Structural Health Monitoring-An International Journal最新文献_第9页

Multi-fault classification of rotor systems based on phase feature of axis trajectory in noisy environments 噪声环境下基于轴轨迹相位特征的转子系统多故障分类

IF 6.6 2区工程技术

Structural Health Monitoring-An International Journal Pub Date : 2023-06-16 DOI: 10.1177/14759217231178652

Chunrong Hua, Libo Xiong, Lumei Lv, Dawei Dong, H. Ouyang

{"title":"Multi-fault classification of rotor systems based on phase feature of axis trajectory in noisy environments","authors":"Chunrong Hua, Libo Xiong, Lumei Lv, Dawei Dong, H. Ouyang","doi":"10.1177/14759217231178652","DOIUrl":"https://doi.org/10.1177/14759217231178652","url":null,"abstract":"As it is difficult to distinguish multiple rotor faults with similar dynamic phenomena in noisy environments, a multi-fault classification method is proposed by combining the extracted trajectory phase feature, a parameter-optimized variational mode decomposition (VMD) method and a light gradient boosting machine (LightGBM) model. The trajectory phase feature is extracted from an axis trajectory by fusing the frequency, amplitude, and phase information related to rotor motion and can comprehensively describe the dynamic characteristics induced by different rotor faults. First, the vibration displacement signals in two orthogonal directions are collected to construct the axis trajectories with 12 rotor states including healthy, unbalance, misalignment, single crack, multiple cracks, and a mixture of them. Second, the trajectory phase feature is extracted from the vectorized axis trajectories, and the frequency spectra of trajectory phase angles under different rotor faults are analyzed through Fourier transform. Finally, a parameter-optimized VMD method combined with a LightGBM model is applied to classify multiple faults of rotor systems in different noisy environments based on the extracted trajectory phase feature. The 12 rotor states can be classified into nine categories based on the harmonic information of 1X–7X components (X is the rotating frequency of a rotor system) and other components with smaller amplitudes in the frequency spectra of trajectory phase angles. The average classification accuracy of the 12 rotor states exceeds 93.0%, and the recognition rate for each kind of fault is greater than 77.5% in noisy environments. The simulated and experimental results demonstrate the effectiveness and adaptability of the proposed multi-fault classification method. This work can provide a reference for the condition monitoring and fault diagnosis of rotor systems in engineering.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":null,"pages":null},"PeriodicalIF":6.6,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42843053","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}

引用次数: 0

Stiffness identification of fixing device in CRTS II slab track based on long-term monitoring data 基于长期监测数据的CRTS II平板轨道固定装置刚度辨识

IF 6.6 2区工程技术

Structural Health Monitoring-An International Journal Pub Date : 2023-06-16 DOI: 10.1177/14759217231177012

Dedao Wang, Shunlong Li, Chao Lin, Senrong Wang

{"title":"Stiffness identification of fixing device in CRTS II slab track based on long-term monitoring data","authors":"Dedao Wang, Shunlong Li, Chao Lin, Senrong Wang","doi":"10.1177/14759217231177012","DOIUrl":"https://doi.org/10.1177/14759217231177012","url":null,"abstract":"The fixing device is an important component in the China Railway Track System (CRTS) II slab track, through which the longitudinal force of the track can be transferred to bridges and piers. The stiffness of the fixing device significantly affects the mechanical properties of the track. However, it is difficult to evaluate stiffness using conventional detection methods, because the fixing device is located under the base plate. In this study, the fixing-device stiffness of simply supported bridges was identified using long-term monitoring data and finite element (FE) analysis. First, a monitoring scheme was proposed based on the principle of track–bridge interaction. The strain in the base plate on both sides of the fixing device, as well as the temperature of the base plate and bridge, were obtained from 2015 to 2018. Then, the stiffness of the fixing device was proved to be closely related to the strain difference of the base plate on both sides of the fixing device, and the corresponding nonlinear mapping relationship was established through FE analysis. Finally, the fixing-device stiffness over 4 years was identified using long-term monitoring data and the established mapping function, and the influence of the stiffness degradation on the mechanical property of the track was discussed. This study can provide a reference for the design and maintenance of CRTS II slab tracks on multi-span simply supported bridges.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":null,"pages":null},"PeriodicalIF":6.6,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45562441","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}

引用次数: 0

An enhanced empirical Fourier decomposition method for bearing fault diagnosis 一种改进的经验傅里叶分解轴承故障诊断方法

IF 6.6 2区工程技术

Structural Health Monitoring-An International Journal Pub Date : 2023-06-16 DOI: 10.1177/14759217231178653

Danchen Zhu, Guoqiang Liu, Xingyu Wu, Bolong Yin

{"title":"An enhanced empirical Fourier decomposition method for bearing fault diagnosis","authors":"Danchen Zhu, Guoqiang Liu, Xingyu Wu, Bolong Yin","doi":"10.1177/14759217231178653","DOIUrl":"https://doi.org/10.1177/14759217231178653","url":null,"abstract":"To address the problem that bearing fault signals are usually contaminated by strong background interference due to multiple structures and complex transmission paths, which affects accurate fault feature extraction, an enhanced empirical Fourier decomposition technique was proposed in this paper. First, in order to weaken the influence of transmission path, the trend-line-extraction-based method was utilized in advance, which suppressed the signal distortion and background noise interference. Then, to achieve the appropriate parameter for the empirical Fourier decomposition, the correlation-coefficient-based decomposition number selection approach was constructed to avoid the existence of irrelevant modal functions. The band improvement strategy was proposed to reduce the invalid frequency bands with too narrow bandwidth during the decomposition process, the weighted harmonics significant index was utilized as the target, and the optimal modal components were also determined. Last, the fast Fourier transform was employed, and the bearing fault signatures were accurately detected. The simulation and experimental bearing fault signals were used for analysis; with the help of some comparisons, the analyzed results show that this method can effectively extract the fault characteristics of rolling element bearing from strong background interference.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":null,"pages":null},"PeriodicalIF":6.6,"publicationDate":"2023-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43687553","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}

引用次数: 0

Vibration-based looseness identification of bolted structures via quasi-analytic wavelet packet and optimized large margin distribution machine 基于准解析小波包和优化大余量分布机的螺栓结构振动松动识别

IF 6.6 2区工程技术

Structural Health Monitoring-An International Journal Pub Date : 2023-06-08 DOI: 10.1177/14759217231159948

Wenzhan Yang, Zhousuo Zhang, Xu Chen

{"title":"Vibration-based looseness identification of bolted structures via quasi-analytic wavelet packet and optimized large margin distribution machine","authors":"Wenzhan Yang, Zhousuo Zhang, Xu Chen","doi":"10.1177/14759217231159948","DOIUrl":"https://doi.org/10.1177/14759217231159948","url":null,"abstract":"Bolted joints are the most widely utilized connection types in industries, and therein looseness identification of bolted structures is of great significance to guarantee structural reliability. In this article, a comprehensive study of bolt looseness identification under random excitation is presented. To fulfill this task, this research focuses on three prominent difficulties, including nonstationary signal processing, subtle feature extraction, and robust state classification. First, a novel filter bank structure of quasi-analytic dual-tree complex wavelet packet transform is constructed to analyze the measured vibration response signals, for purpose of capturing subtle feature information. Then, multiple features are extracted from subband signals to capture the variations of dynamic characteristics, and sensitive features are selected by Laplacian score to construct the low-dimensional feature set. Subsequently, a novel classifier with better generalization performance, named large margin distribution machine, is optimized with the wavelet kernel function and the whale optimization algorithm, in order to handle the intrinsic uncertainty related to the looseness states of bolted structures. After feeding the low-dimensional feature set, the proposed classifier is trained to identify looseness states of bolted structures. Finally, experiments of a two-bolt lapped beam under random excitation are conducted to verify the effectiveness of the proposed method, and two typical loading conditions (paired-bolt looseness and single-bolt looseness) are considered. Besides, the superiority of the proposed method is demonstrated by comparing with other analogical methods. This research can provide a promising implement in practical applications of bolt looseness identification under random excitation.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":null,"pages":null},"PeriodicalIF":6.6,"publicationDate":"2023-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45461220","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}

引用次数: 0

Bandwidth-aware adaptive chirp mode decomposition for railway bearing fault diagnosis 用于铁路轴承故障诊断的带宽感知自适应线性调频模式分解

IF 6.6 2区工程技术

Structural Health Monitoring-An International Journal Pub Date : 2023-06-08 DOI: 10.1177/14759217231174699

Shiqian Chen, Lei Guo, Junjie Fan, Cai Yi, Kaiyun Wang, W. Zhai

{"title":"Bandwidth-aware adaptive chirp mode decomposition for railway bearing fault diagnosis","authors":"Shiqian Chen, Lei Guo, Junjie Fan, Cai Yi, Kaiyun Wang, W. Zhai","doi":"10.1177/14759217231174699","DOIUrl":"https://doi.org/10.1177/14759217231174699","url":null,"abstract":"It is a challenging task to accurately diagnose a railway bearing fault since bearing vibration signals are under strong interferences from wheel–rail excitations. The commonly used Kurtogram-based methods are often trapped in components induced by the wheel–rail excitations while adaptive mode decomposition methods are sensitive to input control parameters. To address these issues, based on a recently developed powerful signal decomposition method, that is, adaptive chirp mode decomposition (ACMD), a novel method called bandwidth-aware ACMD (BA-ACMD) is proposed in this article. First, the filter bank property of ACMD is thoroughly analyzed based on Monte-Carlo simulation and then a bandwidth expression with respect to the penalty parameter is first obtained by fitting a power law model. Then, a weighted spectrum trend (WST) method is proposed to partition frequency bands and then guide the parameter determination of ACMD through the integration of the obtained bandwidth expression. In addition, according to the order of magnitude of the WST in each band, the BA-ACMD adopts a recursive framework to extract signal modes one by one. In this way, dominating signal modes related to wheel–rail excitations can be extracted and then subtracted from the vibration signal in advance so that the bearing faults induced signal modes can be successfully identified. Both simulation and experimental validations are conducted showing that BA-ACMD can effectively detect single and compound faults of railway bearings under strong wheel–rail excitations.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":null,"pages":null},"PeriodicalIF":6.6,"publicationDate":"2023-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49205419","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}

引用次数: 0

Cross-correlation vibro-acoustic modulation method for damage detection 损伤检测的交叉相关振声调制方法

IF 6.6 2区工程技术

Structural Health Monitoring-An International Journal Pub Date : 2023-06-08 DOI: 10.1177/14759217231177005

Dong Liu, D. Donskoy

{"title":"Cross-correlation vibro-acoustic modulation method for damage detection","authors":"Dong Liu, D. Donskoy","doi":"10.1177/14759217231177005","DOIUrl":"https://doi.org/10.1177/14759217231177005","url":null,"abstract":"The vibroacoustic modulation (VAM) method distinguishes itself from other linear or nonlinear acoustic structure health monitoring methods due to its advantages of high sensitivity, relatively simple implementation, and the capability to detect defects in complex structures. However, the damaged index of VAM—modulation index, which is used for quantifying the damage severity, is highly influenced by the frequency of the probe wave. To increase the reliability and robustness of the VAM method, researchers modified it by replacing the single harmonic probe wave with a chirp signal. The instantaneous frequency of the chirp signal linearly increases with time, allowing the modified VAM method to access the broadband frequency range in a relatively short amount of time. However, the sideband that attributes to the acoustic nonlinearity of the defect would be overwhelmed by the broadband frequency content of the probe wave. Hence, a different algorithm is needed to process the received signal of the modified VAM method. Previous studies have employed short-time Fourier transform, wavelet transform, Hilbert transform, and synchronous demodulation for this purpose. Yet, these signal processing methods do not take into account affection of the phase difference between the probe and pump waves to the final result. To address this issue, this paper proposes using a cross-correlation (CC) approach. The performance of the proposed CC approach and existing methods were compared using simulation and experimental testing, evaluating their ability to detect acoustic nonlinearity.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":null,"pages":null},"PeriodicalIF":6.6,"publicationDate":"2023-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47578794","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}

引用次数: 0

Multi-source information fusion meta-learning network with convolutional block attention module for bearing fault diagnosis under limited dataset 基于卷积块注意力模块的多源信息融合元学习网络在有限数据集下的轴承故障诊断

IF 6.6 2区工程技术

Structural Health Monitoring-An International Journal Pub Date : 2023-06-07 DOI: 10.1177/14759217231176045

Shanshan Song, Shuqing Zhang, Wei Dong, Gaochen Li, Chengyu Pan

{"title":"Multi-source information fusion meta-learning network with convolutional block attention module for bearing fault diagnosis under limited dataset","authors":"Shanshan Song, Shuqing Zhang, Wei Dong, Gaochen Li, Chengyu Pan","doi":"10.1177/14759217231176045","DOIUrl":"https://doi.org/10.1177/14759217231176045","url":null,"abstract":"Applications in industrial production have indicated that the challenges of sparse fault samples and singular monitoring data will diminish the performance of deep learning-based diagnostic models to varying degrees. To alleviate the above issues, a multi-source information fusion meta-learning network with convolutional block attention module (CBAM) is proposed in this study for bearing fault diagnosis under limited dataset. This method can fully extract and exploit the complementary and enriched fault-related features in the multi-source monitoring data through the designed multi-branch fusion structure and incorporate metric-based meta-learning to enhance the fault diagnosis performance of the model under limited data samples. Furthermore, the introduction of CBAM can further assist the model to trade-off and focus on more discriminative information in both spatial and channel dimensions. Extensive experiments conducted on two bearing datasets that cover multi-source monitoring data fully demonstrate the validity and superiority of the proposed method.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":null,"pages":null},"PeriodicalIF":6.6,"publicationDate":"2023-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46711869","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}

引用次数: 0

CrackDenseLinkNet: a deep convolutional neural network for semantic segmentation of cracks on concrete surface images CrackDenseLinkNet：一种用于混凝土表面图像裂纹语义分割的深度卷积神经网络

IF 6.6 2区工程技术

Structural Health Monitoring-An International Journal Pub Date : 2023-05-28 DOI: 10.1177/14759217231173305

Preetham Manjunatha, S. Masri, A. Nakano, Landon Carter Wellford

{"title":"CrackDenseLinkNet: a deep convolutional neural network for semantic segmentation of cracks on concrete surface images","authors":"Preetham Manjunatha, S. Masri, A. Nakano, Landon Carter Wellford","doi":"10.1177/14759217231173305","DOIUrl":"https://doi.org/10.1177/14759217231173305","url":null,"abstract":"Cracks are the defects formed by cyclic loading, fatigue, shrinkage, creep, and so on. In addition, they represent the deterioration of the structures over some time. Therefore, it is essential to detect and classify them according to the condition grade at the early stages to prevent the collapse of structures. Deep learning-based semantic segmentation convolutional neural network (CNN) has millions of learnable parameters. However, depending on the complexity of the CNN, it takes hours to days to train the network fully. In this study, an encoder network DenseNet and modified LinkNet with five upsampling blocks were used as a decoder network. The proposed network is referred to as the “CrackDenseLinkNet” in this work. CrackDenseLinkNet has 19.15 million trainable parameters, although the input image size is 512 × 512 and has a deeper encoder. CrackDenseLinkNet and four other state-of-the-art (SOTA) methods were evaluated on three public and one private datasets. The proposed CNN, CrackDenseLinkNet, outperformed the best SOTA method, CrackSegNet, by 2.2% of F1-score on average across the four datasets. Lastly, a crack profile analysis demonstrated that the CrackDenseLinkNet has lesser variance in relative errors for the crack width, length, and area categories against the ground-truth data. The code and datasets can be downloaded at https://github.com/preethamam/CrackDenseLinkNet-DeepLearning-CrackSegmentation .","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":null,"pages":null},"PeriodicalIF":6.6,"publicationDate":"2023-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44983873","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}

引用次数: 1

A deep neural network for electrical resistance calibration of self-sensing carbon fiber polymer composites compatible with edge computing structural monitoring hardware electronics 兼容边缘计算结构监测硬件电子的自感碳纤维聚合物复合材料电阻校正深度神经网络

IF 6.6 2区工程技术

Structural Health Monitoring-An International Journal Pub Date : 2023-05-26 DOI: 10.1177/14759217231170001

M. Tomás, S. Jalali, Kiera Tabatha

{"title":"A deep neural network for electrical resistance calibration of self-sensing carbon fiber polymer composites compatible with edge computing structural monitoring hardware electronics","authors":"M. Tomás, S. Jalali, Kiera Tabatha","doi":"10.1177/14759217231170001","DOIUrl":"https://doi.org/10.1177/14759217231170001","url":null,"abstract":"The self-sensing ability of materials, in particular carbon fiber polymer composites (SSCFPC), is a must-have requirement when designing a structural monitoring network for remote assessment of structural serviceability. This work presents a study using an Artificial Deep Neural Network (ADNN) wherein is evaluated the electrical resistance ( R) output of specimens subjected to an unchanged deformation state of 2.86% strain for prolonged periods of time. Six ADNN architectures are evaluated with varying numbers of neurons on pre-defined hidden layers, sharing the same four data inputs and one output. The dataset is based on 3276 data points collected during the experimental campaign of an innovative electrode design embedded in SSCFPC specimens. The effect of the number of iterations and the architecture of the neural network is investigated in proposed ADNN models. Simple moving average, and moving Standard Deviation, [Formula: see text], are determined and plotted in terms of z-score to assist in performance evaluation of proposed ADNN models. The optimal ADNN architecture is found among six proposed architectures and for each of the four SSCFPC mixtures. Results show the proposed model architectures are able to predict values of R with greater accuracy than traditional regression mathematical methods when traditional statistical coefficients are used. However, when analyzing data in a time-series manner, results show further research is needed to achieve optimal accuracy results. The analysis presented focused on the structural monitoring network infrastructure and hardware electronics compatibility for further development of this type of SSCFPC as a self-sensing composite material with ability of automatic calibration and suitable for real-time data acquisition and artificial intelligence modeling.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":null,"pages":null},"PeriodicalIF":6.6,"publicationDate":"2023-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46920164","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}

引用次数: 1

Incremental learning BiLSTM based on dynamic proportional adjustment mechanism and experience replay for quantitative detection of blade crack propagation 基于动态比例调节机制和经验回放的增量学习BiLSTM用于叶片裂纹扩展的定量检测

IF 6.6 2区工程技术

Structural Health Monitoring-An International Journal Pub Date : 2023-05-26 DOI: 10.1177/14759217231170723

Junxian Shen, Tianchi Ma, Di Song, Feiyun Xu

{"title":"Incremental learning BiLSTM based on dynamic proportional adjustment mechanism and experience replay for quantitative detection of blade crack propagation","authors":"Junxian Shen, Tianchi Ma, Di Song, Feiyun Xu","doi":"10.1177/14759217231170723","DOIUrl":"https://doi.org/10.1177/14759217231170723","url":null,"abstract":"In the traditional quantitative detection model for blade cracks in centrifugal fan, it is assumed that the data distribution is fixed or stable. However, the new data brought by the crack propagation would break the stable distribution, thereby disturbing the old data, and resulting in a decrease in the detection performance of the model. To overcome catastrophic forgetting and reduce the extra computational cost of retaining intact old data, a quantitative detection method based on incremental learning bidirectional long short-term memory (BiLSTM) with dynamic proportional adjustment mechanism and experience replay for blade crack propagation is proposed. First, a basic BiLSTM model is constructed by inputting the data of cracks with a length of 0–5 mm. Second, the fully connected layer features in the model are selected for t-distributed stochastic neighbor embedding (t-SNE) dimensional reduction, and the Kullback–Leibler divergence is used as an indicator of feature distribution evaluating the representative old data. Third, a dynamic proportional adjustment mechanism for the old data retention proportion is constructed according to the feature distribution index and the model detection accuracy. Finally, the data of the crack with a length of 6–10 mm are gradually input to proceed with the incremental learning of the model. Verified by the measured data of the centrifugal fan, the model can adjust the retained number of old crack length data dynamically, and import new crack length data for incremental learning, making it characterized by high detection accuracy, stability, and plasticity for the quantitative detection of crack length propagation in blades.","PeriodicalId":51184,"journal":{"name":"Structural Health Monitoring-An International Journal","volume":null,"pages":null},"PeriodicalIF":6.6,"publicationDate":"2023-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46593990","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}

引用次数: 0