Pub Date : 2025-04-22DOI: 10.1109/TIM.2025.3555701
Dongming Li;Jiamei Wang;Han Wang;Xiaojiang Huang;Jiejie Yang;Yueming Gao;Hung-Chun Li;Mang I Vai;Sio Hang Pun
Conductive intracardiac communication (CIC) is an essential approach for achieving multichamber pacing in leadless pacemakers (LCPs), significantly enhancing the therapeutic outcomes for conditions, such as bradycardia. However, the characteristics of the intracardiac channel are profoundly affected by the heart’s rhythmic contractions. Accurately understanding the dynamic transmission mechanisms and channel parameters under various cardiac pathological states is crucial for enhancing the multichamber pacing functionality of LCPs. In this article, the relationship between cardiac chamber volume and channel impedance is mapped based on the electrocardiogram (ECG) data. This mapping enables precise, real-time adjustments to variable impedance, simulating the impedance changes occurring with each heartbeat. Through this approach, a time-frequency equivalent circuit phantom is proposed to accurately simulate channel characteristics for various pacemaker indications (PIs). Utilizing a quasi-dual-pump structural analogy to the heart, we designed a dynamic experimental measurement platform capable of simulating the cardiac beating process under various PIs, which is employed to validate the accuracy of the circuit phantom. The results demonstrate that the correlation coefficients in the frequency and time domains are greater than 0.9432 and 0.9150, respectively, with a time-domain consistency coefficient of less than 3.25. Through cross validation in both frequency and time domains, the circuit effectively simulates the channel characteristics of normal and PI hearts. The empirical formula established based on the time-domain measurement results can be utilized for the rapid estimation of the right atrium (RA)–right ventricle (RV) channel characteristics. The proposed phantom offers a highly accurate and reproducible experimental method for the design of intracardiac communication transceivers, advancing the development and validation of leadless multichamber pacemaker systems.
{"title":"A Dynamic High-Fidelity Equivalent Circuit Phantom for Intracardiac Communication in Pacemaker Indications","authors":"Dongming Li;Jiamei Wang;Han Wang;Xiaojiang Huang;Jiejie Yang;Yueming Gao;Hung-Chun Li;Mang I Vai;Sio Hang Pun","doi":"10.1109/TIM.2025.3555701","DOIUrl":"https://doi.org/10.1109/TIM.2025.3555701","url":null,"abstract":"Conductive intracardiac communication (CIC) is an essential approach for achieving multichamber pacing in leadless pacemakers (LCPs), significantly enhancing the therapeutic outcomes for conditions, such as bradycardia. However, the characteristics of the intracardiac channel are profoundly affected by the heart’s rhythmic contractions. Accurately understanding the dynamic transmission mechanisms and channel parameters under various cardiac pathological states is crucial for enhancing the multichamber pacing functionality of LCPs. In this article, the relationship between cardiac chamber volume and channel impedance is mapped based on the electrocardiogram (ECG) data. This mapping enables precise, real-time adjustments to variable impedance, simulating the impedance changes occurring with each heartbeat. Through this approach, a time-frequency equivalent circuit phantom is proposed to accurately simulate channel characteristics for various pacemaker indications (PIs). Utilizing a quasi-dual-pump structural analogy to the heart, we designed a dynamic experimental measurement platform capable of simulating the cardiac beating process under various PIs, which is employed to validate the accuracy of the circuit phantom. The results demonstrate that the correlation coefficients in the frequency and time domains are greater than 0.9432 and 0.9150, respectively, with a time-domain consistency coefficient of less than 3.25. Through cross validation in both frequency and time domains, the circuit effectively simulates the channel characteristics of normal and PI hearts. The empirical formula established based on the time-domain measurement results can be utilized for the rapid estimation of the right atrium (RA)–right ventricle (RV) channel characteristics. The proposed phantom offers a highly accurate and reproducible experimental method for the design of intracardiac communication transceivers, advancing the development and validation of leadless multichamber pacemaker systems.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"74 ","pages":"1-12"},"PeriodicalIF":5.6,"publicationDate":"2025-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143870916","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-17DOI: 10.1109/TIM.2025.3557111
Hongpeng Zhang;Xiangming Kan;Chenyong Wang;Zhongyang Cai;Xurui Zhang;Riwei Wang;Chenzhao Bai
As the detection channel of an inductive particle sensor increases, it is inevitable that multiple particles will pass through the sensor simultaneously. However, the movement of multiple particles in fluid and magnetic fields, especially their aggregation behavior, can allow particle sensors to generate misleading wear signals. Therefore, to estimate the influence of the multi-particle aggregation effect on the accuracy of inductive sensors, this study constructed a magnetic coupling model of abrasive particles in strip structure and built the experimental platform required for the study, to investigate the difference in inductance change due to magnetic coupling effect among multiple metal particles at different frequencies. The experimental results reveal the phenomenon of inductance change of the three-particle combinations of the strip structure under two different conditions: in the “sequential entry” type of strip structure, the vortex effect between particles exhibits a mutual weakening, while in the “simultaneous entry” type, the vortex effect between particles exhibits a mutual enhancement. In the “simultaneous entry” type, the eddy current effects between the particles are characterized by mutual enhancement. These findings not only provide a new perspective for understanding the aggregation detection of polymetallic particles but also provide a theoretical basis for improving the accuracy of the sensor.
{"title":"Characterization of Inductive Signals of Polymetallic Particles Under Variable Frequency Conditions","authors":"Hongpeng Zhang;Xiangming Kan;Chenyong Wang;Zhongyang Cai;Xurui Zhang;Riwei Wang;Chenzhao Bai","doi":"10.1109/TIM.2025.3557111","DOIUrl":"https://doi.org/10.1109/TIM.2025.3557111","url":null,"abstract":"As the detection channel of an inductive particle sensor increases, it is inevitable that multiple particles will pass through the sensor simultaneously. However, the movement of multiple particles in fluid and magnetic fields, especially their aggregation behavior, can allow particle sensors to generate misleading wear signals. Therefore, to estimate the influence of the multi-particle aggregation effect on the accuracy of inductive sensors, this study constructed a magnetic coupling model of abrasive particles in strip structure and built the experimental platform required for the study, to investigate the difference in inductance change due to magnetic coupling effect among multiple metal particles at different frequencies. The experimental results reveal the phenomenon of inductance change of the three-particle combinations of the strip structure under two different conditions: in the “sequential entry” type of strip structure, the vortex effect between particles exhibits a mutual weakening, while in the “simultaneous entry” type, the vortex effect between particles exhibits a mutual enhancement. In the “simultaneous entry” type, the eddy current effects between the particles are characterized by mutual enhancement. These findings not only provide a new perspective for understanding the aggregation detection of polymetallic particles but also provide a theoretical basis for improving the accuracy of the sensor.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"74 ","pages":"1-9"},"PeriodicalIF":5.6,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143865372","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The primary challenge facing the field of magnetic anomaly detection (MAD) currently lies in how to effectively improve detection performance in low signal-to-noise ratio (SNR) and real $1/f^{alpha } $ noise scenarios. To overcome these difficulties, this article proposes an optimized MAD method based on a random forest (RF) classifier. This method utilizes an orthonormal basis function (OBF) detector based on Karhunen-Loève expansion (KLE) to extract energy from the raw data as time-domain (TD) features. Meanwhile, the spectral information derived from low-pass filtering (LPF) and fast Fourier transform (FFT) serves as frequency-domain (FD) features of the raw data. The cutoff frequency of the LPF is determined based on a frequency characteristic function that defines the high-frequency boundary of the target signal. Combining these time and FD features, a simulated dataset is constructed for the training and testing of the detection model. Subsequently, the trained model undergoes further validation and evaluation on semi-real and real datasets built upon measured data from a tunneling magnetoresistance (TMR) magnetic sensor. Through a series of simulations, we demonstrate that our designed method exhibits superior detection capability and stronger generalization ability compared to other similar OBF-based methods. Furthermore, the superiority of this method is also confirmed by experimental results based on measured data.
{"title":"Optimization of Magnetic Anomaly Detection Under 1/fα Noise Based on Karhunen–Loève Expansion and Frequency Characteristic Function","authors":"Wenqi Li;Zongtan Zhou;Hongxin Li;Jingsheng Tang;Ming Xu","doi":"10.1109/TIM.2025.3557098","DOIUrl":"https://doi.org/10.1109/TIM.2025.3557098","url":null,"abstract":"The primary challenge facing the field of magnetic anomaly detection (MAD) currently lies in how to effectively improve detection performance in low signal-to-noise ratio (SNR) and real <inline-formula> <tex-math>$1/f^{alpha } $ </tex-math></inline-formula> noise scenarios. To overcome these difficulties, this article proposes an optimized MAD method based on a random forest (RF) classifier. This method utilizes an orthonormal basis function (OBF) detector based on Karhunen-Loève expansion (KLE) to extract energy from the raw data as time-domain (TD) features. Meanwhile, the spectral information derived from low-pass filtering (LPF) and fast Fourier transform (FFT) serves as frequency-domain (FD) features of the raw data. The cutoff frequency of the LPF is determined based on a frequency characteristic function that defines the high-frequency boundary of the target signal. Combining these time and FD features, a simulated dataset is constructed for the training and testing of the detection model. Subsequently, the trained model undergoes further validation and evaluation on semi-real and real datasets built upon measured data from a tunneling magnetoresistance (TMR) magnetic sensor. Through a series of simulations, we demonstrate that our designed method exhibits superior detection capability and stronger generalization ability compared to other similar OBF-based methods. Furthermore, the superiority of this method is also confirmed by experimental results based on measured data.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"74 ","pages":"1-13"},"PeriodicalIF":5.6,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143875180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-16DOI: 10.1109/TIM.2025.3560726
Gui Jin;Wei Zhuang;Mingxin Qin;Feng Wang;Mingsheng Chen;Haocheng Li;Zihong Wang
Currently, electrical capacitance tomography (ECT) is limited to time-differential imaging for monitoring dynamic alterations in cerebral hemorrhage. The inherent constraint of this approach, however, renders it unsuitable for rapid hemorrhage detection, as it requires a reference measurement from a nonhemorrhaging brain. In order to address this limitation, this study proposes a novel approach of frequency-differential ECT (FDECT) for cerebral hemorrhage imaging in practice. The method entails the identification of a frequency range wherein the permittivity variation of cerebral blood with frequency is much greater than the variation of other brain tissues. Within this identified range, two optimal frequencies are selected, and the permittivity difference at these two frequencies is used for imaging. With this method, cerebral hemorrhage is highlighted, and other brain tissues are suppressed, thereby achieving the absolute distribution of cerebral hemorrhage and eliminating the need for nonhemorrhagic baseline data. Simulation results demonstrate that FDECT imaging quality correlates directly with the frequency-dependent permittivity difference between the target and background media, thereby validating FDECT’s theoretical basis and highlighting the critical role of optimal frequency selection. Before conducting in vitro animal imaging, we analyzed the dielectric spectra of ex vivo sheep blood, pig fat, and pig brain tissue to identify the optimal frequency range for differentiating blood from these tissues. In vitro experiments confirmed that FDECT with the optimal frequencies effectively images blood within pig fat or brain tissue, contrasting with the suboptimal results from nonideal frequencies. Although essential for FDECT success, optimal frequency pairing does not eliminate the higher noise levels in FDECT images, largely due to the background brain tissue’s frequency-dependent dielectric characteristics. In order to mitigate this inherent limitation and improve imaging quality, we intend to implement a three-frequency FDECT approach, reducing background tissue interference.
{"title":"Experimental Study of Cerebral Hemorrhage Imaging Based on Frequency-Differential Electrical Capacitance Tomography","authors":"Gui Jin;Wei Zhuang;Mingxin Qin;Feng Wang;Mingsheng Chen;Haocheng Li;Zihong Wang","doi":"10.1109/TIM.2025.3560726","DOIUrl":"https://doi.org/10.1109/TIM.2025.3560726","url":null,"abstract":"Currently, electrical capacitance tomography (ECT) is limited to time-differential imaging for monitoring dynamic alterations in cerebral hemorrhage. The inherent constraint of this approach, however, renders it unsuitable for rapid hemorrhage detection, as it requires a reference measurement from a nonhemorrhaging brain. In order to address this limitation, this study proposes a novel approach of frequency-differential ECT (FDECT) for cerebral hemorrhage imaging in practice. The method entails the identification of a frequency range wherein the permittivity variation of cerebral blood with frequency is much greater than the variation of other brain tissues. Within this identified range, two optimal frequencies are selected, and the permittivity difference at these two frequencies is used for imaging. With this method, cerebral hemorrhage is highlighted, and other brain tissues are suppressed, thereby achieving the absolute distribution of cerebral hemorrhage and eliminating the need for nonhemorrhagic baseline data. Simulation results demonstrate that FDECT imaging quality correlates directly with the frequency-dependent permittivity difference between the target and background media, thereby validating FDECT’s theoretical basis and highlighting the critical role of optimal frequency selection. Before conducting in vitro animal imaging, we analyzed the dielectric spectra of ex vivo sheep blood, pig fat, and pig brain tissue to identify the optimal frequency range for differentiating blood from these tissues. In vitro experiments confirmed that FDECT with the optimal frequencies effectively images blood within pig fat or brain tissue, contrasting with the suboptimal results from nonideal frequencies. Although essential for FDECT success, optimal frequency pairing does not eliminate the higher noise levels in FDECT images, largely due to the background brain tissue’s frequency-dependent dielectric characteristics. In order to mitigate this inherent limitation and improve imaging quality, we intend to implement a three-frequency FDECT approach, reducing background tissue interference.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"74 ","pages":"1-14"},"PeriodicalIF":5.6,"publicationDate":"2025-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143875139","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-16DOI: 10.1109/TIM.2025.3561400
Zhongyu Bai;Hongli Xu;Qichuan Ding;Xiangyue Zhang
Recently, attention-based arbitrary style transfer (AST) techniques have been widely applied in image generation and video processing. However, the scale bias of the attention module used for contextual information extraction and multiscale feature aggregation poses a challenge in balancing the content structure and style patterns of images. In this work, a multiscale channel attention fusion network (MCAFNet) is proposed to generate stylization images with well-coordinated content and style. Specifically, the multiscale channel attention module (MCAM) is introduced to extract both local and global contextual information of style features within the channel dimension and subsequently aggregate this information with content features. Following MCAM, an attentional feature fusion module (AFFM) is adopted to effectively integrate both deep and shallow semantic features. Furthermore, a novel contrastive loss based on multi-source feature enhancement is proposed to optimize the spatial distribution between content and style features. Both qualitative and quantitative experimental results compared to the state-of-the-art (SOTA) baseline approaches indicate the superiority of the proposed method for real-time image and video style transfer.
{"title":"MCAFNet: Multiscale Channel Attention Fusion Network for Arbitrary Style Transfer","authors":"Zhongyu Bai;Hongli Xu;Qichuan Ding;Xiangyue Zhang","doi":"10.1109/TIM.2025.3561400","DOIUrl":"https://doi.org/10.1109/TIM.2025.3561400","url":null,"abstract":"Recently, attention-based arbitrary style transfer (AST) techniques have been widely applied in image generation and video processing. However, the scale bias of the attention module used for contextual information extraction and multiscale feature aggregation poses a challenge in balancing the content structure and style patterns of images. In this work, a multiscale channel attention fusion network (MCAFNet) is proposed to generate stylization images with well-coordinated content and style. Specifically, the multiscale channel attention module (MCAM) is introduced to extract both local and global contextual information of style features within the channel dimension and subsequently aggregate this information with content features. Following MCAM, an attentional feature fusion module (AFFM) is adopted to effectively integrate both deep and shallow semantic features. Furthermore, a novel contrastive loss based on multi-source feature enhancement is proposed to optimize the spatial distribution between content and style features. Both qualitative and quantitative experimental results compared to the state-of-the-art (SOTA) baseline approaches indicate the superiority of the proposed method for real-time image and video style transfer.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"74 ","pages":"1-13"},"PeriodicalIF":5.6,"publicationDate":"2025-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143875106","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Coriolis flowmeters are widely used in the petroleum, chemical, and pharmaceutical industries. To ensure safety and efficiency during production, it is necessary to ensure the measurement accuracy of the flowmeter. The dynamic imbalance of a flowmeter is a key factor that affects the measurement accuracy of the flowmeter. Therefore, a method for evaluating the unbalanced state status of flowmeters is proposed. This study aimed to quickly and accurately identify the dynamic unbalanced state of a flowmeter. First, the frequency response function of the original signal is calculated to reduce the error caused by the influence of excitation. Second, the method of combining empirical wavelet and Laplacian sparse decomposition is used to extract the characteristic information of each order modal. Finally, through experiments, it is found that the second- and third-order modal characteristic details are sensitive to the degree and location of the flowmeter dynamic unbalance fault. Therefore, second- and third-order modal characteristic information were selected to calculate two characteristic indicators that reflect the degree and location of the flowmeter dynamic unbalanced fault to achieve the effect of fault diagnosis. Experiments and simulations show that this method is more accurate and efficient than the other methods.
{"title":"The Unbalanced State Detection of Coriolis Flowmeter Based on Laplace Wavelet Sparse Representation and Customized Detection Indicator","authors":"Yue Si;Yanyi Zhang;Lingfei Kong;Chaohui Zhang;Bohan Zhao","doi":"10.1109/TIM.2025.3558818","DOIUrl":"https://doi.org/10.1109/TIM.2025.3558818","url":null,"abstract":"Coriolis flowmeters are widely used in the petroleum, chemical, and pharmaceutical industries. To ensure safety and efficiency during production, it is necessary to ensure the measurement accuracy of the flowmeter. The dynamic imbalance of a flowmeter is a key factor that affects the measurement accuracy of the flowmeter. Therefore, a method for evaluating the unbalanced state status of flowmeters is proposed. This study aimed to quickly and accurately identify the dynamic unbalanced state of a flowmeter. First, the frequency response function of the original signal is calculated to reduce the error caused by the influence of excitation. Second, the method of combining empirical wavelet and Laplacian sparse decomposition is used to extract the characteristic information of each order modal. Finally, through experiments, it is found that the second- and third-order modal characteristic details are sensitive to the degree and location of the flowmeter dynamic unbalance fault. Therefore, second- and third-order modal characteristic information were selected to calculate two characteristic indicators that reflect the degree and location of the flowmeter dynamic unbalanced fault to achieve the effect of fault diagnosis. Experiments and simulations show that this method is more accurate and efficient than the other methods.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"74 ","pages":"1-10"},"PeriodicalIF":5.6,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143856326","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Continuous motion intention prediction is valuable for human-machine interaction (HMI); however, the accuracy and the efficiency of the existing related results are far from satisfactory. This study proposes a novel continuous motion-intention prediction method that combines multisource information fusion with an improved temporal convolutional neural network (TCN), enhanced by the introduction of the temporal pattern attention (TPA) mechanism. By integrating the temporal features of surface electromyography (sEMG) and mechanomyography (MMG) signals, we fully exploit their synergistic effect in movement intention prediction. Using the TCN network as the continuous motion prediction model improves the training efficiency through parallel computation and a simple network structure. TCN avoids the possible gradient vanishing problem and allows for parameter tuning according to different tasks. By integrating the TPA mechanism with the TCN, the model’s ability to recognize human motion intention in sequential data is significantly improved by focusing on key time steps. This enhancement increases prediction accuracy and strengthens the model’s ability to capture long-term dependencies in time series data. Experiments are conducted to show the effectiveness and the advantages of the proposed TPA-TCN-based motion prediction in comparison with the related results.
{"title":"Multisource Information Fusion for Continuous Prediction of Joint Angles Using TCN Combined With Temporal Pattern Attention Mechanism","authors":"Tairen Sun;Shaozhe Wang;Hongjun Yang;Jiantao Yang;Zeng-Guang Hou","doi":"10.1109/TIM.2025.3560752","DOIUrl":"https://doi.org/10.1109/TIM.2025.3560752","url":null,"abstract":"Continuous motion intention prediction is valuable for human-machine interaction (HMI); however, the accuracy and the efficiency of the existing related results are far from satisfactory. This study proposes a novel continuous motion-intention prediction method that combines multisource information fusion with an improved temporal convolutional neural network (TCN), enhanced by the introduction of the temporal pattern attention (TPA) mechanism. By integrating the temporal features of surface electromyography (sEMG) and mechanomyography (MMG) signals, we fully exploit their synergistic effect in movement intention prediction. Using the TCN network as the continuous motion prediction model improves the training efficiency through parallel computation and a simple network structure. TCN avoids the possible gradient vanishing problem and allows for parameter tuning according to different tasks. By integrating the TPA mechanism with the TCN, the model’s ability to recognize human motion intention in sequential data is significantly improved by focusing on key time steps. This enhancement increases prediction accuracy and strengthens the model’s ability to capture long-term dependencies in time series data. Experiments are conducted to show the effectiveness and the advantages of the proposed TPA-TCN-based motion prediction in comparison with the related results.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"74 ","pages":"1-12"},"PeriodicalIF":5.6,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143875178","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
For aircraft engine predictive maintenance (PdM) programs, accurate remaining useful life (RUL) predictions can significantly reduce unscheduled maintenance downtime and ensure engine safety. To this end, we introduce a novel dual-channel degradation monitoring (DCDM) algorithm designed to minimize RUL prediction errors. Unlike traditional RUL prediction algorithms based on graph neural networks (GNNs), the proposed DCDM model extracts fault features from both node embeddings and changes in graph structures. This dual-channel approach allows for the fusion of fault information, improving the model’s ability to extract features across different fault patterns for RUL prediction. During the graph structure learning (GSL) process, domain-specific knowledge and a dynamic graph learning algorithm are integrated to generate graphs, enhancing the interpretability of graph representation. In addition, by introducing node sparse encoding as a model input, the DCDM model’s capability to discern critical features is significantly improved. The predictive performance of the DCDM model and the effectiveness of its individual components are validated using the C-MAPSS dataset. The results demonstrate the superior accuracy of the proposed method compared to existing approaches.
{"title":"Dual-Channel Degradation Monitoring Based on Graph Neural Network for Aero-Engine Remaining Useful Life Prediction","authors":"Li'ang Cao;Yuanfu Li;Jinwei Chen;Wenjie Wu;Huisheng Zhang","doi":"10.1109/TIM.2025.3560753","DOIUrl":"https://doi.org/10.1109/TIM.2025.3560753","url":null,"abstract":"For aircraft engine predictive maintenance (PdM) programs, accurate remaining useful life (RUL) predictions can significantly reduce unscheduled maintenance downtime and ensure engine safety. To this end, we introduce a novel dual-channel degradation monitoring (DCDM) algorithm designed to minimize RUL prediction errors. Unlike traditional RUL prediction algorithms based on graph neural networks (GNNs), the proposed DCDM model extracts fault features from both node embeddings and changes in graph structures. This dual-channel approach allows for the fusion of fault information, improving the model’s ability to extract features across different fault patterns for RUL prediction. During the graph structure learning (GSL) process, domain-specific knowledge and a dynamic graph learning algorithm are integrated to generate graphs, enhancing the interpretability of graph representation. In addition, by introducing node sparse encoding as a model input, the DCDM model’s capability to discern critical features is significantly improved. The predictive performance of the DCDM model and the effectiveness of its individual components are validated using the C-MAPSS dataset. The results demonstrate the superior accuracy of the proposed method compared to existing approaches.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"74 ","pages":"1-19"},"PeriodicalIF":5.6,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143870982","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-15DOI: 10.1109/TIM.2025.3560754
Dandan Zheng;Jilin Ye;Maosen Wang;Yongtao Chen
In horizontal gas-liquid slug flow, the translational velocity of the liquid slug and its bubble distribution plays a crucial role in guiding pipeline design and understanding the formation and development mechanisms of the liquid slug. This article proposes the use of a single optical fiber probe (OFP) based on fiber optical reflectometer (FOR) to measure the velocity and chord length distribution of dispersed bubbles within the liquid slug. These acquired local parameters are then used to determine the translational slug velocity and local liquid holdup. Experiments under 24 velocity conditions were conducted in a horizontal DN50 pipeline, with the superficial liquid velocity from 0.2 to 0.7 m/s and superficial gas velocity from 1.04 to 8.14 m/s. The wavelet synchrosqueezed transform (WSST) method was applied to extract the frequency of the oscillation signal and to calculate the local velocity. Maximum velocity at the slug head region was identified as the translational velocity of the slug. Compared with Bendiksen’s empirical formula, this method achieved a mean absolute percentage error (MAPE) of 4.38%. Under most velocity conditions, the relative error did not exceed 10%. The trend in local liquid holdup was also consistent with Gregory’s prediction. Subsequently, the FOR technique was used to measure both size and spatial distribution of dispersed bubbles inside a slug. The average chord length of all dispersed bubbles was found to be 1.86 mm, with 93.8% of bubbles exhibiting a chord length between 0 and 5 mm. As superficial gas velocity increased from 3 to 7 m/s, the bubble chord length distribution shifted from 0-4 mm to 0-2 mm, indicating the enhanced bubble fragmentation within the slug at a higher gas velocity. Additionally, 29.8%–39.7% of large dispersed bubbles were concentrated at the slug head and slug tail. Bubble entrainment contributed to the increased large bubble concentration at the slug head, while bubble coalescence at the slug tail led to a reduction in the proportion of small bubbles. Moreover, the influence of gas velocity increase on bubble spatial distribution was analyzed.
{"title":"Translational Velocity and Dispersed Bubble Distribution Measurement in Slug Flow Based on Fiber Optical Reflectometer","authors":"Dandan Zheng;Jilin Ye;Maosen Wang;Yongtao Chen","doi":"10.1109/TIM.2025.3560754","DOIUrl":"https://doi.org/10.1109/TIM.2025.3560754","url":null,"abstract":"In horizontal gas-liquid slug flow, the translational velocity of the liquid slug and its bubble distribution plays a crucial role in guiding pipeline design and understanding the formation and development mechanisms of the liquid slug. This article proposes the use of a single optical fiber probe (OFP) based on fiber optical reflectometer (FOR) to measure the velocity and chord length distribution of dispersed bubbles within the liquid slug. These acquired local parameters are then used to determine the translational slug velocity and local liquid holdup. Experiments under 24 velocity conditions were conducted in a horizontal DN50 pipeline, with the superficial liquid velocity from 0.2 to 0.7 m/s and superficial gas velocity from 1.04 to 8.14 m/s. The wavelet synchrosqueezed transform (WSST) method was applied to extract the frequency of the oscillation signal and to calculate the local velocity. Maximum velocity at the slug head region was identified as the translational velocity of the slug. Compared with Bendiksen’s empirical formula, this method achieved a mean absolute percentage error (MAPE) of 4.38%. Under most velocity conditions, the relative error did not exceed 10%. The trend in local liquid holdup was also consistent with Gregory’s prediction. Subsequently, the FOR technique was used to measure both size and spatial distribution of dispersed bubbles inside a slug. The average chord length of all dispersed bubbles was found to be 1.86 mm, with 93.8% of bubbles exhibiting a chord length between 0 and 5 mm. As superficial gas velocity increased from 3 to 7 m/s, the bubble chord length distribution shifted from 0-4 mm to 0-2 mm, indicating the enhanced bubble fragmentation within the slug at a higher gas velocity. Additionally, 29.8%–39.7% of large dispersed bubbles were concentrated at the slug head and slug tail. Bubble entrainment contributed to the increased large bubble concentration at the slug head, while bubble coalescence at the slug tail led to a reduction in the proportion of small bubbles. Moreover, the influence of gas velocity increase on bubble spatial distribution was analyzed.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"74 ","pages":"1-12"},"PeriodicalIF":5.6,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143871031","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Text spotting in camera-captured document images faces significant challenges, especially with dense text of variable lengths. Existing approaches falter with the long-tailed distribution of word lengths, leading to decreased performance on words with extreme lengths. To address this issue, we present WordLenSpotter, an end-to-end framework incorporating word length awareness to improve detection and recognition across a wide range of word lengths. Our method utilizes a dilated convolutional fusion module in its image encoder and a transformer framework for joint detection and recognition guided by word length priors. Our innovations include a spatial length predictor (SLP) and a length-aware segmentation (LenSeg) proposal head, enhancing the model’s sensitivity to the spatial distribution of text. Evaluated on our newly constructed DSTD1500 dataset and existing public datasets with dense text, WordLenSpotter demonstrates superior text spotting capabilities, especially in handling the diversity of word lengths in dense text scenes. The code is available at https://github.com/unxiaohao/WordLenSpotter
{"title":"Word Length-Aware Text Spotting: Enhancing Dense Text Detection and Recognition for Camera-Captured Document Image","authors":"Hao Wang;Huabing Zhou;Yanduo Zhang;Jiayi Ma;Haibin Ling","doi":"10.1109/TIM.2025.3560748","DOIUrl":"https://doi.org/10.1109/TIM.2025.3560748","url":null,"abstract":"Text spotting in camera-captured document images faces significant challenges, especially with dense text of variable lengths. Existing approaches falter with the long-tailed distribution of word lengths, leading to decreased performance on words with extreme lengths. To address this issue, we present WordLenSpotter, an end-to-end framework incorporating word length awareness to improve detection and recognition across a wide range of word lengths. Our method utilizes a dilated convolutional fusion module in its image encoder and a transformer framework for joint detection and recognition guided by word length priors. Our innovations include a spatial length predictor (SLP) and a length-aware segmentation (LenSeg) proposal head, enhancing the model’s sensitivity to the spatial distribution of text. Evaluated on our newly constructed DSTD1500 dataset and existing public datasets with dense text, WordLenSpotter demonstrates superior text spotting capabilities, especially in handling the diversity of word lengths in dense text scenes. The code is available at <uri>https://github.com/unxiaohao/WordLenSpotter</uri>","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"74 ","pages":"1-15"},"PeriodicalIF":5.6,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143875104","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：