary mixtures of five VOCs—acetone, acetonitrile, isopropanol, methanol, and toluene at 300 °C. The CuO exhibited responses equal to the sum of its steady-state responses to individual VOCs for all the possible combinations of binary and ternary mixtures with 25 ppm of the constituent gases. A systematic study based on the recovery cycle was conducted by retracting the VOCs sequentially from the proximity of CuO surface after recording response cycle. Interestingly, the recovery time constant τ\u0000<sub>rec</sub>\u0000 was found to follow the order—isopropanol (96.93 – 435.45) ≥ methanol (111.82 – 313.21) > toluene (9.9 – 220.49) > acetonitrile (85.96 – 332.32) > acetone (could not be found) in all binary, ternary, and quaternary mixtures of VOCs, irrespective of the sequence of retraction of the VOCs from the mixture. Also, it was found that VOCs with –OH groups have higher adsorption capacity on the sensing layer as compared to –NH\u0000<sub>2</sub>\u0000, – C = O, and aromatic VOCs.","PeriodicalId":13014,"journal":{"name":"IEEE Sensors Letters","volume":"8 12","pages":"1-4"},"PeriodicalIF":2.2,"publicationDate":"2024-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142672206","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this letter, a novel fully flexible piezotronic bipolar junction transistor (n-p-n PBJT) is designed and constructed by configuring two ZnO/Poly(3-hexylthiophene) heterojunction diodes back to back. The n-p-n PBJT acts as a signal-mediated device providing both current and voltage as the outputs. The utilization of the n-p-n PBJT in wearable applications is testified where a unique synergy between energy harvesting and sensing is found. Under mechanical stress, the output signal is amplified (with no preamplifier circuitry), which makes it a proper candidate as a high-performance sensor (voltage-based sensitivity is extracted as 0.49 V/kPa, four times higher than piezotronic p-n heterojunction). As a wearable energy harvester, the output signal is rectified (with no signal regulation circuitry), and it generates a peak output power of 2.9 µW, which is ten times higher than that of the piezotronic p-n diode. The outstanding performance of the n-p-n PBJT provides a new strategy to improve device performance for the emerging application in wearable electronics.
{"title":"A Comparative Study on Synergy Between Energy Harvesting and Pressure Sensing in Piezotronic Heterojunctions","authors":"Zihao Liang;Emad Iranmanesh;Shuxin Lin;Weipeng Xuan;Hang Zhou","doi":"10.1109/LSENS.2024.3491581","DOIUrl":"https://doi.org/10.1109/LSENS.2024.3491581","url":null,"abstract":"In this letter, a novel fully flexible piezotronic bipolar junction transistor (n-p-n PBJT) is designed and constructed by configuring two ZnO/Poly(3-hexylthiophene) heterojunction diodes back to back. The n-p-n PBJT acts as a signal-mediated device providing both current and voltage as the outputs. The utilization of the n-p-n PBJT in wearable applications is testified where a unique synergy between energy harvesting and sensing is found. Under mechanical stress, the output signal is amplified (with no preamplifier circuitry), which makes it a proper candidate as a high-performance sensor (voltage-based sensitivity is extracted as 0.49 V/kPa, four times higher than piezotronic p-n heterojunction). As a wearable energy harvester, the output signal is rectified (with no signal regulation circuitry), and it generates a peak output power of 2.9 µW, which is ten times higher than that of the piezotronic p-n diode. The outstanding performance of the n-p-n PBJT provides a new strategy to improve device performance for the emerging application in wearable electronics.","PeriodicalId":13014,"journal":{"name":"IEEE Sensors Letters","volume":"8 12","pages":"1-4"},"PeriodicalIF":2.2,"publicationDate":"2024-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142679315","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-01DOI: 10.1109/LSENS.2024.3490658
Qihao Zeng;Mingkun Wang;Yupeng Zhang;Hongyi Lin;Wei Qiao;Dong Sun
Optical path difference is commonly used to adjust the signal of coherent light. Current optical systems typically adjust the transmission distance of the beam or the refractive index of the medium to change the optical path. However, the former approach often involves complex operations, risks of mechanical damage, and difficulties in adjustment, while the latter has a limited adjustment range. This letter proposes a Kerr-fiber-based optical path difference adjustment system. In this system, the Kerr liquid inside the fiber induces a change in birefringence due to the electric field's polarization effect, enabling rapid optical path adjustment. The system adjusts the effective refractive index by applying an external voltage: a 10-V voltage induces a one-cycle change in interference fringes, and increasing the voltage to 50 V results in a 0.1 change in the refractive index, with a minimum adjustment precision of 0.01. Experimental results demonstrate a millisecond-level response rate for the overall system. Comparative tests show that this method is similar to classic adjustment methods but offers simplified operation. In addition, the system exhibits enhanced stability in scenarios requiring rapid and precise adjustments.
光路差通常用于调整相干光的信号。目前的光学系统通常通过调整光束的传输距离或介质的折射率来改变光路。然而，前一种方法往往涉及复杂的操作、机械损坏风险和调整困难，而后一种方法的调整范围有限。本文提出了一种基于克尔光纤的光路差调整系统。在该系统中，光纤内部的克尔液体会因电场的偏振效应而引起双折射变化，从而实现快速光路调整。该系统通过施加外部电压来调整有效折射率：10 V 的电压会引起干涉条纹一个周期的变化，而将电压提高到 50 V 则会导致折射率发生 0.1 的变化，最低调整精度为 0.01。实验结果表明，整个系统的响应速度达到毫秒级。对比测试表明，这种方法与传统的调节方法相似，但操作更简便。此外，在需要快速精确调节的情况下，该系统也表现出更高的稳定性。
{"title":"Optical Path Difference Modulation Method Based on the Kerr Effect","authors":"Qihao Zeng;Mingkun Wang;Yupeng Zhang;Hongyi Lin;Wei Qiao;Dong Sun","doi":"10.1109/LSENS.2024.3490658","DOIUrl":"https://doi.org/10.1109/LSENS.2024.3490658","url":null,"abstract":"Optical path difference is commonly used to adjust the signal of coherent light. Current optical systems typically adjust the transmission distance of the beam or the refractive index of the medium to change the optical path. However, the former approach often involves complex operations, risks of mechanical damage, and difficulties in adjustment, while the latter has a limited adjustment range. This letter proposes a Kerr-fiber-based optical path difference adjustment system. In this system, the Kerr liquid inside the fiber induces a change in birefringence due to the electric field's polarization effect, enabling rapid optical path adjustment. The system adjusts the effective refractive index by applying an external voltage: a 10-V voltage induces a one-cycle change in interference fringes, and increasing the voltage to 50 V results in a 0.1 change in the refractive index, with a minimum adjustment precision of 0.01. Experimental results demonstrate a millisecond-level response rate for the overall system. Comparative tests show that this method is similar to classic adjustment methods but offers simplified operation. In addition, the system exhibits enhanced stability in scenarios requiring rapid and precise adjustments.","PeriodicalId":13014,"journal":{"name":"IEEE Sensors Letters","volume":"8 12","pages":"1-4"},"PeriodicalIF":2.2,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142672011","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-30DOI: 10.1109/LSENS.2024.3488652
Jair A.Lima Silva;Wesley Costa;Khan Md Mazedul Islam;Jan Kleine Deters;Ewout Bergsma;Helder R. Oliveira Rocha;Patrick Noordhoek;Heinrich Wörtche
Reliability is a constraint of low-power wireless connectivity, commonly addressed by the deployment of mesh topology. Accordingly, power consumption becomes a major concern during the design and implementation of such networks. Thus, a mono-objective optimization was implemented in this work to decrease the total amount of power consumed by a low-power wireless mesh network based on Thread protocol. Using a genetic algorithm, the optimization procedure takes into account a predefined connectivity matrix, in which the possible distances between all network devices are considered. The experimental proof-of-concept shows that a mean gain of 26.45 dB is achievable in a specific scenario. Through our experimental results, we conclude that the Thread mesh protocol has much leeway to meet the low-power consumption requirement of wireless sensor networks.
可靠性是低功耗无线连接的一个制约因素，通常通过部署网状拓扑结构来解决。因此，在此类网络的设计和实施过程中，功耗成为一个主要问题。因此，本研究采用了单目标优化方法，以降低基于 Thread 协议的低功耗无线网状网络的总功耗。优化程序采用遗传算法，考虑了预定义的连接矩阵，其中考虑了所有网络设备之间的可能距离。概念验证实验表明，在特定场景下可实现 26.45 dB 的平均增益。通过实验结果，我们得出结论：Thread 网状协议在满足无线传感器网络的低功耗要求方面有很大的余地。
{"title":"Improving the Power Consumption of Thread Mesh Networks Through Genetic Algorithm Optimization","authors":"Jair A.Lima Silva;Wesley Costa;Khan Md Mazedul Islam;Jan Kleine Deters;Ewout Bergsma;Helder R. Oliveira Rocha;Patrick Noordhoek;Heinrich Wörtche","doi":"10.1109/LSENS.2024.3488652","DOIUrl":"https://doi.org/10.1109/LSENS.2024.3488652","url":null,"abstract":"Reliability is a constraint of low-power wireless connectivity, commonly addressed by the deployment of mesh topology. Accordingly, power consumption becomes a major concern during the design and implementation of such networks. Thus, a mono-objective optimization was implemented in this work to decrease the total amount of power consumed by a low-power wireless mesh network based on Thread protocol. Using a genetic algorithm, the optimization procedure takes into account a predefined connectivity matrix, in which the possible distances between all network devices are considered. The experimental proof-of-concept shows that a mean gain of 26.45 dB is achievable in a specific scenario. Through our experimental results, we conclude that the Thread mesh protocol has much leeway to meet the low-power consumption requirement of wireless sensor networks.","PeriodicalId":13014,"journal":{"name":"IEEE Sensors Letters","volume":"8 12","pages":"1-4"},"PeriodicalIF":2.2,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142679323","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-30DOI: 10.1109/LSENS.2024.3488560
Hao Jia;Pere Marti-Puig;Cesar F Caiafa;Moises Serra-Serra;Zhe Sun;Jordi Solé-Casals
The large number of greenhouse gas emissions caused by human activities, and their harmful effect on the earth’s climate, have reached a point where actions are needed. Wind energy is one of the available green energies that can be used to mitigate this problem. Predictive maintenance is of vital importance to ensure continuous wind power generation and is typically based on the use of sensor data from all wind turbine systems. But in some cases, data contain outliers or are not available at all due to sensor or system failures. In this letter, we explore the use of tensor completion methods to estimate missing data in this field. Experimental results demonstrate the usefulness of the proposed tensor completion algorithms, especially the high-accuracy low-rank tensor completion (HaLRTC) method, which outperforms the interpolation method used as a reference.
{"title":"Exploring Tensor Completion for Missing Data Estimation in Wind Farms","authors":"Hao Jia;Pere Marti-Puig;Cesar F Caiafa;Moises Serra-Serra;Zhe Sun;Jordi Solé-Casals","doi":"10.1109/LSENS.2024.3488560","DOIUrl":"https://doi.org/10.1109/LSENS.2024.3488560","url":null,"abstract":"The large number of greenhouse gas emissions caused by human activities, and their harmful effect on the earth’s climate, have reached a point where actions are needed. Wind energy is one of the available green energies that can be used to mitigate this problem. Predictive maintenance is of vital importance to ensure continuous wind power generation and is typically based on the use of sensor data from all wind turbine systems. But in some cases, data contain outliers or are not available at all due to sensor or system failures. In this letter, we explore the use of tensor completion methods to estimate missing data in this field. Experimental results demonstrate the usefulness of the proposed tensor completion algorithms, especially the high-accuracy low-rank tensor completion (HaLRTC) method, which outperforms the interpolation method used as a reference.","PeriodicalId":13014,"journal":{"name":"IEEE Sensors Letters","volume":"8 12","pages":"1-4"},"PeriodicalIF":2.2,"publicationDate":"2024-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142672008","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","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：