Pub Date : 2023-10-10DOI: 10.1177/14759217231197806
Haiyang Pan, Hong Feng, Jian Cheng, Jinde Zheng
Under the influence of strong noise, period fault features of rolling bearing are not obvious, which increases the difficulty of accurately extracting period fault features. An autonomous weak period fault extraction method under strong noise named Ramanujan-gram is proposed in this paper. The greatest advantage of Ramanujan-gram is that it uses the Ramanujan feature extraction technique to reconstruct the components in each frequency band, which can overcome the weakness of the weak noise robustness of the filter methods used by the traditional kurtogram methods and improve the accuracy of period fault feature extraction. Meanwhile, the adaptive frequency band segmentation method based on the order statistical filter is used for adaptive frequency band segmentation, which overcomes the defect that the binary tree structure of fixed frequency band segmentation may destroy the optimal demodulated frequency band. Considering that kurtosis index is difficult to accurately evaluate period fault information in components, Ramanujan-gram adopts adaptive square envelope spectrum weighted kurtosis index to improve the evaluation accuracy of period fault information. The test signals of rolling bearing verify that Ramanujan-gram has strong noise robustness and is an effective method for weak period fault extraction under strong noise.
{"title":"Ramanujan-gram: an autonomous weak period fault extraction method under strong noise","authors":"Haiyang Pan, Hong Feng, Jian Cheng, Jinde Zheng","doi":"10.1177/14759217231197806","DOIUrl":"https://doi.org/10.1177/14759217231197806","url":null,"abstract":"Under the influence of strong noise, period fault features of rolling bearing are not obvious, which increases the difficulty of accurately extracting period fault features. An autonomous weak period fault extraction method under strong noise named Ramanujan-gram is proposed in this paper. The greatest advantage of Ramanujan-gram is that it uses the Ramanujan feature extraction technique to reconstruct the components in each frequency band, which can overcome the weakness of the weak noise robustness of the filter methods used by the traditional kurtogram methods and improve the accuracy of period fault feature extraction. Meanwhile, the adaptive frequency band segmentation method based on the order statistical filter is used for adaptive frequency band segmentation, which overcomes the defect that the binary tree structure of fixed frequency band segmentation may destroy the optimal demodulated frequency band. Considering that kurtosis index is difficult to accurately evaluate period fault information in components, Ramanujan-gram adopts adaptive square envelope spectrum weighted kurtosis index to improve the evaluation accuracy of period fault information. The test signals of rolling bearing verify that Ramanujan-gram has strong noise robustness and is an effective method for weak period fault extraction under strong noise.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":"108 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136357450","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 : 2023-10-10DOI: 10.1177/14759217231186957
Davide Piciucco, Francesco Foti, Margaux Geuzaine, Vincent Denoël
The regular monitoring of cable forces is essential for ensuring the safety of cable structures both during construction and throughout their lifetime. This paper aims at developing a vibration-based identification procedure of the axial forces, bending stiffness, and, secondarily, the crossing point position of cable networks. A model constituted by two crossing stays having small bending stiffness and negligible sag effects is considered. The in-plane direct dynamic problem is solved both numerically and through a perturbation approach. The obtained results are compared to the outcomes of a finite element model for verification purposes. The theoretical studies are also supported by experimental tests performed on a real cable-stayed bridge (Haccourt bridge), which provide insights into the dynamics of the system showing that models of cables with small bending stiffness are more appropriate than taut string models. The inverse analysis based on non-linear Bayesian regression is developed and the closed-form asymptotic formulations are used to prove that the bending stiffness, the cable forces, and the crossing point position can be separately identified from a set of observed frequencies. The implemented procedure is then applied to the tested bridge as a proof of concept, showing that the proposed in-plane identification strategy provides satisfactory results.
{"title":"Bayesian forces identification in cable networks with small bending stiffness","authors":"Davide Piciucco, Francesco Foti, Margaux Geuzaine, Vincent Denoël","doi":"10.1177/14759217231186957","DOIUrl":"https://doi.org/10.1177/14759217231186957","url":null,"abstract":"The regular monitoring of cable forces is essential for ensuring the safety of cable structures both during construction and throughout their lifetime. This paper aims at developing a vibration-based identification procedure of the axial forces, bending stiffness, and, secondarily, the crossing point position of cable networks. A model constituted by two crossing stays having small bending stiffness and negligible sag effects is considered. The in-plane direct dynamic problem is solved both numerically and through a perturbation approach. The obtained results are compared to the outcomes of a finite element model for verification purposes. The theoretical studies are also supported by experimental tests performed on a real cable-stayed bridge (Haccourt bridge), which provide insights into the dynamics of the system showing that models of cables with small bending stiffness are more appropriate than taut string models. The inverse analysis based on non-linear Bayesian regression is developed and the closed-form asymptotic formulations are used to prove that the bending stiffness, the cable forces, and the crossing point position can be separately identified from a set of observed frequencies. The implemented procedure is then applied to the tested bridge as a proof of concept, showing that the proposed in-plane identification strategy provides satisfactory results.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136358080","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 : 2023-10-10DOI: 10.1177/14759217231198101
Shuo Zhang, Zhiwen Liu, Sihai He, Yunping Chen
Aiming at the problem that it is difficult to detect effective transient impact characteristics of wind turbine generator bearing fault signals due to non-stationary and strong noise, a fault diagnosis method based on adaptive redundant lifting wavelet dictionary and Bayesian biorthogonal sparse representation (SR) algorithm is proposed. First, a Bayesian model is integrated into the biorthogonal matching pursuit (MP) algorithm to improve the use of dictionary atoms in the effective support set. Then, an adaptive redundant lifting wavelet is used to construct a dictionary matching the transient characteristics of the signal. Finally, the SR algorithm is established by integrating the Bayesian biorthogonal MP model and adaptive redundant lifting wavelet dictionary. Simulation and experimental results show that the proposed method can improve the accuracy of signal reconstruction of transient components and effectively extract bearing fault features, thus verifying the effectiveness and robustness of the method.
{"title":"Transient fault extraction for wind turbine generator bearing based on Bayesian biorthogonal sparse representation using adaptive redundant lifting wavelet dictionary","authors":"Shuo Zhang, Zhiwen Liu, Sihai He, Yunping Chen","doi":"10.1177/14759217231198101","DOIUrl":"https://doi.org/10.1177/14759217231198101","url":null,"abstract":"Aiming at the problem that it is difficult to detect effective transient impact characteristics of wind turbine generator bearing fault signals due to non-stationary and strong noise, a fault diagnosis method based on adaptive redundant lifting wavelet dictionary and Bayesian biorthogonal sparse representation (SR) algorithm is proposed. First, a Bayesian model is integrated into the biorthogonal matching pursuit (MP) algorithm to improve the use of dictionary atoms in the effective support set. Then, an adaptive redundant lifting wavelet is used to construct a dictionary matching the transient characteristics of the signal. Finally, the SR algorithm is established by integrating the Bayesian biorthogonal MP model and adaptive redundant lifting wavelet dictionary. Simulation and experimental results show that the proposed method can improve the accuracy of signal reconstruction of transient components and effectively extract bearing fault features, thus verifying the effectiveness and robustness of the method.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136294313","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 : 2023-10-10DOI: 10.1177/14759217231194222
Shiji Ma, Lan Qiao, Qingwen Li
The disengagement of bridge bearings is a pervasive issue encountered in the realm of bridges, which can potentially lead to changes in operational circumstances, diminished longevity, and compromised traffic safety. The current methods employed for detecting such disconnections primarily rely on force sensors, cameras, and acceleration sensors. However, their practical implementation on-site and effectiveness in accurately identifying disengagement require enhancement. To address the challenges associated with the installation and layout of conventional contact sensors, as well as the potential introduction of additional mass, a sophisticated “bridge-bearing disconnection detection system” has been devised. This innovative system is based on laser Doppler vibrometer technology, which eliminates the need for physical contact. The feasibility of employing non-contact laser Doppler vibration measurement technology in the detection of bridge-bearing disconnection has been successfully verified within the framework of this study. Furthermore, a comprehensive analysis of the sensitivity of key dynamic parameters, specifically natural frequencies and vibration modes, to bridge-bearing disengagement has been conducted. The verification process included evaluating the identification effectiveness of regularized combined absolute changes in vibration modes and flexibility matrix diagonal matrix change rate (FDMCR) under diverse working conditions simulating complete disconnection. This assessment involved using both finite element analysis and empirical measurements. The findings unequivocally demonstrate that the disconnection of bridge bearings results in a reduction in the natural frequencies for each mode order, with an observed cumulative effect. In addition, it is noteworthy that the vibration mode indices typically exhibit greater sensitivity toward the disconnection of outer bearings. By contrast, FDMCR demonstrates commendable positioning capabilities and exceptional noise resistance in identifying bridge-bearing disengagement. The empirical insights gleaned from these research findings hold significant value in terms of on-site identification of bridge-bearing disengagement, ultimately contributing to the preservation of bridges’ long-term operational integrity.
{"title":"Bridge-bearing disengagement identification based on flexibility matrix diagonal matrix change rate: an indoor physical simulation experiment","authors":"Shiji Ma, Lan Qiao, Qingwen Li","doi":"10.1177/14759217231194222","DOIUrl":"https://doi.org/10.1177/14759217231194222","url":null,"abstract":"The disengagement of bridge bearings is a pervasive issue encountered in the realm of bridges, which can potentially lead to changes in operational circumstances, diminished longevity, and compromised traffic safety. The current methods employed for detecting such disconnections primarily rely on force sensors, cameras, and acceleration sensors. However, their practical implementation on-site and effectiveness in accurately identifying disengagement require enhancement. To address the challenges associated with the installation and layout of conventional contact sensors, as well as the potential introduction of additional mass, a sophisticated “bridge-bearing disconnection detection system” has been devised. This innovative system is based on laser Doppler vibrometer technology, which eliminates the need for physical contact. The feasibility of employing non-contact laser Doppler vibration measurement technology in the detection of bridge-bearing disconnection has been successfully verified within the framework of this study. Furthermore, a comprehensive analysis of the sensitivity of key dynamic parameters, specifically natural frequencies and vibration modes, to bridge-bearing disengagement has been conducted. The verification process included evaluating the identification effectiveness of regularized combined absolute changes in vibration modes and flexibility matrix diagonal matrix change rate (FDMCR) under diverse working conditions simulating complete disconnection. This assessment involved using both finite element analysis and empirical measurements. The findings unequivocally demonstrate that the disconnection of bridge bearings results in a reduction in the natural frequencies for each mode order, with an observed cumulative effect. In addition, it is noteworthy that the vibration mode indices typically exhibit greater sensitivity toward the disconnection of outer bearings. By contrast, FDMCR demonstrates commendable positioning capabilities and exceptional noise resistance in identifying bridge-bearing disengagement. The empirical insights gleaned from these research findings hold significant value in terms of on-site identification of bridge-bearing disengagement, ultimately contributing to the preservation of bridges’ long-term operational integrity.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136295319","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 : 2023-10-10DOI: 10.1177/14759217231202964
Mengchao Zhang, Kai Jiang, Shuai Zhao, Nini Hao, Yuan Zhang
During transportation, bulk materials are susceptible to spillage due to equipment instability and environmental factors, resulting in increased maintenance costs and environmental pollution. Thus, intelligent and efficient condition monitoring is crucial for maintaining operational efficiency of transfer equipment. It facilitates the timely identification of potential safety hazards, preventing accidents from occurring or their impact from spreading, thereby minimizing production and maintenance costs. This study presents a deep-learning-based multioperation synchronous monitoring method suitable for belt conveyors that integrate target segmentation and detection networks to simultaneously diagnose belt deviation, measure conveying load, identify idlers, and do other tasks on a self-made dataset. This method effectively reduces the complexity of multistate simultaneous monitoring and monitoring costs, thereby avoiding environmental pollution caused by transportation accidents. Experimental results show that the segmentation accuracy of the proposed method can be up to 88.72%, with a detection accuracy of 91.3% and an overall inference speed of 90.9 frames per second. Furthermore, by extending the dataset, the proposed method can incorporate additional tasks, such as belt damage, scattered material, and foreign object identifications. This study has practical significance in ensuring the normal and eco-friendly operation of bulk material transportation. Our source dataset is available at https://github.com/zhangzhangzhang1618/dataset-for-turnning-section
{"title":"Deep-learning-based multistate monitoring method of belt conveyor turning section","authors":"Mengchao Zhang, Kai Jiang, Shuai Zhao, Nini Hao, Yuan Zhang","doi":"10.1177/14759217231202964","DOIUrl":"https://doi.org/10.1177/14759217231202964","url":null,"abstract":"During transportation, bulk materials are susceptible to spillage due to equipment instability and environmental factors, resulting in increased maintenance costs and environmental pollution. Thus, intelligent and efficient condition monitoring is crucial for maintaining operational efficiency of transfer equipment. It facilitates the timely identification of potential safety hazards, preventing accidents from occurring or their impact from spreading, thereby minimizing production and maintenance costs. This study presents a deep-learning-based multioperation synchronous monitoring method suitable for belt conveyors that integrate target segmentation and detection networks to simultaneously diagnose belt deviation, measure conveying load, identify idlers, and do other tasks on a self-made dataset. This method effectively reduces the complexity of multistate simultaneous monitoring and monitoring costs, thereby avoiding environmental pollution caused by transportation accidents. Experimental results show that the segmentation accuracy of the proposed method can be up to 88.72%, with a detection accuracy of 91.3% and an overall inference speed of 90.9 frames per second. Furthermore, by extending the dataset, the proposed method can incorporate additional tasks, such as belt damage, scattered material, and foreign object identifications. This study has practical significance in ensuring the normal and eco-friendly operation of bulk material transportation. Our source dataset is available at https://github.com/zhangzhangzhang1618/dataset-for-turnning-section","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":"238 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136295723","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 : 2023-10-10DOI: 10.1177/14759217231197265
Dineo A Ramatlo, Daniel N Wilke, Philip W Loveday
Developing reliable ultrasonic-guided wave monitoring systems requires a significant amount of inspection data for each application scenario. Experimental investigations are fundamental but require a long period and are costly, especially for real-life testing. This is exacerbated by a lack of experimental data that includes damage. In some guided wave applications, such as pipelines, it is possible to introduce artificial damage and perform lab experiments on the test structure. However, in rail track applications, laboratory experiments are either not possible or meaningful. The generation of synthetic data using modelling capabilities thus becomes increasingly important. This paper presents a variational autoencoder (VAE)-based deep learning approach for generating synthetic ultrasonic inspection data for welded railway tracks. The primary aim is to use a VAE model to generate synthetic data containing damage signatures at specified positions along the length of a rail track. The VAE is trained to encode an input damage-free baseline signal and decode to reconstruct an inspection signal with damage by adding a damage signature on either side of the transducer by specifying the distance to the damage signature as an additional variable in the latent space. The training data was produced from a physics-based model that computes virtual experimental response signals using the semi-analytical finite element and the traditional finite element procedures. The VAE reconstructed response signals containing damage signatures were almost identical to the original target signals simulated using the physics-based model. The VAE was able to capture the complex features in the signals resulting from the interaction of multiple propagating modes in a multi-discontinuous waveguide. The VAE model successfully generated synthetic inspection data by fusing reflections from welds with the reflection from a crack model at specified distances from the transducer on either the right or left side. In some cases, the VAE did not exactly reconstruct the peak amplitude of the reflections. This study demonstrated the potential and highlighted the benefit of using a VAE to generate synthetic data with damage signatures as opposed to using superposition to fuse the damage-free responses containing reflections from welds with a damage signature. The results show that it is possible to generate realistic inspection data for unavailable damage scenarios.
{"title":"A data-driven hybrid approach to generate synthetic data for unavailable damage scenarios in welded rails for ultrasonic guided wave monitoring","authors":"Dineo A Ramatlo, Daniel N Wilke, Philip W Loveday","doi":"10.1177/14759217231197265","DOIUrl":"https://doi.org/10.1177/14759217231197265","url":null,"abstract":"Developing reliable ultrasonic-guided wave monitoring systems requires a significant amount of inspection data for each application scenario. Experimental investigations are fundamental but require a long period and are costly, especially for real-life testing. This is exacerbated by a lack of experimental data that includes damage. In some guided wave applications, such as pipelines, it is possible to introduce artificial damage and perform lab experiments on the test structure. However, in rail track applications, laboratory experiments are either not possible or meaningful. The generation of synthetic data using modelling capabilities thus becomes increasingly important. This paper presents a variational autoencoder (VAE)-based deep learning approach for generating synthetic ultrasonic inspection data for welded railway tracks. The primary aim is to use a VAE model to generate synthetic data containing damage signatures at specified positions along the length of a rail track. The VAE is trained to encode an input damage-free baseline signal and decode to reconstruct an inspection signal with damage by adding a damage signature on either side of the transducer by specifying the distance to the damage signature as an additional variable in the latent space. The training data was produced from a physics-based model that computes virtual experimental response signals using the semi-analytical finite element and the traditional finite element procedures. The VAE reconstructed response signals containing damage signatures were almost identical to the original target signals simulated using the physics-based model. The VAE was able to capture the complex features in the signals resulting from the interaction of multiple propagating modes in a multi-discontinuous waveguide. The VAE model successfully generated synthetic inspection data by fusing reflections from welds with the reflection from a crack model at specified distances from the transducer on either the right or left side. In some cases, the VAE did not exactly reconstruct the peak amplitude of the reflections. This study demonstrated the potential and highlighted the benefit of using a VAE to generate synthetic data with damage signatures as opposed to using superposition to fuse the damage-free responses containing reflections from welds with a damage signature. The results show that it is possible to generate realistic inspection data for unavailable damage scenarios.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136357752","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 : 2023-10-10DOI: 10.1177/14759217231199569
Changwei Liu, Jianwen Pan, Jinting Wang
Anomaly detection in deformation is important for structural health monitoring and safety evaluation of dams. In this paper, an anomaly detection model for the deformation of arch dams is presented. It combines the long short-term memory network (LSTM)-based behavior model for dam deformation prediction and the small probability method for control limits determination, and thus is called an LSTM-based anomaly detection model. To demonstrate the advantages of the LSTM-based anomaly detection model, the traditional hydrostatic-seasonal-time behavior model and the confidence interval method are considered for comparison. The 178 m-high Longyangxia Arch Dam is taken as a case study. The results show that the LSTM-based model has sufficiently high accuracy for dam deformation prediction, especially can accurately predict displacement peaks and troughs. The LSTM-based anomaly detection model can significantly avoid false warnings and missing alarms and is able to send alarms in time when the occurrence of adverse conditions causes abnormal deformation of the dam.
{"title":"An LSTM-based anomaly detection model for the deformation of concrete dams","authors":"Changwei Liu, Jianwen Pan, Jinting Wang","doi":"10.1177/14759217231199569","DOIUrl":"https://doi.org/10.1177/14759217231199569","url":null,"abstract":"Anomaly detection in deformation is important for structural health monitoring and safety evaluation of dams. In this paper, an anomaly detection model for the deformation of arch dams is presented. It combines the long short-term memory network (LSTM)-based behavior model for dam deformation prediction and the small probability method for control limits determination, and thus is called an LSTM-based anomaly detection model. To demonstrate the advantages of the LSTM-based anomaly detection model, the traditional hydrostatic-seasonal-time behavior model and the confidence interval method are considered for comparison. The 178 m-high Longyangxia Arch Dam is taken as a case study. The results show that the LSTM-based model has sufficiently high accuracy for dam deformation prediction, especially can accurately predict displacement peaks and troughs. The LSTM-based anomaly detection model can significantly avoid false warnings and missing alarms and is able to send alarms in time when the occurrence of adverse conditions causes abnormal deformation of the dam.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136295332","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 : 2023-10-10DOI: 10.1177/14759217231193702
Alvin Chen, Fotis Kopsaftopoulos, Sandipan Mishra
Anomalies often occur in metal additive manufacturing from processing inconsistencies and uncertainty. A robust fault detection system that uses sensor measurements such as melt pool imaging has the potential to improve part quality and save production time by anticipating print failure. Toward this goal, we develop and validate a fault detection technique using melt pool geometry-related measurements from an in situ near-infrared optical camera. This method is unsupervised and is trained on a small dataset, mitigating human error in classifying fault types, and reducing lead times for preparing training datasets. Furthermore, this method uses learned geometry-informed nominal behavior of the melt pool signal to make informed decisions on the process health. There are spatial-temporal characteristics embedded in the melt pool images, caused by the periodicity in the geometry-dependent raster pattern. These characteristics can be captured in the frequency domain using the signal spectrogram, a representation of the frequency content over time. Defects will appear in the spectrogram, disrupting the healthy spectral response. To quantify healthy spectrograms, we use principal component (PC) decomposition to extract the features of these spectrograms as a set of nominal basis vectors. Anomaly detection is then performed by calculating the error between the original and reconstructed spectrogram vector by projection of the spectrogram PCs onto the nominal basis. The reconstruction error for anomalous signals is larger than that from healthy signals, which is then used for fault detection. A one-tailed statistical test is used to determine the fault detection threshold for the reconstruction error signal. This method is tested on three raster patterns and performs better than a comparative time-series thresholding method. We demonstrate that this time-frequency algorithm can detect both temporal faults (which occur at a single time instant) and spatial faults (such as those introduced by an improper sintering), differentiating them from nominal operation.
{"title":"An unsupervised online anomaly detection method for metal additive manufacturing processes via a statistical time-frequency domain algorithm","authors":"Alvin Chen, Fotis Kopsaftopoulos, Sandipan Mishra","doi":"10.1177/14759217231193702","DOIUrl":"https://doi.org/10.1177/14759217231193702","url":null,"abstract":"Anomalies often occur in metal additive manufacturing from processing inconsistencies and uncertainty. A robust fault detection system that uses sensor measurements such as melt pool imaging has the potential to improve part quality and save production time by anticipating print failure. Toward this goal, we develop and validate a fault detection technique using melt pool geometry-related measurements from an in situ near-infrared optical camera. This method is unsupervised and is trained on a small dataset, mitigating human error in classifying fault types, and reducing lead times for preparing training datasets. Furthermore, this method uses learned geometry-informed nominal behavior of the melt pool signal to make informed decisions on the process health. There are spatial-temporal characteristics embedded in the melt pool images, caused by the periodicity in the geometry-dependent raster pattern. These characteristics can be captured in the frequency domain using the signal spectrogram, a representation of the frequency content over time. Defects will appear in the spectrogram, disrupting the healthy spectral response. To quantify healthy spectrograms, we use principal component (PC) decomposition to extract the features of these spectrograms as a set of nominal basis vectors. Anomaly detection is then performed by calculating the error between the original and reconstructed spectrogram vector by projection of the spectrogram PCs onto the nominal basis. The reconstruction error for anomalous signals is larger than that from healthy signals, which is then used for fault detection. A one-tailed statistical test is used to determine the fault detection threshold for the reconstruction error signal. This method is tested on three raster patterns and performs better than a comparative time-series thresholding method. We demonstrate that this time-frequency algorithm can detect both temporal faults (which occur at a single time instant) and spatial faults (such as those introduced by an improper sintering), differentiating them from nominal operation.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136295710","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 : 2023-10-10DOI: 10.1177/14759217231182305
Chenfei Du, Jianhua Liu, Hao Gong, Jiayu Huang, Wentao Zhang
Threaded fasteners are widely applied in mechanical systems, providing the functions of connection, fastening, and sealing. However, loosening is vulnerable to occurring in harsh environment. The importance of loosening detection cannot be emphasized. Percussion-based loosening detection method has attracted much attention due to the convenience and low cost. However, the simultaneous loosening detection of multiple-threaded fasteners based on percussion method is still a challenging issue that needs to be addressed. This study proposes a novel multi-bolt loosening detection method combining percussion method, and deep learning. The method consists of three integrated modules, that is, signal preprocessing, loosening information enhancement, and loosening detection modules. In the first module, variational mode decomposition is used to decompose the original signal into a series of intrinsic mode function to eliminate the interference of noise. In the second module, compressive sampling matching pursuit is applied to represent the denoised signal sparsely, and the sparse signal is fused with the denoised signal to enhance loosening information in the signal. Last, DenseNet-CBAM network structure combining attention mechanism is proposed for multiple classification task. Experimental results showed that the proposed method achieved the detection accuracy of more than 97% in three different types of mechanical structures with multiple-threaded fasteners, indicating its great potentials in engineering applications.
{"title":"Percussion-based loosening detection method for multi-bolt structure using convolutional neural network DenseNet-CBAM","authors":"Chenfei Du, Jianhua Liu, Hao Gong, Jiayu Huang, Wentao Zhang","doi":"10.1177/14759217231182305","DOIUrl":"https://doi.org/10.1177/14759217231182305","url":null,"abstract":"Threaded fasteners are widely applied in mechanical systems, providing the functions of connection, fastening, and sealing. However, loosening is vulnerable to occurring in harsh environment. The importance of loosening detection cannot be emphasized. Percussion-based loosening detection method has attracted much attention due to the convenience and low cost. However, the simultaneous loosening detection of multiple-threaded fasteners based on percussion method is still a challenging issue that needs to be addressed. This study proposes a novel multi-bolt loosening detection method combining percussion method, and deep learning. The method consists of three integrated modules, that is, signal preprocessing, loosening information enhancement, and loosening detection modules. In the first module, variational mode decomposition is used to decompose the original signal into a series of intrinsic mode function to eliminate the interference of noise. In the second module, compressive sampling matching pursuit is applied to represent the denoised signal sparsely, and the sparse signal is fused with the denoised signal to enhance loosening information in the signal. Last, DenseNet-CBAM network structure combining attention mechanism is proposed for multiple classification task. Experimental results showed that the proposed method achieved the detection accuracy of more than 97% in three different types of mechanical structures with multiple-threaded fasteners, indicating its great potentials in engineering applications.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":"123 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136357953","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 : 2023-10-10DOI: 10.1177/14759217231203241
Zhongjie Zhang, Liang Zeng, Nan Zhang
This paper presents a damage sparse imaging method using multipath-scattered Lamb waves. It leverages a large number of echoes and reverberations in the recorded signal that may be usually ignored in conventional methods. First, reflections of Lamb waves at free edges are viewed as waves transmitted from a virtual transducer which is located at the mirror point of the actual one. On this basis, an optimized transducers-layout strategy is proposed based on the multipath propagation model of the Lamb wave. Benefiting from that, the direct damage-scattered wave and several waves scattered by both the damage and edges could be separately identified in the time domain, and further, each wave could be matched with a sensing path (either actual or virtual) in the expanded sensor network. Subsequently, a dictionary is constructed from the Lamb wave propagation and scattering model. By solving the sparse reconstruction problem, the pixel value of each point in the region of interest is obtained, and the whole area can be finally visualized. The proposed method is validated using experiments conducted on an aluminum plate with simulated damages. Results show that the damages can be correctly detected and accurately localized with only a single transmitter–receiver pair.
{"title":"Damage imaging using multipath-scattered Lamb waves under a sparse reconstruction framework","authors":"Zhongjie Zhang, Liang Zeng, Nan Zhang","doi":"10.1177/14759217231203241","DOIUrl":"https://doi.org/10.1177/14759217231203241","url":null,"abstract":"This paper presents a damage sparse imaging method using multipath-scattered Lamb waves. It leverages a large number of echoes and reverberations in the recorded signal that may be usually ignored in conventional methods. First, reflections of Lamb waves at free edges are viewed as waves transmitted from a virtual transducer which is located at the mirror point of the actual one. On this basis, an optimized transducers-layout strategy is proposed based on the multipath propagation model of the Lamb wave. Benefiting from that, the direct damage-scattered wave and several waves scattered by both the damage and edges could be separately identified in the time domain, and further, each wave could be matched with a sensing path (either actual or virtual) in the expanded sensor network. Subsequently, a dictionary is constructed from the Lamb wave propagation and scattering model. By solving the sparse reconstruction problem, the pixel value of each point in the region of interest is obtained, and the whole area can be finally visualized. The proposed method is validated using experiments conducted on an aluminum plate with simulated damages. Results show that the damages can be correctly detected and accurately localized with only a single transmitter–receiver pair.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136295725","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：