Nowadays, the feature extraction method of multichannel acoustic emission (AE) signal provides a solid research foundation for digital and intelligent fault diagnosis of the hoisting system. More specifically, AE signal collected from the hoisting system is generally characterized by nonlinear and non‐stationary, thus making the traditional intelligent fault diagnosis methods cannot accurately extract the inherent fault features. To alleviate this problem and improve the accuracy of multichannel fault diagnosis, a new fault diagnosis method for hoisting system based on differential search algorithm‐variational mode decomposition (DSA‐VMD) and improved deep convolutional neural network (IDCNN) is proposed in this paper. Specifically, the proposed DSA‐VMD and IDCNN method is divided into two main components: (i) the inside parameters (K, a) of VMD is optimized to effectively extract the multichannel AE fault feature via DSA‐VMD and (ii) the extracted multichannel fault components are fed into the designed IDCNN algorithm to accomplish fault identification automatically. Experimental results from the hoisting system demonstrate the effectiveness of the proposed approach. Additionally, the superiority of the proposed approach has also been verified in extracting fault information and fault identification compared to the other multichannel fault diagnosis methods.
{"title":"Multichannel intelligent fault diagnosis of hoisting system using differential search algorithm‐variational mode decomposition and improved deep convolutional neural network","authors":"Yang Li, Chi-Guhn Lee, Feiyun Xu","doi":"10.1002/stc.3023","DOIUrl":"https://doi.org/10.1002/stc.3023","url":null,"abstract":"Nowadays, the feature extraction method of multichannel acoustic emission (AE) signal provides a solid research foundation for digital and intelligent fault diagnosis of the hoisting system. More specifically, AE signal collected from the hoisting system is generally characterized by nonlinear and non‐stationary, thus making the traditional intelligent fault diagnosis methods cannot accurately extract the inherent fault features. To alleviate this problem and improve the accuracy of multichannel fault diagnosis, a new fault diagnosis method for hoisting system based on differential search algorithm‐variational mode decomposition (DSA‐VMD) and improved deep convolutional neural network (IDCNN) is proposed in this paper. Specifically, the proposed DSA‐VMD and IDCNN method is divided into two main components: (i) the inside parameters (K, a) of VMD is optimized to effectively extract the multichannel AE fault feature via DSA‐VMD and (ii) the extracted multichannel fault components are fed into the designed IDCNN algorithm to accomplish fault identification automatically. Experimental results from the hoisting system demonstrate the effectiveness of the proposed approach. Additionally, the superiority of the proposed approach has also been verified in extracting fault information and fault identification compared to the other multichannel fault diagnosis methods.","PeriodicalId":22049,"journal":{"name":"Structural Control and Health Monitoring","volume":"8 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84151484","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}
Vision‐based displacement measurement for structural health monitoring has gained popularity in recent years but still has several limitations in practical applications. Unmanned aerial vehicles (UAVs) provide opportunities to address the bottleneck problems of camera resolution insufficiency and mounting inconvenience due to their high maneuverability. However, existing methods using UAVs for structural displacement measurement are often complicated to operate due to the use of multiple stationary markers or multiple UAVs. This paper describes a novel vision‐based displacement measurement approach, using only one UAV, along with a motionless laser spot projected from a distance away as a reference. The positions of the marker and the laser spot are precisely calculated using a two‐step strategy, in which a designed black and white marker of known size is applied to the structure for scale definition and precise positioning. The adaptive region of interest (ROI) and adaptive binarization methods are utilized to improve the automatic applicability of the proposed approach with various background and brightness values. In this way, the motion of the UAV parallel and perpendicular to the plane of the structure can be eliminated by the stationary reference laser spot and the constantly updated scaling factors, respectively. The performance of the proposed method is validated on a two‐story frame and a suspension bridge. The results show that the displacement measured using the UAV agrees with the reference data obtained using the laser displacement sensor and the stationary camera, thereby demonstrating the accuracy and feasibility of the proposed method for displacement measurement for small‐ and large‐scale infrastructure.
{"title":"Vision‐based displacement measurement using an unmanned aerial vehicle","authors":"Yitian Han, Gang Wu, Dongming Feng","doi":"10.1002/stc.3025","DOIUrl":"https://doi.org/10.1002/stc.3025","url":null,"abstract":"Vision‐based displacement measurement for structural health monitoring has gained popularity in recent years but still has several limitations in practical applications. Unmanned aerial vehicles (UAVs) provide opportunities to address the bottleneck problems of camera resolution insufficiency and mounting inconvenience due to their high maneuverability. However, existing methods using UAVs for structural displacement measurement are often complicated to operate due to the use of multiple stationary markers or multiple UAVs. This paper describes a novel vision‐based displacement measurement approach, using only one UAV, along with a motionless laser spot projected from a distance away as a reference. The positions of the marker and the laser spot are precisely calculated using a two‐step strategy, in which a designed black and white marker of known size is applied to the structure for scale definition and precise positioning. The adaptive region of interest (ROI) and adaptive binarization methods are utilized to improve the automatic applicability of the proposed approach with various background and brightness values. In this way, the motion of the UAV parallel and perpendicular to the plane of the structure can be eliminated by the stationary reference laser spot and the constantly updated scaling factors, respectively. The performance of the proposed method is validated on a two‐story frame and a suspension bridge. The results show that the displacement measured using the UAV agrees with the reference data obtained using the laser displacement sensor and the stationary camera, thereby demonstrating the accuracy and feasibility of the proposed method for displacement measurement for small‐ and large‐scale infrastructure.","PeriodicalId":22049,"journal":{"name":"Structural Control and Health Monitoring","volume":"225 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89185732","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}
Lijun Wu, Yajin Wang, Xu Lin, Zhicong Chen, Qiao Zheng, Shuying Cheng, P. Lin
Vision‐based displacement measurement is promising and emerging for structural monitoring. However, the accuracy of visual measurement is commonly limited by the resolution of the camera. The super‐resolution (SR) technique can reconstruct high‐resolution images from the corresponding low‐resolution images within the constraints of prior knowledge. Existing SR algorithms mainly focus on improving the overall quality of the image. By contrast, the accurate extraction of the coordinates of feature points is the most important for the visual measurement. Besides, the SR network is usually trained by an artificial dataset whose low‐resolution images are obtained by artificially degrading the corresponding high‐resolution images, instead of those directly captured by cameras. However, this degradation usually is only a simple bicubic downsampling that cannot reflect the real degradation, which will provide inaccurate constraints to the network training. Therefore, this paper proposes a novel SR framework that can significantly preserve the feature coordinators for visual measurement (SRFCP). First, a deep learning‐based SR network that focuses on feature preservation is proposed, which introduces both feature weighted branch and feature preserving loss. Second, an image degradation model is built based on the blur kernel and noise extracted from the images captured in real scene. Experiments on public datasets show that the proposed SRFCP performs well both in terms of the objective evaluation index and the subjective visual effect. Then, a binocular visual measurement platform is set up to measure the distance of adjacent feature points on a chessboard. Lastly, several SR algorithms are evaluated by the improvement they bring to the measurement accuracy. Experimental results show that the distance measurement performance can be significantly improved by the images reconstructed by the SRFCP. Therefore, the proposed SRFCP can accurately reconstruct the high‐resolution images while preserving the features coordinates, which is crucial for the visual measurement in structural monitoring.
{"title":"Deep learning‐based super‐resolution with feature coordinators preservation for vision‐based measurement","authors":"Lijun Wu, Yajin Wang, Xu Lin, Zhicong Chen, Qiao Zheng, Shuying Cheng, P. Lin","doi":"10.1002/stc.3107","DOIUrl":"https://doi.org/10.1002/stc.3107","url":null,"abstract":"Vision‐based displacement measurement is promising and emerging for structural monitoring. However, the accuracy of visual measurement is commonly limited by the resolution of the camera. The super‐resolution (SR) technique can reconstruct high‐resolution images from the corresponding low‐resolution images within the constraints of prior knowledge. Existing SR algorithms mainly focus on improving the overall quality of the image. By contrast, the accurate extraction of the coordinates of feature points is the most important for the visual measurement. Besides, the SR network is usually trained by an artificial dataset whose low‐resolution images are obtained by artificially degrading the corresponding high‐resolution images, instead of those directly captured by cameras. However, this degradation usually is only a simple bicubic downsampling that cannot reflect the real degradation, which will provide inaccurate constraints to the network training. Therefore, this paper proposes a novel SR framework that can significantly preserve the feature coordinators for visual measurement (SRFCP). First, a deep learning‐based SR network that focuses on feature preservation is proposed, which introduces both feature weighted branch and feature preserving loss. Second, an image degradation model is built based on the blur kernel and noise extracted from the images captured in real scene. Experiments on public datasets show that the proposed SRFCP performs well both in terms of the objective evaluation index and the subjective visual effect. Then, a binocular visual measurement platform is set up to measure the distance of adjacent feature points on a chessboard. Lastly, several SR algorithms are evaluated by the improvement they bring to the measurement accuracy. Experimental results show that the distance measurement performance can be significantly improved by the images reconstructed by the SRFCP. Therefore, the proposed SRFCP can accurately reconstruct the high‐resolution images while preserving the features coordinates, which is crucial for the visual measurement in structural monitoring.","PeriodicalId":22049,"journal":{"name":"Structural Control and Health Monitoring","volume":"82 12","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91426448","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}
To improve the applicability of mass dampers in control problems facing different types of uncertainties, asymmetric nonlinear energy sink‐inerters (ANESIs) are proposed. Previous studies reveal that ANESIs have excellent control effect in mitigating impulsive responses but only briefly touch upon their seismic application. This study is intended to provide better understanding in this regard by analyzing the dynamics of the ANESI system when subjected to harmonic and seismic ground excitations. First, the development of ANESIs is described and the theoretical model of ANESIs is updated in accordance with the experimental validation on a three‐story steel‐frame structure. Then analytical investigations are carried out and subsequently a design method is proposed for the ANESI system when subjected to harmonic ground excitation. Finally, 12 seismic ground motions are applied to the ANESI system to assess its control effectiveness, energy and frequency robustness, and space demand. The results show that devices with inerter are in general more effective than those without inerter and the ANESI outperforms its linear and nonlinear counterparts with stronger robustness against changes in structural property and energy level, demonstrating great potential in seismic application.
{"title":"An inerter‐enhanced asymmetric nonlinear energy sink for response mitigation of structures subjected to harmonic and seismic ground excitations","authors":"Jingjing Wang, Chao Zhang, Yuqiang Zheng","doi":"10.1002/stc.3104","DOIUrl":"https://doi.org/10.1002/stc.3104","url":null,"abstract":"To improve the applicability of mass dampers in control problems facing different types of uncertainties, asymmetric nonlinear energy sink‐inerters (ANESIs) are proposed. Previous studies reveal that ANESIs have excellent control effect in mitigating impulsive responses but only briefly touch upon their seismic application. This study is intended to provide better understanding in this regard by analyzing the dynamics of the ANESI system when subjected to harmonic and seismic ground excitations. First, the development of ANESIs is described and the theoretical model of ANESIs is updated in accordance with the experimental validation on a three‐story steel‐frame structure. Then analytical investigations are carried out and subsequently a design method is proposed for the ANESI system when subjected to harmonic ground excitation. Finally, 12 seismic ground motions are applied to the ANESI system to assess its control effectiveness, energy and frequency robustness, and space demand. The results show that devices with inerter are in general more effective than those without inerter and the ANESI outperforms its linear and nonlinear counterparts with stronger robustness against changes in structural property and energy level, demonstrating great potential in seismic application.","PeriodicalId":22049,"journal":{"name":"Structural Control and Health Monitoring","volume":"15 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75787294","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}
This study proposes a novel shape memory alloy (SMA)‐based self‐centering (SC) precast segmental concrete column (PSCC). The entire structural system, called SSC‐PSCC, can be constructed via off‐site precasting followed by on‐site assembly. By using connecting SMA bolts and steel angles, wherein SMA and steel elements contribute to SC and energy dissipation (ED) capacities, respectively, SSC‐PSCC is free from post‐tensioned anchorages and wet joints, and the construction or repair process is considerably simplified. The behavior of SSC‐PSCC was systematically studied through refined finite element models (FEMs) that were verified by good agreements with previous testing results. Cyclic behavior was simulated, and SSC‐PSCC exhibits desirable SC and ED capacities. Compared with traditional post‐tensioning (PT)‐based SC‐PSCC, SSC‐PSCC has numerous advantages, including easy construction, quick repair and replacement of ED elements after earthquakes, and free from buckling of ED bars observed in traditional PT‐based segmental columns. Furthermore, parametric studies were conducted to identify several crucial design parameters, such as prestrain in SMA bolts, axial load ratio, and segment reinforcement diameter. Design recommendations are provided on the basis of the results of the parametric studies.
{"title":"Development of a novel shape memory alloy‐based self‐centering precast segmental concrete column","authors":"Zhi-Peng Chen, Songye Zhu","doi":"10.1002/stc.3099","DOIUrl":"https://doi.org/10.1002/stc.3099","url":null,"abstract":"This study proposes a novel shape memory alloy (SMA)‐based self‐centering (SC) precast segmental concrete column (PSCC). The entire structural system, called SSC‐PSCC, can be constructed via off‐site precasting followed by on‐site assembly. By using connecting SMA bolts and steel angles, wherein SMA and steel elements contribute to SC and energy dissipation (ED) capacities, respectively, SSC‐PSCC is free from post‐tensioned anchorages and wet joints, and the construction or repair process is considerably simplified. The behavior of SSC‐PSCC was systematically studied through refined finite element models (FEMs) that were verified by good agreements with previous testing results. Cyclic behavior was simulated, and SSC‐PSCC exhibits desirable SC and ED capacities. Compared with traditional post‐tensioning (PT)‐based SC‐PSCC, SSC‐PSCC has numerous advantages, including easy construction, quick repair and replacement of ED elements after earthquakes, and free from buckling of ED bars observed in traditional PT‐based segmental columns. Furthermore, parametric studies were conducted to identify several crucial design parameters, such as prestrain in SMA bolts, axial load ratio, and segment reinforcement diameter. Design recommendations are provided on the basis of the results of the parametric studies.","PeriodicalId":22049,"journal":{"name":"Structural Control and Health Monitoring","volume":"11 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84549917","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}
This study proposes a statistical‐based detection method that is responsive to damage and not to environmental and operational conditions. The method serves as a damage recognition system for structural health monitoring using field measurements from real bridges. Vehicle‐induced bridge and ambient vibration measurements collected from the benchmark Old ADA steel truss bridge situated in Japan were utilized to validate the proposed method. The steel truss members in the bridge were subjected to five different damage scenarios to represent common potential problems in structural health monitoring of real‐life applications. The collected measurements have been completely published and made available online. A combination of principal component analysis (PCA) and linear discriminant analysis (LDA) transformation is utilized as a statistical‐based recognition technique. Vibration data representing power spectral density (PSD) functions were tested as damage‐sensitive features from identified condition sources. The proposed combination of the PCA‐LDA transformation system outperforms the popular PCA transformation as a statistical model for classification of state conditions. Although the first two principal components of PCA hold 50–85% of the variation in data, the first two components from PCA‐ LDA hold about 95% of the total variation. As a result, the three PCs, of PCA‐LDA, visualization successfully managed to classify the five structural damage scenarios into their five individual subgroups.
{"title":"Structural damage detection of old ADA steel truss bridge using vibration data","authors":"Ali A. Al‐Ghalib","doi":"10.1002/stc.3098","DOIUrl":"https://doi.org/10.1002/stc.3098","url":null,"abstract":"This study proposes a statistical‐based detection method that is responsive to damage and not to environmental and operational conditions. The method serves as a damage recognition system for structural health monitoring using field measurements from real bridges. Vehicle‐induced bridge and ambient vibration measurements collected from the benchmark Old ADA steel truss bridge situated in Japan were utilized to validate the proposed method. The steel truss members in the bridge were subjected to five different damage scenarios to represent common potential problems in structural health monitoring of real‐life applications. The collected measurements have been completely published and made available online. A combination of principal component analysis (PCA) and linear discriminant analysis (LDA) transformation is utilized as a statistical‐based recognition technique. Vibration data representing power spectral density (PSD) functions were tested as damage‐sensitive features from identified condition sources. The proposed combination of the PCA‐LDA transformation system outperforms the popular PCA transformation as a statistical model for classification of state conditions. Although the first two principal components of PCA hold 50–85% of the variation in data, the first two components from PCA‐ LDA hold about 95% of the total variation. As a result, the three PCs, of PCA‐LDA, visualization successfully managed to classify the five structural damage scenarios into their five individual subgroups.","PeriodicalId":22049,"journal":{"name":"Structural Control and Health Monitoring","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90725831","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}
Identifying fundamental characteristics of a structure provide key information for structural health monitoring (SHM). To date, numerous researchers have reported tools and algorithms that can perform system identification, with a large portion of their application being contact‐based sensors. Although dynamic responses of structures can be directly measured from contact‐based sensors, the lifespan of those sensors being much shorter than that of the structure, requiring labor to deploy and maintain the sensors, etc., has led to the use of non‐contact‐based sensors. Among various non‐contact‐based sensors, some researchers have investigated the use of light detection and ranging (Lidar) sensors, which remotely acquire three‐dimensional ranging information, mostly for static displacement measurement during construction. Thus, this paper presents an approach for system identification of structures using dynamic displacement measured from a multi‐channel Lidar sensor. Hardware and mechanical attributes that limit the direct use of raw data from the Lidar are explored. Then, strategies to adjust the tilt axes and reduce the range uncertainties and data synchronization are proposed. Subsequently, two types of laboratory‐scale structures are prepared for validation: a flexible cantilever beam and a four‐story shear building. In both of the structures, the Lidar showed less than 5% error in estimating the first natural frequencies. Also, the mode shape has been estimated with high precision. The results demonstrate the ability of the Lidar for identifying dynamic characteristics of a structure. The potable feature of the Lidar will further allow full‐scale monitoring of a large‐scale civil infrastructure for SHM.
{"title":"Noncontact dynamic displacements measurements for structural identification using a multi‐channel Lidar","authors":"Jaehun Lee, R. Kim","doi":"10.1002/stc.3100","DOIUrl":"https://doi.org/10.1002/stc.3100","url":null,"abstract":"Identifying fundamental characteristics of a structure provide key information for structural health monitoring (SHM). To date, numerous researchers have reported tools and algorithms that can perform system identification, with a large portion of their application being contact‐based sensors. Although dynamic responses of structures can be directly measured from contact‐based sensors, the lifespan of those sensors being much shorter than that of the structure, requiring labor to deploy and maintain the sensors, etc., has led to the use of non‐contact‐based sensors. Among various non‐contact‐based sensors, some researchers have investigated the use of light detection and ranging (Lidar) sensors, which remotely acquire three‐dimensional ranging information, mostly for static displacement measurement during construction. Thus, this paper presents an approach for system identification of structures using dynamic displacement measured from a multi‐channel Lidar sensor. Hardware and mechanical attributes that limit the direct use of raw data from the Lidar are explored. Then, strategies to adjust the tilt axes and reduce the range uncertainties and data synchronization are proposed. Subsequently, two types of laboratory‐scale structures are prepared for validation: a flexible cantilever beam and a four‐story shear building. In both of the structures, the Lidar showed less than 5% error in estimating the first natural frequencies. Also, the mode shape has been estimated with high precision. The results demonstrate the ability of the Lidar for identifying dynamic characteristics of a structure. The potable feature of the Lidar will further allow full‐scale monitoring of a large‐scale civil infrastructure for SHM.","PeriodicalId":22049,"journal":{"name":"Structural Control and Health Monitoring","volume":"37 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77812030","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}
It is difficult to establish a classification and recognition model of machinery and equipment based on labeled samples in the actual industrial environment because of the imperfect fault modes and data missing. To solve this problem, a semisupervised anomaly detection method based on masked autoencoders of distribution estimation (MADE) is designed. First, the Mel‐frequency cepstrum coefficient (MFCC) is employed to extract fault features from vibration signals of rolling bearings. Then, a group of mask matrices are set on each hidden layer to overcome the perfect reconstruction problem of the autoencoders' input, and the full‐connection probability of reconstruction is used to replace the reconstruction error and adopted as the anomaly score. Finally, the diagnostic threshold is determined according to the Youden index. Experimental results show that the MADE method can extract fault‐sensitive features from a noisy industrial environment and introduce mask matrices renders to make the network autoregressive, thus solving the problem of perfect reconstruction of autoencoders. It is verified based on three rolling bearing datasets that the accuracy, precision, recall, and F1‐score of the proposed method are confirmed to be all 100%. Moreover, the accuracy of the proposed method is 17.19% higher than that of the memory‐inhibition method on the rolling bearing dataset provided by the Center for Intelligent Maintenance Systems (IMS) in University of Cincinnati (USA). The accuracy of the proposed method is also improved compared with other state‐of‐the‐art anomaly detection methods.
在实际工业环境中，由于故障模式不完善和数据缺失，难以建立基于标记样本的机械设备分类识别模型。为了解决这一问题，设计了一种基于分布估计掩码自编码器的半监督异常检测方法。首先，利用Mel - frequency倒频谱系数(MFCC)从滚动轴承振动信号中提取故障特征;然后，在每个隐藏层上设置一组掩码矩阵来克服自编码器输入的完美重构问题，并用重构的全连接概率代替重构误差作为异常评分。最后根据约登指数确定诊断阈值。实验结果表明，该方法可以从噪声工业环境中提取故障敏感特征，并引入掩模矩阵渲染使网络自回归，从而解决了自编码器的完美重构问题。基于三个滚动轴承数据集验证了该方法的准确率、精密度、召回率和F1 - score均为100%。此外，在美国辛辛那提大学(University of Cincinnati)智能维护系统中心(IMS)提供的滚动轴承数据集上，该方法的准确率比记忆抑制方法高17.19%。与其他最先进的异常检测方法相比，该方法的准确性也得到了提高。
{"title":"Anomaly detection in rolling bearings based on the Mel‐frequency cepstrum coefficient and masked autoencoder for distribution estimation","authors":"Suchao Xie, Runda Liu, Leilei Du, Hongchuang Tan","doi":"10.1002/stc.3096","DOIUrl":"https://doi.org/10.1002/stc.3096","url":null,"abstract":"It is difficult to establish a classification and recognition model of machinery and equipment based on labeled samples in the actual industrial environment because of the imperfect fault modes and data missing. To solve this problem, a semisupervised anomaly detection method based on masked autoencoders of distribution estimation (MADE) is designed. First, the Mel‐frequency cepstrum coefficient (MFCC) is employed to extract fault features from vibration signals of rolling bearings. Then, a group of mask matrices are set on each hidden layer to overcome the perfect reconstruction problem of the autoencoders' input, and the full‐connection probability of reconstruction is used to replace the reconstruction error and adopted as the anomaly score. Finally, the diagnostic threshold is determined according to the Youden index. Experimental results show that the MADE method can extract fault‐sensitive features from a noisy industrial environment and introduce mask matrices renders to make the network autoregressive, thus solving the problem of perfect reconstruction of autoencoders. It is verified based on three rolling bearing datasets that the accuracy, precision, recall, and F1‐score of the proposed method are confirmed to be all 100%. Moreover, the accuracy of the proposed method is 17.19% higher than that of the memory‐inhibition method on the rolling bearing dataset provided by the Center for Intelligent Maintenance Systems (IMS) in University of Cincinnati (USA). The accuracy of the proposed method is also improved compared with other state‐of‐the‐art anomaly detection methods.","PeriodicalId":22049,"journal":{"name":"Structural Control and Health Monitoring","volume":"108 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79515621","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}
Shibin Lin, J. Ashlock, Sadegh Shams, Fan Shi, Yujin Wang
Lamb waves have infinite number dispersion modes; however, no every mode is excitable and detectable. Traditional matrix methods can calculate the dispersion curve of each mode over the full range of possible frequencies. However, the resulting numerically calculated multimodal dispersion curves do not fully represent the dispersion curves measured in real experiments, which are most often dominated by energy from specific modes. An improved dynamic stiffness matrix method is proposed herein to overcome such challenges of the traditional matrix methods. The first step of the improved method is to calculate the displacement response of a plate‐like structure under a vertical dynamic loading using the global stiffness matrix of the structure, then the dominant dispersion trend is extracted from the displacement using the phase‐velocity scanning scheme. The improved method is verified with three case studies representing typical plate‐like structures in structural engineering. The results demonstrate that dispersion trends calculated with the improved method have good agreement with those obtained from experimental measurements.
{"title":"Analytical computation of the dominant dispersion trend of Lamb waves in plate‐like structures with an improved dynamic stiffness matrix method","authors":"Shibin Lin, J. Ashlock, Sadegh Shams, Fan Shi, Yujin Wang","doi":"10.1002/stc.3103","DOIUrl":"https://doi.org/10.1002/stc.3103","url":null,"abstract":"Lamb waves have infinite number dispersion modes; however, no every mode is excitable and detectable. Traditional matrix methods can calculate the dispersion curve of each mode over the full range of possible frequencies. However, the resulting numerically calculated multimodal dispersion curves do not fully represent the dispersion curves measured in real experiments, which are most often dominated by energy from specific modes. An improved dynamic stiffness matrix method is proposed herein to overcome such challenges of the traditional matrix methods. The first step of the improved method is to calculate the displacement response of a plate‐like structure under a vertical dynamic loading using the global stiffness matrix of the structure, then the dominant dispersion trend is extracted from the displacement using the phase‐velocity scanning scheme. The improved method is verified with three case studies representing typical plate‐like structures in structural engineering. The results demonstrate that dispersion trends calculated with the improved method have good agreement with those obtained from experimental measurements.","PeriodicalId":22049,"journal":{"name":"Structural Control and Health Monitoring","volume":"65 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78445978","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}
J. E. Ramón, J. M. Gandía-Romero, R. Bataller, J. A. López, M. Valcuende, J. Soto
The excellent high‐durability features of ultra‐high performance fibre‐reinforced concrete (UHPFRC) have been verified in laboratory studies, but its performance under service conditions are being studied. Indeed, structural health monitoring (SHM) can be considered an efficient strategy to assess built structures in which concrete matrix performance differs from that those found when assessing laboratory samples (variable actions, cracking, etc.). This work presents INESSCOM, an automated corrosion rate monitoring system, as an innovative support to SHM strategy to monitor UHPFRC structures in terms of durability. Its innovation lies in its durable and multi‐parametric sensor designed to be embedded in multiple parts of a structure. The results from previous laboratory tests and those obtained during real‐time monitoring of an offshore UHPFRC raft are presented. Acceptable deviation of 20% was obtained in corrosion rate measurements with the advantageous reference‐electrode‐free cell of the sensor with respect to the classical three‐electrode cell. Furthermore, sensor provided accurate corrosion measurements in UHPFRC despite its extremely high electrical resistivity and large amount of steel fibres. After 17‐month monitoring of the UHPFRC raft, excellent performance was evidenced under service conditions with corrosion rate values always <0.1 μA/cm2. Conversely, corrosion rate reached 0.4 μA/cm2 in a conventional concrete specimen installed for comparison. Corrosion initiation and propagation stages were clearly defined through the corrosion‐penetration‐damage (μm) diagram obtained for the specimen. Present work positions INESSCOM as an innovative support to structural health monitoring strategy in UHPFRC structures.
{"title":"Real‐time corrosion monitoring of an ultra‐high performance fibre‐reinforced concrete offshore raft by using an autonomous sensor system","authors":"J. E. Ramón, J. M. Gandía-Romero, R. Bataller, J. A. López, M. Valcuende, J. Soto","doi":"10.1002/stc.3102","DOIUrl":"https://doi.org/10.1002/stc.3102","url":null,"abstract":"The excellent high‐durability features of ultra‐high performance fibre‐reinforced concrete (UHPFRC) have been verified in laboratory studies, but its performance under service conditions are being studied. Indeed, structural health monitoring (SHM) can be considered an efficient strategy to assess built structures in which concrete matrix performance differs from that those found when assessing laboratory samples (variable actions, cracking, etc.). This work presents INESSCOM, an automated corrosion rate monitoring system, as an innovative support to SHM strategy to monitor UHPFRC structures in terms of durability. Its innovation lies in its durable and multi‐parametric sensor designed to be embedded in multiple parts of a structure. The results from previous laboratory tests and those obtained during real‐time monitoring of an offshore UHPFRC raft are presented. Acceptable deviation of 20% was obtained in corrosion rate measurements with the advantageous reference‐electrode‐free cell of the sensor with respect to the classical three‐electrode cell. Furthermore, sensor provided accurate corrosion measurements in UHPFRC despite its extremely high electrical resistivity and large amount of steel fibres. After 17‐month monitoring of the UHPFRC raft, excellent performance was evidenced under service conditions with corrosion rate values always <0.1 μA/cm2. Conversely, corrosion rate reached 0.4 μA/cm2 in a conventional concrete specimen installed for comparison. Corrosion initiation and propagation stages were clearly defined through the corrosion‐penetration‐damage (μm) diagram obtained for the specimen. Present work positions INESSCOM as an innovative support to structural health monitoring strategy in UHPFRC structures.","PeriodicalId":22049,"journal":{"name":"Structural Control and Health Monitoring","volume":"72 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81875097","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：