Pub Date : 2023-12-01DOI: 10.1784/insi.2023.65.12.689
Polyenergetic forward projection has great significance in inspecting hazardous materials, establishing optimal radiographic variables and investigating beam hardening effects. However, it is computationally intensive to perform polyenergetic forward-projection calculations for high-resolution phantoms. To address this issue, a rapid polyenergetic forward-projection algorithm is proposed for a 9 MeV industrial computed tomography (CT) system. The FLUktuierende KAskade (FLUKA) software package is used to generate the 9 MeV X-ray spectrum data. Two voxelised phantoms are used to model scanned objects, one being a multi-material cylinder and the other a single-material turbine blade. An incremental version of Siddon's algorithm is adopted to calculate the intersection lengths between the X-rays and the auxiliary phantoms. Three strategies are utilised to accelerate the calculation, in which: the intersection lengths do not vary with the energy bins and can be used repeatedly until all the energy bins are counted; a graphics processing unit (GPU) is used to accelerate the ray tracing algorithm by utilising a parallel computing technique; and faster memory access is achieved by binding the auxiliary phantoms to texture objects. The simulation results in this paper show that the GPU-based approach not only maintains the image precision but also gains significant speed-ups over the conventional central processing unit (CPU)-based Siddon method. Furthermore, beam hardening artefacts can clearly be seen from the profile curves of the reconstructed slices, indicating that this method is effective.
多能前向投影在检测危险材料、确定最佳射线变量和研究光束硬化效应方面具有重要意义。然而，对高分辨率模型进行多能正向投影计算需要大量计算。为了解决这个问题，我们为 9 MeV 工业计算机断层扫描（CT）系统提出了一种快速多能正向投影算法。FLUktuierende KAskade (FLUKA) 软件包用于生成 9 MeV X 射线光谱数据。两个体素化模型用于模拟扫描对象，一个是多材料圆柱体，另一个是单材料涡轮叶片。采用增量版 Siddon 算法计算 X 射线与辅助模型之间的交点长度。本文采用了三种策略来加快计算速度，其中：交点长度不会随能量箱的变化而变化，可以重复使用，直到计算完所有能量箱为止；利用并行计算技术，使用图形处理器（GPU）来加快光线跟踪算法；通过将辅助模型与纹理对象绑定，实现更快的内存访问速度。本文的仿真结果表明，与传统的基于中央处理器（CPU）的 Siddon 方法相比，基于 GPU 的方法不仅保持了图像精度，而且还显著提高了速度。此外，从重建切片的轮廓曲线上可以清楚地看到光束硬化伪影，这表明该方法是有效的。
{"title":"GPU-accelerated polyenergetic forward projection for 9 MeV industrial CT system","authors":"","doi":"10.1784/insi.2023.65.12.689","DOIUrl":"https://doi.org/10.1784/insi.2023.65.12.689","url":null,"abstract":"Polyenergetic forward projection has great significance in inspecting hazardous materials, establishing optimal radiographic variables and investigating beam hardening effects. However, it is computationally intensive to perform polyenergetic forward-projection calculations for high-resolution\u0000 phantoms. To address this issue, a rapid polyenergetic forward-projection algorithm is proposed for a 9 MeV industrial computed tomography (CT) system. The FLUktuierende KAskade (FLUKA) software package is used to generate the 9 MeV X-ray spectrum data. Two voxelised phantoms are used to model\u0000 scanned objects, one being a multi-material cylinder and the other a single-material turbine blade. An incremental version of Siddon's algorithm is adopted to calculate the intersection lengths between the X-rays and the auxiliary phantoms. Three strategies are utilised to accelerate the calculation,\u0000 in which: the intersection lengths do not vary with the energy bins and can be used repeatedly until all the energy bins are counted; a graphics processing unit (GPU) is used to accelerate the ray tracing algorithm by utilising a parallel computing technique; and faster memory access is achieved\u0000 by binding the auxiliary phantoms to texture objects. The simulation results in this paper show that the GPU-based approach not only maintains the image precision but also gains significant speed-ups over the conventional central processing unit (CPU)-based Siddon method. Furthermore, beam\u0000 hardening artefacts can clearly be seen from the profile curves of the reconstructed slices, indicating that this method is effective.","PeriodicalId":344397,"journal":{"name":"Insight - Non-Destructive Testing and Condition Monitoring","volume":"18 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139015191","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-12-01DOI: 10.1784/insi.2023.65.12.695
Chaozhi Cai, Renlong Li, Qiang Ma, Hongfeng Gao
Fault diagnosis methods for rolling bearings based on deep learning have become a research hotspot. However, these methods mostly use convolutional neural networks (CNNs), which have the problem of gradient dispersion or disappearance as the network deepens. Moreover, directly converting vibration signals into images as network input cannot preserve the temporal correlation between signals. In the case of small datasets and complex and variable working conditions, the accuracy of fault diagnosis is low and the generalisation ability is poor. To solve the above problems, a rolling bearing fault diagnosis method based on the Gramian angular field (GAF) and an SE-ResNeXt50 transfer learning model is proposed. Firstly, the parameters of the GAF obtained from multiple experiments are selected and the one-dimensional time-series vibration signal is encoded by combining the data enhancement method, and converted into a Gramian angular difference field (GADF) diagram and a Gramian angular sum field (GASF) diagram with local time information and uniqueness. Then, a fine-tuning transfer learning strategy is used to transfer the pre-trained model parameters to an SE-ResNeXt50 model, which improves the training speed of the model and improves the overfitting problem of the model on small target datasets. Finally, the GAF diagram is used as the input to the model and a feature recalibration strategy is used to adaptively obtain the importance of each feature channel, which further improves the feature utilisation. To verify the effectiveness and superiority of the proposed method, the rolling bearing data from Case Western Reserve University are selected for experimental verification and the generalisation performance of the proposed method is tested under varying loads and different dataset scales. The results show that when there is only a small amount of data, the proposed method can still achieve high diagnosis accuracy for different loads and has better recognition accuracy and generalisation compared to other fault diagnosis methods.
基于深度学习的滚动轴承故障诊断方法已成为研究热点。然而，这些方法大多使用卷积神经网络（CNN），随着网络的深入，存在梯度分散或消失的问题。此外，直接将振动信号转换成图像作为网络输入无法保留信号之间的时间相关性。在数据集较小、工况复杂多变的情况下，故障诊断的准确性较低，泛化能力较差。为解决上述问题，本文提出了一种基于格拉米安角场（GAF）和 SE-ResNeXt50 转移学习模型的滚动轴承故障诊断方法。首先，选取多次实验得到的格拉西亚角场（GAF）参数，结合数据增强方法对一维时间序列振动信号进行编码，转换成具有局部时间信息和唯一性的格拉西亚角差场（GADF）图和格拉西亚角和场（GASF）图。然后，使用微调转移学习策略将预训练模型参数转移到 SE-ResNeXt50 模型中，从而提高了模型的训练速度，并改善了模型在小目标数据集上的过拟合问题。最后，将 GAF 图作为模型的输入，并使用特征重新校准策略自适应地获取每个特征通道的重要性，从而进一步提高了特征利用率。为了验证所提方法的有效性和优越性，我们选取了凯斯西储大学的滚动轴承数据进行实验验证，并测试了所提方法在不同载荷和不同数据集规模下的泛化性能。结果表明，在只有少量数据的情况下，与其他故障诊断方法相比，所提出的方法在不同载荷下仍能达到较高的诊断精度，并且具有更好的识别精度和泛化性能。
{"title":"Bearing fault diagnosis method based on the Gramian angular field and an SE-ResNeXt50 transfer learning model","authors":"Chaozhi Cai, Renlong Li, Qiang Ma, Hongfeng Gao","doi":"10.1784/insi.2023.65.12.695","DOIUrl":"https://doi.org/10.1784/insi.2023.65.12.695","url":null,"abstract":"Fault diagnosis methods for rolling bearings based on deep learning have become a research hotspot. However, these methods mostly use convolutional neural networks (CNNs), which have the problem of gradient dispersion or disappearance as the network deepens. Moreover, directly converting\u0000 vibration signals into images as network input cannot preserve the temporal correlation between signals. In the case of small datasets and complex and variable working conditions, the accuracy of fault diagnosis is low and the generalisation ability is poor. To solve the above problems, a\u0000 rolling bearing fault diagnosis method based on the Gramian angular field (GAF) and an SE-ResNeXt50 transfer learning model is proposed. Firstly, the parameters of the GAF obtained from multiple experiments are selected and the one-dimensional time-series vibration signal is encoded by combining\u0000 the data enhancement method, and converted into a Gramian angular difference field (GADF) diagram and a Gramian angular sum field (GASF) diagram with local time information and uniqueness. Then, a fine-tuning transfer learning strategy is used to transfer the pre-trained model parameters to\u0000 an SE-ResNeXt50 model, which improves the training speed of the model and improves the overfitting problem of the model on small target datasets. Finally, the GAF diagram is used as the input to the model and a feature recalibration strategy is used to adaptively obtain the importance of each\u0000 feature channel, which further improves the feature utilisation. To verify the effectiveness and superiority of the proposed method, the rolling bearing data from Case Western Reserve University are selected for experimental verification and the generalisation performance of the proposed method\u0000 is tested under varying loads and different dataset scales. The results show that when there is only a small amount of data, the proposed method can still achieve high diagnosis accuracy for different loads and has better recognition accuracy and generalisation compared to other fault diagnosis\u0000 methods.","PeriodicalId":344397,"journal":{"name":"Insight - Non-Destructive Testing and Condition Monitoring","volume":"85 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139017408","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-12-01DOI: 10.1784/insi.2023.65.12.682
O. Kara, H. H. Çelik
Magnetic flux leakage (MFL) is a non-destructive testing method used to detect railhead surface defects. For effective MFL testing, the homogeneity of the magnetic flux density distribution (MFDD) within the railhead is crucial. Inhomogeneous formation of the MFDD within the railhead reduces the efficiency of the MFL testing. The homogeneity of the MFDD depends on the distance between poles (DBP) of the MFL testing system. According to the literature, as the DBP parameter increases, the MFDD becomes more homogeneous. In this study, four different homogeneity levels of the MFDD are introduced based on the DBP parameter. 3D finite element method (FEM) modelling simulation is conducted to obtain MFL testing analyses. The analyses are performed on a rail that contains rectangular surface defects of varying depth and length. The results of this study are evaluated using characteristic features of the MFL signal BX component, namely the slope of the baseline, the bottom value and the peak-to-peak value. The results show that if the homogeneity level of the MFDD within the railhead is higher, the bottom value and slope of the baseline decrease and the peak-to-peak value increases. This indicates that higher homogeneity of the MFDD enhances the detection efficiency of the MFL testing. Eventually, it is found that with the formation of nearly 100% homogeneous MFDD in the railhead, the slope of the baseline, the bottom value and the peak-to-peak value are enhanced by up to 83%, 77% and 12%, respectively.
{"title":"The homogeneity effects of magnetic flux density distribution on the detection of railhead surface defects via the magnetic flux leakage method","authors":"O. Kara, H. H. Çelik","doi":"10.1784/insi.2023.65.12.682","DOIUrl":"https://doi.org/10.1784/insi.2023.65.12.682","url":null,"abstract":"Magnetic flux leakage (MFL) is a non-destructive testing method used to detect railhead surface defects. For effective MFL testing, the homogeneity of the magnetic flux density distribution (MFDD) within the railhead is crucial. Inhomogeneous formation of the MFDD within the railhead\u0000 reduces the efficiency of the MFL testing. The homogeneity of the MFDD depends on the distance between poles (DBP) of the MFL testing system. According to the literature, as the DBP parameter increases, the MFDD becomes more homogeneous. In this study, four different homogeneity levels of\u0000 the MFDD are introduced based on the DBP parameter. 3D finite element method (FEM) modelling simulation is conducted to obtain MFL testing analyses. The analyses are performed on a rail that contains rectangular surface defects of varying depth and length. The results of this study are evaluated\u0000 using characteristic features of the MFL signal BX component, namely the slope of the baseline, the bottom value and the peak-to-peak value. The results show that if the homogeneity level of the MFDD within the railhead is higher, the bottom value and slope of the baseline decrease\u0000 and the peak-to-peak value increases. This indicates that higher homogeneity of the MFDD enhances the detection efficiency of the MFL testing. Eventually, it is found that with the formation of nearly 100% homogeneous MFDD in the railhead, the slope of the baseline, the bottom value and the\u0000 peak-to-peak value are enhanced by up to 83%, 77% and 12%, respectively.","PeriodicalId":344397,"journal":{"name":"Insight - Non-Destructive Testing and Condition Monitoring","volume":"149 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139016991","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-12-01DOI: 10.1784/insi.2023.65.12.675
Feng Jiang, Rongxi Hou, Li Tao
In order to obtain defect information quickly and effectively and improve the accuracy and evaluation ability of traditional electromagnetic non-destructive testing (NDT), an electromagnetic image recognition method for the defect profile based on magnetic field disturbance is proposed in this paper. The excitation coil structure is designed, the excitation mode of the signal source is optimised and a three-dimensional electromagnetic transient analysis model is established for defect profile identification of a metal surface with an anti-corrosion protective layer. The research shows that the disturbed magnetic field Bz has the characteristics of high-resolution imaging and symmetry. The orientation of the defect on the surface has different effects on the clarity of image recognition. The larger the angle between the defect boundary and the induced current, the more complete and clear the image formed by the disturbed magnetic field Bz . A rectangular square wave is the best excitation signal for defect recognition. Its Bz image at t = 0 can present complete shape and position information about the defect. In addition, the excitation coil structure based on the principle of the disturbed magnetic field must provide a uniform induced current to produce a pronounced disturbed magnetic field. It is concluded that electromagnetic imaging technology based on the disturbed magnetic field Bz can better detect and characterise the shape of metal surface defects without damaging the metal protective layer and has good application potential for NDT and safety evaluation of in-service equipment.
{"title":"Electromagnetic image recognition of a defect profile on a metal surface with a protective layer based on magnetic disturbance","authors":"Feng Jiang, Rongxi Hou, Li Tao","doi":"10.1784/insi.2023.65.12.675","DOIUrl":"https://doi.org/10.1784/insi.2023.65.12.675","url":null,"abstract":"In order to obtain defect information quickly and effectively and improve the accuracy and evaluation ability of traditional electromagnetic non-destructive testing (NDT), an electromagnetic image recognition method for the defect profile based on magnetic field disturbance is proposed\u0000 in this paper. The excitation coil structure is designed, the excitation mode of the signal source is optimised and a three-dimensional electromagnetic transient analysis model is established for defect profile identification of a metal surface with an anti-corrosion protective layer. The\u0000 research shows that the disturbed magnetic field Bz has the characteristics of high-resolution imaging and symmetry. The orientation of the defect on the surface has different effects on the clarity of image recognition. The larger the angle between the defect boundary and the induced\u0000 current, the more complete and clear the image formed by the disturbed magnetic field Bz . A rectangular square wave is the best excitation signal for defect recognition. Its Bz image at t = 0 can present complete shape and position information about the defect. In addition,\u0000 the excitation coil structure based on the principle of the disturbed magnetic field must provide a uniform induced current to produce a pronounced disturbed magnetic field. It is concluded that electromagnetic imaging technology based on the disturbed magnetic field Bz can better\u0000 detect and characterise the shape of metal surface defects without damaging the metal protective layer and has good application potential for NDT and safety evaluation of in-service equipment.","PeriodicalId":344397,"journal":{"name":"Insight - Non-Destructive Testing and Condition Monitoring","volume":"51 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138992656","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-12-01DOI: 10.1784/insi.2023.65.12.666
Yumei Ye, Jingang Zhang, Qiang Yang, Songhe Meng, Jun Wang
The dynamic responses of key regions are critical inputs for the structural life estimation of spacecraft. Response reconstruction methods are needed for structural locations where sensors are not placed due to resource limitations. In this paper, a reconstruction method based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is proposed. CEEMDAN can eliminate the mode-mixing phenomenon of traditional empirical mode decomposition (EMD) during signal decompositions to improve the reconstruction accuracy. The proposed method is applied to the reconstruction of acceleration and strain responses at critical locations of a load-bearing structure under sinusoidal and random vibration loads. Numerical and experimental validation are carried out. The numerical results show that the reconstructions are almost unaffected by the selected white noise levels of CEEMDAN and the locations of measured and targeted points. The experimental results show that compared with traditional EMD, the reconstruction accuracy of CEEMDAN is improved by a maximum of 79.94% with almost no additional computational cost. The proposed reconstruction method shows efficiency and accuracy for a wide range of applications.
{"title":"Complete ensemble empirical mode decomposition with adaptive noise for dynamic response reconstruction of spacecraft structures under random vibration","authors":"Yumei Ye, Jingang Zhang, Qiang Yang, Songhe Meng, Jun Wang","doi":"10.1784/insi.2023.65.12.666","DOIUrl":"https://doi.org/10.1784/insi.2023.65.12.666","url":null,"abstract":"The dynamic responses of key regions are critical inputs for the structural life estimation of spacecraft. Response reconstruction methods are needed for structural locations where sensors are not placed due to resource limitations. In this paper, a reconstruction method based on complete\u0000 ensemble empirical mode decomposition with adaptive noise (CEEMDAN) is proposed. CEEMDAN can eliminate the mode-mixing phenomenon of traditional empirical mode decomposition (EMD) during signal decompositions to improve the reconstruction accuracy. The proposed method is applied to the reconstruction\u0000 of acceleration and strain responses at critical locations of a load-bearing structure under sinusoidal and random vibration loads. Numerical and experimental validation are carried out. The numerical results show that the reconstructions are almost unaffected by the selected white noise levels\u0000 of CEEMDAN and the locations of measured and targeted points. The experimental results show that compared with traditional EMD, the reconstruction accuracy of CEEMDAN is improved by a maximum of 79.94% with almost no additional computational cost. The proposed reconstruction method shows efficiency\u0000 and accuracy for a wide range of applications.","PeriodicalId":344397,"journal":{"name":"Insight - Non-Destructive Testing and Condition Monitoring","volume":"353 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139022968","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-01DOI: 10.1784/insi.2023.65.11.600
Xiangqian Fan, Jueding Liu
In order to identify the dynamic fracture characteristics of fibre-reinforced polymer (FRP)-reinforced concrete, three-point bending fracture tests of FRP-reinforced concrete with five different strain loading rates (10−1/s, 10−2/s, 10−3/s, 10−4/s and 10−5/s) are carried out. The fracture behaviour of FRP-reinforced concrete is analysed using digital image correlation (DIC) and acoustic emission (AE) techniques. Crack propagation in FRP-reinforced concrete beams during the fracture process is observed and a quantitative relationship is established between the AE parameters and the load level before the ultimate load is reached. Test results on the variation of the AE parameters can characterise the internal damage behaviour of the FRP-reinforced concrete beams and monitor the distribution characteristics of any cracks. The crack propagation trajectory of the FRP-reinforced concrete beams can be observed using the DIC technique and quantitative analysis of the crack opening displacement (COD) can be achieved.
{"title":"Fracture behaviour characterisation of FRP-reinforced concrete based on digital image correlation and acoustic emission technology","authors":"Xiangqian Fan, Jueding Liu","doi":"10.1784/insi.2023.65.11.600","DOIUrl":"https://doi.org/10.1784/insi.2023.65.11.600","url":null,"abstract":"In order to identify the dynamic fracture characteristics of fibre-reinforced polymer (FRP)-reinforced concrete, three-point bending fracture tests of FRP-reinforced concrete with five different strain loading rates (10−1/s, 10−2/s, 10−3/s, 10−4/s and 10−5/s) are carried out. The fracture behaviour of FRP-reinforced concrete is analysed using digital image correlation (DIC) and acoustic emission (AE) techniques. Crack propagation in FRP-reinforced concrete beams during the fracture process is observed and a quantitative relationship is established between the AE parameters and the load level before the ultimate load is reached. Test results on the variation of the AE parameters can characterise the internal damage behaviour of the FRP-reinforced concrete beams and monitor the distribution characteristics of any cracks. The crack propagation trajectory of the FRP-reinforced concrete beams can be observed using the DIC technique and quantitative analysis of the crack opening displacement (COD) can be achieved.","PeriodicalId":344397,"journal":{"name":"Insight - Non-Destructive Testing and Condition Monitoring","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139299575","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-01DOI: 10.1784/insi.2023.65.11.631
Tianyi Yu, Shunming Li, Jiantao Lu
The use of neural network models to monitor bearing vibration signals can easily be affected by noise, which leads to a decrease in the model test accuracy. Therefore, the existence of noise problems increases the requirements for non-linear mapping capability and robustness of deep neural network (DNN) models. In order to deal with the noise problem, the concept of qubit neurons is introduced into a deep learning stacked autoencoder (SAE) model, and a quantum stacked autoencoder (QSAE) model based on qubits and quantum gates is proposed. The properties of SAE layer-by-layer coding and the arithmetic of qubit neurons are combined in the QSAE. The quantum state signal is taken as the input signal to the encoder and the coding activation function and coding weight matrix are redefined by quantum-controlled non-gates and quantum revolving gates, so that the quantum state signal can be coded layer by layer. Experimental results show that the QSAE can train and diagnose noisy experimental data and maintain high test accuracy in an anti-attack test. This shows that the QSAE has non-linear mapping capability and robustness.
{"title":"Quantum stacked autoencoder fault diagnosis model for bearing faults","authors":"Tianyi Yu, Shunming Li, Jiantao Lu","doi":"10.1784/insi.2023.65.11.631","DOIUrl":"https://doi.org/10.1784/insi.2023.65.11.631","url":null,"abstract":"The use of neural network models to monitor bearing vibration signals can easily be affected by noise, which leads to a decrease in the model test accuracy. Therefore, the existence of noise problems increases the requirements for non-linear mapping capability and robustness of deep neural network (DNN) models. In order to deal with the noise problem, the concept of qubit neurons is introduced into a deep learning stacked autoencoder (SAE) model, and a quantum stacked autoencoder (QSAE) model based on qubits and quantum gates is proposed. The properties of SAE layer-by-layer coding and the arithmetic of qubit neurons are combined in the QSAE. The quantum state signal is taken as the input signal to the encoder and the coding activation function and coding weight matrix are redefined by quantum-controlled non-gates and quantum revolving gates, so that the quantum state signal can be coded layer by layer. Experimental results show that the QSAE can train and diagnose noisy experimental data and maintain high test accuracy in an anti-attack test. This shows that the QSAE has non-linear mapping capability and robustness.","PeriodicalId":344397,"journal":{"name":"Insight - Non-Destructive Testing and Condition Monitoring","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139301846","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-01DOI: 10.1784/insi.2023.65.11.618
Penghao Liu, Juwei Zhang, Bo Liu
Based on the metal magnetic memory effect, this paper improves the Jiles-Atheron (J-A) force-magnetic coupling model under the condition of applying a weak magnetic excitation, introduces the relative magnetic induction parameter into the theoretical derivation process and explores the stress magnetisation law for the damaged part of a broken wire rope. The Ansys simulation platform is used to construct a force-magnetic coupling simulation model of the steel wire rope and finite element analysis is carried out. Combined with a static tensile test, the validity of the resulting theoretical model is verified. The results show that there is a certain relationship between the relative magnetic induction of the steel wire rope sample and the tensile stress. Its distribution can be used to evaluate the stress development trend and provide early warning of the final failure of the specimen, which provides a basis for future quantitative evaluation of damage status.
{"title":"Study of the law of stress magnetisation based on the magnetic memory effect under weak magnetic excitation","authors":"Penghao Liu, Juwei Zhang, Bo Liu","doi":"10.1784/insi.2023.65.11.618","DOIUrl":"https://doi.org/10.1784/insi.2023.65.11.618","url":null,"abstract":"Based on the metal magnetic memory effect, this paper improves the Jiles-Atheron (J-A) force-magnetic coupling model under the condition of applying a weak magnetic excitation, introduces the relative magnetic induction parameter into the theoretical derivation process and explores the stress magnetisation law for the damaged part of a broken wire rope. The Ansys simulation platform is used to construct a force-magnetic coupling simulation model of the steel wire rope and finite element analysis is carried out. Combined with a static tensile test, the validity of the resulting theoretical model is verified. The results show that there is a certain relationship between the relative magnetic induction of the steel wire rope sample and the tensile stress. Its distribution can be used to evaluate the stress development trend and provide early warning of the final failure of the specimen, which provides a basis for future quantitative evaluation of damage status.","PeriodicalId":344397,"journal":{"name":"Insight - Non-Destructive Testing and Condition Monitoring","volume":"18 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139304008","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In response to the frequent occurrence of leakage accidents in heating pipelines, timely detection of leakage points in such pipelines is of great significance to ensure the safe operation of heating systems. This article proposes a method for detecting leakage points in heating pipelines using drones equipped with infrared thermal imagers, employing a combination of the improved RR3DETDet algorithm and the adaptive threshold method. Firstly, the algorithm identifies the area of the heating pipeline and then employs the adaptive threshold method to detect the presence of leakage points in the identified pipeline area. Additionally, taking into account the morphological characteristics of heating pipelines, the RR3DETDet network is enhanced by introducing variable convolution, enabling more precise extraction of pipeline features. To reduce model overfitting and enhance network expression capabilities, the H-swish activation function is employed to replace the original activation function. Furthermore, candidate anchor boxes are clustered using the K-means++ clustering algorithm to obtain better position regression results and improve training efficiency. The improved algorithm demonstrates significantly better positioning precision compared to the original network. Moreover, an adaptive threshold algorithm is proposed for leak detection and labelling, utilising the original temperature information contained in infrared images. The experimental results demonstrate that this method achieves higher accuracy in detecting leaks in heating pipelines.
{"title":"Heating pipeline identification and leakage detection method based on improved R3 Det","authors":"Jiayan Chen, Zhiqian Li, Ping Tang, Shuai Kong, Jiansheng Hu, Qiang Wang","doi":"10.1784/insi.2023.65.11.609","DOIUrl":"https://doi.org/10.1784/insi.2023.65.11.609","url":null,"abstract":"In response to the frequent occurrence of leakage accidents in heating pipelines, timely detection of leakage points in such pipelines is of great significance to ensure the safe operation of heating systems. This article proposes a method for detecting leakage points in heating pipelines using drones equipped with infrared thermal imagers, employing a combination of the improved RR3DETDet algorithm and the adaptive threshold method. Firstly, the algorithm identifies the area of the heating pipeline and then employs the adaptive threshold method to detect the presence of leakage points in the identified pipeline area. Additionally, taking into account the morphological characteristics of heating pipelines, the RR3DETDet network is enhanced by introducing variable convolution, enabling more precise extraction of pipeline features. To reduce model overfitting and enhance network expression capabilities, the H-swish activation function is employed to replace the original activation function. Furthermore, candidate anchor boxes are clustered using the K-means++ clustering algorithm to obtain better position regression results and improve training efficiency. The improved algorithm demonstrates significantly better positioning precision compared to the original network. Moreover, an adaptive threshold algorithm is proposed for leak detection and labelling, utilising the original temperature information contained in infrared images. The experimental results demonstrate that this method achieves higher accuracy in detecting leaks in heating pipelines.","PeriodicalId":344397,"journal":{"name":"Insight - Non-Destructive Testing and Condition Monitoring","volume":"64 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139295647","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-01DOI: 10.1784/insi.2023.65.11.625
N. U. I. Wani, C. Ranga
In this paper, the most significant diagnostic parameters of transformers and the corresponding asset management strategies adopted by the leading power utilities of various countries are discussed. The analytical techniques used include Monte Carlo simulations (MCSs), fuzzy inference systems, neuro-fuzzy systems, regression analysis, multi-criterion analysis (MCA) and so on. These techniques incorporate information related to transformer design and fabrication, operation history and maintenance, visual inspection and various diagnostic tests and measurements. The health assessment methodologies for power and distribution transformers may require improvement. Cost-effective techniques involving few critical parameters have to be used for distribution transformers. Suitable communication technologies need to be deployed for remote monitoring and assessment. Appropriate interfacing software can be developed for analysis. There are many issues and operating conditions associated with Indian transmission and distribution systems, which call for the adaptation of the discussed techniques for their effective use.
{"title":"A review of health index-based condition assessment techniques for power and distribution transformers","authors":"N. U. I. Wani, C. Ranga","doi":"10.1784/insi.2023.65.11.625","DOIUrl":"https://doi.org/10.1784/insi.2023.65.11.625","url":null,"abstract":"In this paper, the most significant diagnostic parameters of transformers and the corresponding asset management strategies adopted by the leading power utilities of various countries are discussed. The analytical techniques used include Monte Carlo simulations (MCSs), fuzzy inference systems, neuro-fuzzy systems, regression analysis, multi-criterion analysis (MCA) and so on. These techniques incorporate information related to transformer design and fabrication, operation history and maintenance, visual inspection and various diagnostic tests and measurements. The health assessment methodologies for power and distribution transformers may require improvement. Cost-effective techniques involving few critical parameters have to be used for distribution transformers. Suitable communication technologies need to be deployed for remote monitoring and assessment. Appropriate interfacing software can be developed for analysis. There are many issues and operating conditions associated with Indian transmission and distribution systems, which call for the adaptation of the discussed techniques for their effective use.","PeriodicalId":344397,"journal":{"name":"Insight - Non-Destructive Testing and Condition Monitoring","volume":"32 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139299166","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：