Pub Date : 2024-10-21DOI: 10.1109/TIM.2024.3476598
Shaoliang Gong;Jianwu He;Chao Yang;Yumei Wen;Li Duan;Qi Kang;Yixin Ma
Micro-thrusters play a pivotal role in achieving drag-free control in space-based gravitational wave detection missions. Dynamic micro-thrust reconstruction is crucial for comprehensive evaluation and understanding of the operational process of micro-thrusters and for the advancement of higher performance micro-thrusters. The conventional dynamic thrust reconstruction method, inversion of dynamic equation (IODE), heavily relies on the dynamic equation model. This dependence leads to performance decline when discrepancies emerge between the model and the actual system, coupled with a time-consuming parameter estimation process. To overcome these limitations, this article introduces an innovative algorithm breaking free from the constraints of parametric models, named the dynamic matrix control algorithm-based dynamic thrust reconstruction (DMC-DTR) algorithm. In this article, the underlying principles and algorithm process of DMC-DTR are presented. Comparative analyses of DMC-DTR and IODE are conducted through both simulated and real dynamic thrust measurement experiments. The results indicate that DMC-DTR closely matches the performance of IODE when the dynamic equation model aligns with the actual system. Notably, when the model diverges from reality, DMC-DTR outperforms IODE significantly, showing the enhanced universality of the proposed algorithm. In addition, the simulation reveals that DMC-DTR exhibits better performance than IODE at lower sampling frequencies. Moreover, the dynamic thrust from a cold gas micro-thruster is reconstructed by DMC-DTR to demonstrate its practical application.
{"title":"A Dynamic Micro-Thrust Reconstruction Algorithm Independent of Parametric Models","authors":"Shaoliang Gong;Jianwu He;Chao Yang;Yumei Wen;Li Duan;Qi Kang;Yixin Ma","doi":"10.1109/TIM.2024.3476598","DOIUrl":"https://doi.org/10.1109/TIM.2024.3476598","url":null,"abstract":"Micro-thrusters play a pivotal role in achieving drag-free control in space-based gravitational wave detection missions. Dynamic micro-thrust reconstruction is crucial for comprehensive evaluation and understanding of the operational process of micro-thrusters and for the advancement of higher performance micro-thrusters. The conventional dynamic thrust reconstruction method, inversion of dynamic equation (IODE), heavily relies on the dynamic equation model. This dependence leads to performance decline when discrepancies emerge between the model and the actual system, coupled with a time-consuming parameter estimation process. To overcome these limitations, this article introduces an innovative algorithm breaking free from the constraints of parametric models, named the dynamic matrix control algorithm-based dynamic thrust reconstruction (DMC-DTR) algorithm. In this article, the underlying principles and algorithm process of DMC-DTR are presented. Comparative analyses of DMC-DTR and IODE are conducted through both simulated and real dynamic thrust measurement experiments. The results indicate that DMC-DTR closely matches the performance of IODE when the dynamic equation model aligns with the actual system. Notably, when the model diverges from reality, DMC-DTR outperforms IODE significantly, showing the enhanced universality of the proposed algorithm. In addition, the simulation reveals that DMC-DTR exhibits better performance than IODE at lower sampling frequencies. Moreover, the dynamic thrust from a cold gas micro-thruster is reconstructed by DMC-DTR to demonstrate its practical application.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-12"},"PeriodicalIF":5.6,"publicationDate":"2024-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142517977","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 : 2024-10-21DOI: 10.1109/TIM.2024.3472910
Jaewook Oh;Inhwan Kim;Inhyeok Hwang;Bowook Choi;Namsu Kim
Prognostics and health monitoring of insulated gate bipolar transistors (IGBTs) is a primary concern when determining the reliability of inverter systems. The rising popularity of electric vehicles (EVs) has increased the investigations on the reliability of power modules, particularly that of IGBTs. Condition monitoring of IGBT can be performed using various IGBT characteristic parameters, such as collector-emitter saturation voltage ( ${V} _{text {ce,sat}}$ ), collector current ( ${I} _{text {C}}$ ), and gate-emitter threshold voltage ( ${V} _{text {ge,th}}$ ). This study proposes a programmable method for detecting and pinpointing the location of bond-wire lift-off without accessing the gate signal or the collector and emitter terminals of the targeted IGBT chip or freewheeling diode (FWD). The methodology of collecting and processing the collector-emitter voltage data from the three-phase motor phase terminal voltage is discussed. The proposed approach is validated through simulations of a motor drive system, and its adaptability and sensitivity are confirmed through fault emulation tests and power cycling tests in actual motor drive systems.
{"title":"Programmable Online Bond-Wire Fault Detection and Location Method for Insulated Gate Bipolar Transistor Using Inverter Output Parameters","authors":"Jaewook Oh;Inhwan Kim;Inhyeok Hwang;Bowook Choi;Namsu Kim","doi":"10.1109/TIM.2024.3472910","DOIUrl":"https://doi.org/10.1109/TIM.2024.3472910","url":null,"abstract":"Prognostics and health monitoring of insulated gate bipolar transistors (IGBTs) is a primary concern when determining the reliability of inverter systems. The rising popularity of electric vehicles (EVs) has increased the investigations on the reliability of power modules, particularly that of IGBTs. Condition monitoring of IGBT can be performed using various IGBT characteristic parameters, such as collector-emitter saturation voltage (\u0000<inline-formula> <tex-math>${V} _{text {ce,sat}}$ </tex-math></inline-formula>\u0000), collector current (\u0000<inline-formula> <tex-math>${I} _{text {C}}$ </tex-math></inline-formula>\u0000), and gate-emitter threshold voltage (\u0000<inline-formula> <tex-math>${V} _{text {ge,th}}$ </tex-math></inline-formula>\u0000). This study proposes a programmable method for detecting and pinpointing the location of bond-wire lift-off without accessing the gate signal or the collector and emitter terminals of the targeted IGBT chip or freewheeling diode (FWD). The methodology of collecting and processing the collector-emitter voltage data from the three-phase motor phase terminal voltage is discussed. The proposed approach is validated through simulations of a motor drive system, and its adaptability and sensitivity are confirmed through fault emulation tests and power cycling tests in actual motor drive systems.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-8"},"PeriodicalIF":5.6,"publicationDate":"2024-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142518051","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 : 2024-10-18DOI: 10.1109/TIM.2024.3472790
Jun Jiang;Yaqian He;Yu Song;Qiang Wu;Qian Wang
Medium voltage power electronic transformer (PET), composed of power electronics modules and a high-frequency transformer (HFT), plays a significant role in the hybrid ac/dc distribution power grid, with renewable energy sources and electrified loads. Stressed by the high frequency, high voltage, and high temperature, the insulation of HFT is the weak point. However, it is difficult to get access to the discharge signal under severe electromagnetic interference (EMI) due to the fast-switching of the power electronics modules. An optical interference-based partial discharge (PD) detection is proposed and verified to evaluate the insulation status of HFT and PET. In this article, the method of detecting and extracting PD signals under high-voltage and high-frequency pulse waveform is studied by using a skeleton optical fiber sensor (OFS), Sagnac interference structure, and noncorrelation coefficient algorithm. The algorithm distinguishes PD activities from interference signals by analyzing the correlation of signals. The ultrasonic interference characteristics of frequency below 20 kHz and pulse voltage rise time above 50 ns are analyzed. The results demonstrate that the main frequency range of the disturbing signal is from 20 to 100 kHz, which overlaps with the frequency distribution of the PD signal. The noncorrelation coefficient between the signals is proportional to the PD strength. The PD signal is detected successfully on the HFT prototype. It is expected to provide a convenient approach to insulation detection and evaluation of high power, high frequency, and high voltage power apparatus.
{"title":"Optical Ultrasonic Detection for Partial Discharge Under Pulsed High Voltage","authors":"Jun Jiang;Yaqian He;Yu Song;Qiang Wu;Qian Wang","doi":"10.1109/TIM.2024.3472790","DOIUrl":"https://doi.org/10.1109/TIM.2024.3472790","url":null,"abstract":"Medium voltage power electronic transformer (PET), composed of power electronics modules and a high-frequency transformer (HFT), plays a significant role in the hybrid ac/dc distribution power grid, with renewable energy sources and electrified loads. Stressed by the high frequency, high voltage, and high temperature, the insulation of HFT is the weak point. However, it is difficult to get access to the discharge signal under severe electromagnetic interference (EMI) due to the fast-switching of the power electronics modules. An optical interference-based partial discharge (PD) detection is proposed and verified to evaluate the insulation status of HFT and PET. In this article, the method of detecting and extracting PD signals under high-voltage and high-frequency pulse waveform is studied by using a skeleton optical fiber sensor (OFS), Sagnac interference structure, and noncorrelation coefficient algorithm. The algorithm distinguishes PD activities from interference signals by analyzing the correlation of signals. The ultrasonic interference characteristics of frequency below 20 kHz and pulse voltage rise time above 50 ns are analyzed. The results demonstrate that the main frequency range of the disturbing signal is from 20 to 100 kHz, which overlaps with the frequency distribution of the PD signal. The noncorrelation coefficient between the signals is proportional to the PD strength. The PD signal is detected successfully on the HFT prototype. It is expected to provide a convenient approach to insulation detection and evaluation of high power, high frequency, and high voltage power apparatus.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-8"},"PeriodicalIF":5.6,"publicationDate":"2024-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142595941","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 : 2024-10-18DOI: 10.1109/TIM.2024.3472779
Jaehoon Jung;Jihye Kim;Seong-Cheol Kim;Sohee Lim
In this article, we propose an eye-gaze tracking method based on head orientation estimation that uses a single 60-GHz frequency-modulated continuous-wave (FMCW) radar sensor. The FMCW radar data are acquired for cases, in which the radar is illuminating the front or side of the face. Because the variation in facial muscles caused by eye blinking is more pronounced when a human gazes at the radar, the orientation of the human head can be estimated by analyzing the received radar signal. First, an approximate range-angle map is generated to identify whether a human exists. When a human is detected, the received signal is projected onto the lower subspace of interest. Subsequently, a super-resolution algorithm is applied to the projected signal to obtain a precise target spectrum. The accumulated spectrogram is used as input to MobileNet to classify radar signal images corresponding to different orientations of the human head. The classification results show that the proposed method can identify the orientation of a human head with an accuracy exceeding 90%.
{"title":"Eye-Gaze Tracking Based on Head Orientation Estimation Using FMCW Radar Sensor","authors":"Jaehoon Jung;Jihye Kim;Seong-Cheol Kim;Sohee Lim","doi":"10.1109/TIM.2024.3472779","DOIUrl":"https://doi.org/10.1109/TIM.2024.3472779","url":null,"abstract":"In this article, we propose an eye-gaze tracking method based on head orientation estimation that uses a single 60-GHz frequency-modulated continuous-wave (FMCW) radar sensor. The FMCW radar data are acquired for cases, in which the radar is illuminating the front or side of the face. Because the variation in facial muscles caused by eye blinking is more pronounced when a human gazes at the radar, the orientation of the human head can be estimated by analyzing the received radar signal. First, an approximate range-angle map is generated to identify whether a human exists. When a human is detected, the received signal is projected onto the lower subspace of interest. Subsequently, a super-resolution algorithm is applied to the projected signal to obtain a precise target spectrum. The accumulated spectrogram is used as input to MobileNet to classify radar signal images corresponding to different orientations of the human head. The classification results show that the proposed method can identify the orientation of a human head with an accuracy exceeding 90%.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-10"},"PeriodicalIF":5.6,"publicationDate":"2024-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142518093","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 : 2024-10-17DOI: 10.1109/TIM.2024.3472901
Huihui Li;Yulin Feng;Yi Zhao;Qianru Wen;Hang Liu
Pointer instruments are widely used in various fields due to their simplicity, low cost, strong anti-interference capabilities, and durability. However, manual reading and recognition of these instruments can be complex and labor-intensive. Existing meter recognition algorithms have limitations in dial extraction and reading calculation, which restrict their practical applications. In this article, we propose an instrument detection method based on arbitrary quadrilateral positioning. It directly locates key areas such as pointers and scales, eliminating irrelevant information and simplifying the reading process. By applying perspective transformation to the detected quadrilateral, the instrument is transformed into a front-view state, eliminating the need for time-consuming postprocessing operations. We also propose a reading algorithm that handles missing scales, improving reading accuracy. Experimental results demonstrate that our method outperforms existing algorithms in terms of detection and reading accuracy. Using a self-built instrument dataset, the mean average precision (mAP) is 92.4, and the absolute error of the readings is less than 0.02. It is particularly effective in challenging scenarios involving tilt, rotation, changing lighting, and small targets, making it suitable for real-world applications.
{"title":"Instrument Detection and Reading Based on Arbitrary Quadrilateral","authors":"Huihui Li;Yulin Feng;Yi Zhao;Qianru Wen;Hang Liu","doi":"10.1109/TIM.2024.3472901","DOIUrl":"https://doi.org/10.1109/TIM.2024.3472901","url":null,"abstract":"Pointer instruments are widely used in various fields due to their simplicity, low cost, strong anti-interference capabilities, and durability. However, manual reading and recognition of these instruments can be complex and labor-intensive. Existing meter recognition algorithms have limitations in dial extraction and reading calculation, which restrict their practical applications. In this article, we propose an instrument detection method based on arbitrary quadrilateral positioning. It directly locates key areas such as pointers and scales, eliminating irrelevant information and simplifying the reading process. By applying perspective transformation to the detected quadrilateral, the instrument is transformed into a front-view state, eliminating the need for time-consuming postprocessing operations. We also propose a reading algorithm that handles missing scales, improving reading accuracy. Experimental results demonstrate that our method outperforms existing algorithms in terms of detection and reading accuracy. Using a self-built instrument dataset, the mean average precision (mAP) is 92.4, and the absolute error of the readings is less than 0.02. It is particularly effective in challenging scenarios involving tilt, rotation, changing lighting, and small targets, making it suitable for real-world applications.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-9"},"PeriodicalIF":5.6,"publicationDate":"2024-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142595878","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 : 2024-10-17DOI: 10.1109/TIM.2024.3481554
Kai Zhong;Jiefei Yu;Song Zhu;Xiaoming Zhang
Capturing the root cause is crucial for ensuring the safety and efficiency of industrial processes. Nevertheless, most of the existing methods are unavailable for resultful causality mining, particularly those exhibiting nonuniform characteristics. Besides, insufficient utilization of process knowledge and data leads to suboptimal model performance. To fill this gap, a unified root cause identification framework with collaborative reasoning capacity is proposed for nonuniform processes. Initially, the nonuniform dynamic time warping (NDTW) and maximal relevance minimal redundancy (mRMR) methods are designed to acquire highly uniform data with less computational burden. Subsequently, the semantic-aided path search algorithm is proposed for gaining the refined knowledge causal graph (RKCG), enhancing the interpretability and providing a preliminary reference for subsequent causality identification. After that, integrating the highly uniform data and RKCG was into the advantageous attention-causal-aware gated recurrent unit (ACGRU) model for a more reliable causality matrix. Then, the comprehensive causal graph is established with the aid of knowledge-data collaborative reasoning. Finally, the proposed method is verified by ASHRAE RP-1043 centrifugal chiller and real-world coal mill fault datasets.
{"title":"A Unified Framework With Collaborative Reasoning Capacity for Nonuniform Industrial Processes Root Cause Identification","authors":"Kai Zhong;Jiefei Yu;Song Zhu;Xiaoming Zhang","doi":"10.1109/TIM.2024.3481554","DOIUrl":"https://doi.org/10.1109/TIM.2024.3481554","url":null,"abstract":"Capturing the root cause is crucial for ensuring the safety and efficiency of industrial processes. Nevertheless, most of the existing methods are unavailable for resultful causality mining, particularly those exhibiting nonuniform characteristics. Besides, insufficient utilization of process knowledge and data leads to suboptimal model performance. To fill this gap, a unified root cause identification framework with collaborative reasoning capacity is proposed for nonuniform processes. Initially, the nonuniform dynamic time warping (NDTW) and maximal relevance minimal redundancy (mRMR) methods are designed to acquire highly uniform data with less computational burden. Subsequently, the semantic-aided path search algorithm is proposed for gaining the refined knowledge causal graph (RKCG), enhancing the interpretability and providing a preliminary reference for subsequent causality identification. After that, integrating the highly uniform data and RKCG was into the advantageous attention-causal-aware gated recurrent unit (ACGRU) model for a more reliable causality matrix. Then, the comprehensive causal graph is established with the aid of knowledge-data collaborative reasoning. Finally, the proposed method is verified by ASHRAE RP-1043 centrifugal chiller and real-world coal mill fault datasets.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-11"},"PeriodicalIF":5.6,"publicationDate":"2024-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142565530","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 : 2024-10-16DOI: 10.1109/TIM.2024.3481537
Minjie Bao;Junguo Fan;Zhendong Fan;Runze Xu;Ke Wang;Chaonan Mu;Ruifeng Li;Hewen Zhou;Peng Kang
Existing voxel-hashing (VH)-based dense reconstruction methods have shown impressive results on datasets collected by hand-held cameras. Large-scale scenes are maintained with a truncated signed distance function (TSDF) volumetric representation. However, practically deploying such methods on low-cost embedded mobile robots remains challenging due to heavy computational burdens and various camera perception degeneration cases. In this work, we propose FROMFusion, a fast and robust on-manifold dense reconstruction framework based on multisensor fusion, which systematically solves how to align the point cloud with the hashed TSDF volume (HTV). Its purely geometric nature ensures the robustness to image motion blur and poor lighting conditions. To reduce memory overhead, we propose a spherical-coordinate-based HTV segmentation algorithm. To surmount missing geometric features, camera occlusion, and over range, a loosely coupled LiDAR-wheel-inertial odometry (LWIO) is applied for trustworthy initial guesses in camera pose optimization. At its core is a two-stage depth-to-HTV matching algorithm, which includes a coarse voxel-level pose ergodic search and a fine subvoxel-level Gauss-Newton (GN) solver with Anderson acceleration (AA) strategy for faster convergence. We evenly distribute heavy computational workloads to heterogeneous computing systems. Extensive field experiments on a low-cost wheeled robot cleaner demonstrate our method models continuous surfaces of large-scale scenes with high quality in both geometry and texture, outperforming current state-of-the-art methods in robustness to camera perception degeneration cases by a significant margin. The frame rate of online embedded implementation can reach up to 47.21 Hz maximum.
{"title":"FROMFusion: Fast and Robust On-Manifold Dense Reconstruction for Low-Cost Wheeled Robots","authors":"Minjie Bao;Junguo Fan;Zhendong Fan;Runze Xu;Ke Wang;Chaonan Mu;Ruifeng Li;Hewen Zhou;Peng Kang","doi":"10.1109/TIM.2024.3481537","DOIUrl":"https://doi.org/10.1109/TIM.2024.3481537","url":null,"abstract":"Existing voxel-hashing (VH)-based dense reconstruction methods have shown impressive results on datasets collected by hand-held cameras. Large-scale scenes are maintained with a truncated signed distance function (TSDF) volumetric representation. However, practically deploying such methods on low-cost embedded mobile robots remains challenging due to heavy computational burdens and various camera perception degeneration cases. In this work, we propose FROMFusion, a fast and robust on-manifold dense reconstruction framework based on multisensor fusion, which systematically solves how to align the point cloud with the hashed TSDF volume (HTV). Its purely geometric nature ensures the robustness to image motion blur and poor lighting conditions. To reduce memory overhead, we propose a spherical-coordinate-based HTV segmentation algorithm. To surmount missing geometric features, camera occlusion, and over range, a loosely coupled LiDAR-wheel-inertial odometry (LWIO) is applied for trustworthy initial guesses in camera pose optimization. At its core is a two-stage depth-to-HTV matching algorithm, which includes a coarse voxel-level pose ergodic search and a fine subvoxel-level Gauss-Newton (GN) solver with Anderson acceleration (AA) strategy for faster convergence. We evenly distribute heavy computational workloads to heterogeneous computing systems. Extensive field experiments on a low-cost wheeled robot cleaner demonstrate our method models continuous surfaces of large-scale scenes with high quality in both geometry and texture, outperforming current state-of-the-art methods in robustness to camera perception degeneration cases by a significant margin. The frame rate of online embedded implementation can reach up to 47.21 Hz maximum.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-18"},"PeriodicalIF":5.6,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142524161","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}
Obtaining current fault data for mechanical equipment is a challenging endeavor. Despite some successes in anomaly detection, achieving satisfactory results remains difficult, particularly when dealing with datasets containing few instances of anomalies and significant distribution differences. To address this challenge, a novel deep residual Wasserstein generative adversarial network, named RWGAN, is designed to effectively detect anomalies of unseen samples in rotary machines under varying working conditions. Initially, an encoder-decoder–encoder pipeline is constructed based on the convolutional autoencoder (AE) module to extract deep feature representations from the time-frequency transformation of vibration signals. In addition, the ResNet structure with skip connections is embedded into the model to enhance feature learning and model performance. Furthermore, a Wasserstein distance module is developed, integrating loss-specific feature learning networks and adversarial training techniques to address large distribution discrepancies across data from varying working conditions. Finally, the network is updated in an end-to-end manner to generate real-like output by fitting the probability distribution of time-frequency images. To validate the effectiveness and superiority of the proposed method, three cases across 15 tasks under varying working conditions are designed. The results demonstrate that the proposed approach achieves satisfactory anomaly detection performance and outperforms other state-of-the-art methods.
{"title":"Time-Frequency RWGAN for Machine Anomaly Detection Under Varying Working Conditions","authors":"Haiyang Wan;Weihua Li;Jian Jiao;Chuanpeng Ji;Weidong Xu;Yi He;Zhuyun Chen","doi":"10.1109/TIM.2024.3481539","DOIUrl":"https://doi.org/10.1109/TIM.2024.3481539","url":null,"abstract":"Obtaining current fault data for mechanical equipment is a challenging endeavor. Despite some successes in anomaly detection, achieving satisfactory results remains difficult, particularly when dealing with datasets containing few instances of anomalies and significant distribution differences. To address this challenge, a novel deep residual Wasserstein generative adversarial network, named RWGAN, is designed to effectively detect anomalies of unseen samples in rotary machines under varying working conditions. Initially, an encoder-decoder–encoder pipeline is constructed based on the convolutional autoencoder (AE) module to extract deep feature representations from the time-frequency transformation of vibration signals. In addition, the ResNet structure with skip connections is embedded into the model to enhance feature learning and model performance. Furthermore, a Wasserstein distance module is developed, integrating loss-specific feature learning networks and adversarial training techniques to address large distribution discrepancies across data from varying working conditions. Finally, the network is updated in an end-to-end manner to generate real-like output by fitting the probability distribution of time-frequency images. To validate the effectiveness and superiority of the proposed method, three cases across 15 tasks under varying working conditions are designed. The results demonstrate that the proposed approach achieves satisfactory anomaly detection performance and outperforms other state-of-the-art methods.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-11"},"PeriodicalIF":5.6,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142565612","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 : 2024-10-16DOI: 10.1109/TIM.2024.3481552
Yuhong Liu;Chang Wu;Xiaotong Kong;Xiaoqi Du;Yuan You
Nowadays, smartphone-based localization, which is the most common way for car drivers to navigate, has attracted considerable attention. In environments where global positioning system (GPS) signals are obstructed, such as indoor parking lots and large tunnels, smartphone-based localization methods still face challenges in accuracy and update frequency. Current methods utilize inertial measurement units (IMUs) to achieve short-term, high-accuracy position estimation. However, most smartphones are equipped with low-precision IMU due to cost constraints, leading to significant cumulative errors over time caused by sensor noise and external interference. Moreover, indoor positioning methods based on Bluetooth low energy (BLE) can achieve long-term and higher precision positioning estimation. However, due to the limitation of smartphone power consumption, real-time scanning and response of BLE cannot be achieved. BLE scans can only be returned after the scanning cycle is completed. Therefore, when the vehicle is moving at high speed, the update frequency of the BLE-based positioning method is lower. Furthermore, due to the asynchronous reception of BLE, motion distortion will occur, seriously affecting positioning accuracy. To address the above problems, we propose a tightly coupled indoor vehicle positioning and navigation algorithm based on smartphone IMU and BLE, named the CFBS (Synthesis of Coordinate transformation and Forward-Backward propagation) algorithm. The coordinate transformation algorithm converts the inertial dynamics from the smartphone to the vehicle, and the backward propagation algorithm compensates for BLE motion distortion and enhances positioning frequency. Finally, IMU bias and position estimation are achieved using a tightly coupled extended Kalman filter (EKF). Our algorithm is tested in real business scenarios to verify its effect, which effectively improves the positioning frequency and accuracy of indoor positioning services (ILBS).
如今，基于智能手机的定位技术已引起广泛关注，它是汽车驾驶员最常用的导航方式。在全球定位系统（GPS）信号受阻的环境中，如室内停车场和大型隧道，基于智能手机的定位方法在精度和更新频率方面仍面临挑战。目前的方法利用惯性测量单元（IMU）来实现短期、高精度的位置估计。然而，由于成本限制，大多数智能手机都配备了低精度的惯性测量单元（IMU），这导致传感器噪声和外部干扰会随着时间的推移产生显著的累积误差。此外，基于蓝牙低功耗（BLE）的室内定位方法可以实现长期和更高精度的定位估算。然而，由于智能手机功耗的限制，BLE 无法实现实时扫描和响应。BLE 扫描只能在扫描周期结束后返回。因此，当车辆高速行驶时，基于 BLE 的定位方法的更新频率较低。此外，由于 BLE 的异步接收，会出现运动失真，严重影响定位精度。针对上述问题，我们提出了一种基于智能手机 IMU 和 BLE 的紧密耦合的室内车辆定位和导航算法，命名为 CFBS（坐标变换和前向-后向传播合成）算法。坐标转换算法将智能手机的惯性动态转换为车辆的惯性动态，后向传播算法补偿 BLE 运动失真并提高定位频率。最后，使用紧密耦合的扩展卡尔曼滤波器（EKF）实现 IMU 偏差和位置估计。我们的算法在实际业务场景中进行了测试，验证了其效果，有效提高了室内定位服务（ILBS）的定位频率和精度。
{"title":"A Tightly Coupled Method for Indoor Vehicle Navigation Based on Smartphone IMU and BLE","authors":"Yuhong Liu;Chang Wu;Xiaotong Kong;Xiaoqi Du;Yuan You","doi":"10.1109/TIM.2024.3481552","DOIUrl":"https://doi.org/10.1109/TIM.2024.3481552","url":null,"abstract":"Nowadays, smartphone-based localization, which is the most common way for car drivers to navigate, has attracted considerable attention. In environments where global positioning system (GPS) signals are obstructed, such as indoor parking lots and large tunnels, smartphone-based localization methods still face challenges in accuracy and update frequency. Current methods utilize inertial measurement units (IMUs) to achieve short-term, high-accuracy position estimation. However, most smartphones are equipped with low-precision IMU due to cost constraints, leading to significant cumulative errors over time caused by sensor noise and external interference. Moreover, indoor positioning methods based on Bluetooth low energy (BLE) can achieve long-term and higher precision positioning estimation. However, due to the limitation of smartphone power consumption, real-time scanning and response of BLE cannot be achieved. BLE scans can only be returned after the scanning cycle is completed. Therefore, when the vehicle is moving at high speed, the update frequency of the BLE-based positioning method is lower. Furthermore, due to the asynchronous reception of BLE, motion distortion will occur, seriously affecting positioning accuracy. To address the above problems, we propose a tightly coupled indoor vehicle positioning and navigation algorithm based on smartphone IMU and BLE, named the CFBS (Synthesis of Coordinate transformation and Forward-Backward propagation) algorithm. The coordinate transformation algorithm converts the inertial dynamics from the smartphone to the vehicle, and the backward propagation algorithm compensates for BLE motion distortion and enhances positioning frequency. Finally, IMU bias and position estimation are achieved using a tightly coupled extended Kalman filter (EKF). Our algorithm is tested in real business scenarios to verify its effect, which effectively improves the positioning frequency and accuracy of indoor positioning services (ILBS).","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-12"},"PeriodicalIF":5.6,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142595887","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 : 2024-10-16DOI: 10.1109/TIM.2024.3481535
Wen Liu;Lishuai Liu;Yanxun Xiang;Fu-Zhen Xuan
As the core component of key equipment in the fields of aerospace, military, and advanced energy, the multilayer heterogeneous metal-bonded structures (MLHMBSs) are required to serve for a long time in harsh environments such as extreme temperature and stress. Detecting potential damages in MLHMBS is crucial to ensure the reliability of key equipment and prevent further catastrophic accidents. In this regard, a major category of nondestructive testing techniques based on ultrasonic guided waves has gained increasing popularity for damage evaluation of various industrial structures. Despite this, it remains an ongoing challenge to inspect MLHMBS using ultrasonic guided waves due to the complicated wave propagation characteristics induced by the serious impedance mismatch within the multilayer heterogeneous structures. To address this problem, a reconstruction algorithm for probabilistic inspection of defects (RAPID) based on the hierarchical energy distribution is proposed to detect and locate the defects of MLHMBS. Compared with the previous traditional energy-based damage index (DI), the method proposed in this article enhances the response of defects in the structure to the DI and effectively avoids missed and erroneous defect detection because it takes into account the uneven energy distribution between layers and the energy distribution law of the frequency band. From the validation experiments, the proposed approach enables precise damage localization in MLHMBS using Lamb waves, promoting the preferable performance of ultrasonic guided wave techniques in practical industrial applications.
作为航空航天、军事和先进能源领域关键设备的核心部件，多层异质金属结合结构（MLHMBS）需要在极端温度和应力等恶劣环境中长期服役。检测多层异质金属结合结构中的潜在损坏对于确保关键设备的可靠性和防止进一步的灾难性事故至关重要。在这方面，基于超声导波的一大类无损检测技术在各种工业结构的损坏评估中越来越受欢迎。尽管如此，由于多层异质结构内部存在严重的阻抗失配，导致波的传播特性复杂，因此使用超声导波检测多层异质结构仍是一项持续的挑战。针对这一问题，提出了一种基于分层能量分布的缺陷概率检测重构算法（RAPID）来检测和定位 MLHMBS 的缺陷。与以往传统的基于能量的损伤指数（DI）相比，本文提出的方法由于考虑了层间能量分布的不均匀性和频带的能量分布规律，增强了结构中缺陷对 DI 的响应，有效避免了缺陷检测的漏检和误检。从验证实验来看，本文提出的方法能利用 Lamb 波对 MLHMBS 中的损伤进行精确定位，促进了超声导波技术在实际工业应用中的优越性能。
{"title":"Hierarchical Energy Distribution-Based Lamb Wave Tomography for Damage Localization in Multilayer Heterogeneous Metallic Bonded Structures","authors":"Wen Liu;Lishuai Liu;Yanxun Xiang;Fu-Zhen Xuan","doi":"10.1109/TIM.2024.3481535","DOIUrl":"https://doi.org/10.1109/TIM.2024.3481535","url":null,"abstract":"As the core component of key equipment in the fields of aerospace, military, and advanced energy, the multilayer heterogeneous metal-bonded structures (MLHMBSs) are required to serve for a long time in harsh environments such as extreme temperature and stress. Detecting potential damages in MLHMBS is crucial to ensure the reliability of key equipment and prevent further catastrophic accidents. In this regard, a major category of nondestructive testing techniques based on ultrasonic guided waves has gained increasing popularity for damage evaluation of various industrial structures. Despite this, it remains an ongoing challenge to inspect MLHMBS using ultrasonic guided waves due to the complicated wave propagation characteristics induced by the serious impedance mismatch within the multilayer heterogeneous structures. To address this problem, a reconstruction algorithm for probabilistic inspection of defects (RAPID) based on the hierarchical energy distribution is proposed to detect and locate the defects of MLHMBS. Compared with the previous traditional energy-based damage index (DI), the method proposed in this article enhances the response of defects in the structure to the DI and effectively avoids missed and erroneous defect detection because it takes into account the uneven energy distribution between layers and the energy distribution law of the frequency band. From the validation experiments, the proposed approach enables precise damage localization in MLHMBS using Lamb waves, promoting the preferable performance of ultrasonic guided wave techniques in practical industrial applications.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-9"},"PeriodicalIF":5.6,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142595912","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：