Nothing Special   »   [go: up one dir, main page]

You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Sign in to use this feature.

Years

Between: -

Subjects

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Journals

Article Types

Countries / Regions

Search Results (7)

Search Parameters:
Keywords = multi-bolt loosening

Order results
Result details
Results per page
Select all
Export citation of selected articles as:
16 pages, 7626 KiB  
Article
Distributed Acoustic Sensing: A Promising Tool for Finger-Band Anomaly Detection
by Kunpeng Zhang, Haochu Ku, Su Wang, Min Zhang, Xiangge He and Hailong Lu
Photonics 2024, 11(10), 896; https://doi.org/10.3390/photonics11100896 - 24 Sep 2024
Viewed by 631
Abstract
The straddle-type monorail is an electric-powered public vehicle widely known for its versatility and ease of maintenance. The finger-band is a critical connecting structure for the straddle-type monorail, but issues such as loose bolts are inevitable over time. Manual inspection is the primary [...] Read more.
The straddle-type monorail is an electric-powered public vehicle widely known for its versatility and ease of maintenance. The finger-band is a critical connecting structure for the straddle-type monorail, but issues such as loose bolts are inevitable over time. Manual inspection is the primary method for detecting bolt looseness in the finger-band, but this approach could be more efficient and resistant to missed detections. In this study, we conducted a straddle-type monorail finger-band-anomaly-monitoring experiment using Distributed Acoustic Sensing (DAS), a distributed multi-point-monitoring system widely used in railway monitoring. We analyzed track vibration signals’ time-domain and frequency-domain characteristics under different monorail operating conditions. Our findings revealed the following: 1. DAS can effectively identify the monorail’s operating status, including travel direction, starting and braking, and real-time train speed measurement. 2. Time-domain signals can accurately pinpoint special track structures such as turnouts and finger-bands. Passing trains over finger-bands also results in notable energy reflections in the frequency domain. 3. After the finger-band bolts loosen, there is a significant increase in vibration energy at the finger-band position, with the degree of energy increase corresponding to the extent of loosening. Full article
(This article belongs to the Special Issue Distributed Optical Fiber Sensing Technology)
Show Figures

Figure 1

Figure 1
<p>HD-DAS device illustration.</p>
Full article ">Figure 2
<p>Structure of HD-DAS system (Redrawn from Ref. [<a href="#B21-photonics-11-00896" class="html-bibr">21</a>]).</p>
Full article ">Figure 3
<p>Actual photo of experimental factory area: white dashed lines indicate finger-bands, green dashed boxes indicate turnouts, and white arrows point to fiber optics.</p>
Full article ">Figure 4
<p>Schematic of the experimental rail system.</p>
Full article ">Figure 5
<p>Field image of fiber optic cable deployment.</p>
Full article ">Figure 6
<p>Actual photo of finger-band assembly: (<b>a</b>) shows the overall finger-band assembly, while (<b>b</b>) is an enlarged view highlighting the finger-band details. In (<b>b</b>), the white dashed lines indicate loosened screws, with the screws of both pairs of finger-bands being loosened to the same extent.</p>
Full article ">Figure 7
<p>DAS waterfall plots at 3 km/h and 4.5 km/h speeds.</p>
Full article ">Figure 8
<p>WSST diagram and relative position of monorail and finger-band: (<b>a</b>) bolts fastened, (<b>b</b>) one buckle loosened, (<b>c</b>) two buckles loosened. (<b>a</b>–<b>c</b>) represent the configurations with the bolt fastened, one buckle unfastened, and two buckles unfastened, respectively. In (<b>c</b>), the two red dashed lines divide the WSST into three sections: approaching, crossing, and passed. (<b>d</b>) shows the relative position of the monorail to the finger-band, corresponding to the three sections in (<b>c</b>).</p>
Full article ">Figure 9
<p>PSD for different bolt-loosening conditions.</p>
Full article ">Figure 10
<p>Accumulated PSD across different frequency ranges for various bolt-loosening conditions: ΣPSD represents accumulated PSD.</p>
Full article ">Figure 11
<p>RMS across different frequency ranges for various bolt-loosening conditions: (<b>a</b>) full frequency range and (<b>b</b>) 20–50 Hz frequency band.</p>
Full article ">Figure 12
<p>Signal-processing flowchart.</p>
Full article ">Figure 13
<p>DAS waterfall plot for experimental system’s noise floor.</p>
Full article ">Figure 14
<p>DAS PSD plot for experimental system’s noise floor.</p>
Full article ">
22 pages, 4973 KiB  
Article
Escalator Foundation Bolt Loosening Fault Recognition Based on Empirical Wavelet Transform and Multi-Scale Gray-Gradient Co-Occurrence Matrix
by Xuezhuang E and Wenbo Wang
Sensors 2023, 23(15), 6801; https://doi.org/10.3390/s23156801 - 30 Jul 2023
Viewed by 1089
Abstract
An escalator is an essential large-scale public transport equipment; once it fails, this inevitably affects the operation of the escalator and even leads to safety concerns, or perhaps accidents. As an important structural part of the escalator, the foundation of the main engine [...] Read more.
An escalator is an essential large-scale public transport equipment; once it fails, this inevitably affects the operation of the escalator and even leads to safety concerns, or perhaps accidents. As an important structural part of the escalator, the foundation of the main engine can cause the operation of the escalator to become abnormal when its fixing bolts become loose. Aiming to reduce the difficulty of extracting the fault features of the footing bolt when it loosens, a fault feature extraction method is proposed in this paper based on empirical wavelet transform (EWT) and the gray-gradient co-occurrence matrix (GGCM). Firstly, the Teager energy operator and multi-scale peak determination are used to improve the spectral partitioning ability of EWT, and the improved EWT is used to decompose the original foundation vibration signal into a series of empirical mode functions (EMFs). Then, the gray-gradient co-occurrence matrix of each EMF is constructed, and six texture features of the gray-gradient co-occurrence matrix are calculated as the fault feature vectors of this EMF. Finally, the fault features of all EMFs are fused, and the degree of the loosening of the escalator foundation bolt is identified using the fused multi-scale feature vector and BiLSTM. The experimental results show that the proposed method based on EWT and GGCM feature extraction can diagnose the loosening degree of foundation bolts more effectively and has a certain engineering application value. Full article
(This article belongs to the Section Fault Diagnosis & Sensors)
Show Figures

Figure 1

Figure 1
<p>Simulated signal and FFT spectrum.</p>
Full article ">Figure 2
<p>Comparison of spectrum division results of EWT.</p>
Full article ">Figure 3
<p>LSTM network structure diagram.</p>
Full article ">Figure 4
<p>BI-LSTM network structure diagram.</p>
Full article ">Figure 5
<p>Original machine feet vibration signal.</p>
Full article ">Figure 6
<p>Results of EWT decomposition. (<b>a</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>1</mn> </msub> </mrow> </semantics></math>; (<b>b</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>2</mn> </msub> </mrow> </semantics></math>; (<b>c</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>3</mn> </msub> </mrow> </semantics></math>; (<b>d</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>4</mn> </msub> </mrow> </semantics></math>.</p>
Full article ">Figure 7
<p>Contour of normal (<b>a</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>2</mn> </msub> </mrow> </semantics></math>; (<b>b</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>3</mn> </msub> </mrow> </semantics></math>; (<b>c</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>4</mn> </msub> </mrow> </semantics></math>.</p>
Full article ">Figure 8
<p>Contour of loosening 1 lap: (<b>a</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>2</mn> </msub> </mrow> </semantics></math>; (<b>b</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>3</mn> </msub> </mrow> </semantics></math>; (<b>c</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>4</mn> </msub> </mrow> </semantics></math>.</p>
Full article ">Figure 9
<p>Contour of loosening 2 laps (<b>a</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>2</mn> </msub> </mrow> </semantics></math>; (<b>b</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>3</mn> </msub> </mrow> </semantics></math>; (<b>c</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>4</mn> </msub> </mrow> </semantics></math>.</p>
Full article ">Figure 10
<p>Contour of loosening 3 laps (<b>a</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>2</mn> </msub> </mrow> </semantics></math>; (<b>b</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>3</mn> </msub> </mrow> </semantics></math>; (<b>c</b>) <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>EMF</mi> </mrow> <mn>4</mn> </msub> </mrow> </semantics></math>.</p>
Full article ">Figure 11
<p>Graphs of six feature vectors based on GGCM.</p>
Full article ">Figure 12
<p>The fused 18-dimensional GGCM feature vector.</p>
Full article ">Figure 13
<p>Process diagram of bolt loosening fault diagnosis.</p>
Full article ">Figure 14
<p>Histogram of the classification performance of four methods.</p>
Full article ">Figure 15
<p>Confusion matrix from ten-fold CV with four methods.</p>
Full article ">Figure 16
<p>Histogram of the classification performance of four ML methods.</p>
Full article ">Figure 17
<p>Confusion matrix from ten-fold CV with four ML methods.</p>
Full article ">
22 pages, 8007 KiB  
Article
Structural Nonlinear Damage Identification Method Based on the Kullback–Leibler Distance of Time Domain Model Residuals
by Heng Zuo and Huiyong Guo
Remote Sens. 2023, 15(4), 1135; https://doi.org/10.3390/rs15041135 - 19 Feb 2023
Cited by 2 | Viewed by 1633
Abstract
Under external load excitation, damage such as breathing cracks and bolt loosening will cause structural time domain acceleration to have nonlinear features. To solve the problem of time domain nonlinear damage identification, a damage identification method based on the Kullback–Leibler (KL) distance of [...] Read more.
Under external load excitation, damage such as breathing cracks and bolt loosening will cause structural time domain acceleration to have nonlinear features. To solve the problem of time domain nonlinear damage identification, a damage identification method based on the Kullback–Leibler (KL) distance of time domain model residuals is proposed in this paper. First, an autoregressive (AR) model order was selected using the autocorrelation function (ACF) and Akaike information criterion (AIC). Then, an AR model was obtained based on the structural acceleration response time series, and the AR model residual was extracted. Finally, the KL distance was used as a damage indicator to judge the structural damage source location. The effectiveness of the proposed method was verified by using a multi-story, multi-span stand model experiment and a simulated eight-story shear structure. The results show that the proposed structural nonlinear damage identification method can effectively distinguish the structural damage location of multi-degree-of-freedom shear structures and complex stand structures, and it is robust enough to detect environmental noise and small damage. Full article
(This article belongs to the Special Issue Remote Sensing in Structural Health Monitoring)
Show Figures

Figure 1

Figure 1
<p>Damage identification using the KL distance of the AR model residual.</p>
Full article ">Figure 2
<p>Simulation of eight-story shear structure model.</p>
Full article ">Figure 3
<p>White noise loading curves.</p>
Full article ">Figure 3 Cont.
<p>White noise loading curves.</p>
Full article ">Figure 4
<p>Acceleration curves of baseline state with excitation amplitude of 250 kN.</p>
Full article ">Figure 4 Cont.
<p>Acceleration curves of baseline state with excitation amplitude of 250 kN.</p>
Full article ">Figure 5
<p>ACF curves of the baseline state with excitation amplitude of 250 kN.</p>
Full article ">Figure 6
<p>Damage identification results of damage scenarios 1–8.</p>
Full article ">Figure 6 Cont.
<p>Damage identification results of damage scenarios 1–8.</p>
Full article ">Figure 7
<p>Damage identification results of damage scenarios 9–16.</p>
Full article ">Figure 7 Cont.
<p>Damage identification results of damage scenarios 9–16.</p>
Full article ">Figure 8
<p>Damage identification results of damage scenarios 17–24.</p>
Full article ">Figure 8 Cont.
<p>Damage identification results of damage scenarios 17–24.</p>
Full article ">Figure 9
<p>The stand structure experimental model.</p>
Full article ">Figure 10
<p>The region divisions and damaged brace locations in the experimental model.</p>
Full article ">Figure 11
<p>The acceleration curves of the baseline state.</p>
Full article ">Figure 12
<p>The ACF curves of the baseline state.</p>
Full article ">Figure 12 Cont.
<p>The ACF curves of the baseline state.</p>
Full article ">Figure 13
<p>The damage identification results of the stand structure model experiment.</p>
Full article ">Figure 13 Cont.
<p>The damage identification results of the stand structure model experiment.</p>
Full article ">
18 pages, 4290 KiB  
Article
Loosening Identification of Multi-Bolt Connections Based on Wavelet Transform and ResNet-50 Convolutional Neural Network
by Xiao-Xue Li, Dan Li, Wei-Xin Ren and Jun-Shu Zhang
Sensors 2022, 22(18), 6825; https://doi.org/10.3390/s22186825 - 9 Sep 2022
Cited by 22 | Viewed by 2503
Abstract
A high-strength bolt connection is the key component of large-scale steel structures. Bolt loosening and preload loss during operation can reduce the load-carrying capacity, safety, and durability of the structures. In order to detect loosening damage in multi-bolt connections of large-scale civil engineering [...] Read more.
A high-strength bolt connection is the key component of large-scale steel structures. Bolt loosening and preload loss during operation can reduce the load-carrying capacity, safety, and durability of the structures. In order to detect loosening damage in multi-bolt connections of large-scale civil engineering structures, we proposed a multi-bolt loosening identification method based on time-frequency diagrams and a convolutional neural network (CNN) using vi-bro-acoustic modulation (VAM) signals. Continuous wavelet transform was employed to obtain the time-frequency diagrams of VAM signals as the features. Afterward, the CNN model was trained to identify the multi-bolt loosening conditions from the raw time-frequency diagrams intelligently. It helps to get rid of the dependence on traditional manual selection of simplex and ineffective damage index and to eliminate the influence of operational noise of structures on the identification accuracy. A laboratory test was carried out on bolted connection specimens with four high-strength bolts of different degrees of loosening. The effects of different excitations, CNN models, and dataset sizes were investigated. We found that the ResNet-50 CNN model taking time-frequency diagrams of the hammer excited VAM signals, as the input had better performance in identifying the loosened bolts with various degrees of loosening at different positions. The results indicate that the proposed multi-bolt loosening identification method based on VAM and ResNet-50 CNN can identify bolt loosening with a reasonable accuracy, computational efficiency, and robustness. Full article
Show Figures

Figure 1

Figure 1
<p>Principle of vibro-acoustic modulation (VAM) technique.</p>
Full article ">Figure 2
<p>ResNet-50 residual block.</p>
Full article ">Figure 3
<p>ResNet-50 neural network configuration.</p>
Full article ">Figure 4
<p>Multi-bolt loosening identification flowchart.</p>
Full article ">Figure 5
<p>Experimental setup.</p>
Full article ">Figure 6
<p>Time-frequency diagram obtained using different test methods excitation frequencies: (<b>a</b>) results of vibration generator test under case J; (<b>b</b>) results of hammer test under case J; (<b>c</b>) results of vibration generator test under case A; (<b>d</b>) results of hammer test under case A.</p>
Full article ">Figure 7
<p>Image augmentation.</p>
Full article ">Figure 8
<p>Comparison of training results between ResNet-50 and other CNN models: (<b>a</b>) training accuracy; (<b>b</b>) validation accuracy; (<b>c</b>) training loss; and (<b>d</b>) validation loss.</p>
Full article ">Figure 9
<p>Comparison of different excitation forms on model performance of ResNet-50 model. (<b>a</b>) training accuracy; (<b>b</b>) validation accuracy; (<b>c</b>) training loss; and (<b>d</b>) validation loss.</p>
Full article ">Figure 10
<p>Confusion matrix.</p>
Full article ">
12 pages, 3345 KiB  
Article
Investigation on Vibration Signal Characteristics in a Centrifugal Pump Using EMD-LS-MFDFA
by Xing Liang, Yuanxing Luo, Fei Deng and Yan Li
Processes 2022, 10(6), 1169; https://doi.org/10.3390/pr10061169 - 10 Jun 2022
Cited by 5 | Viewed by 1861
Abstract
Vibration signals from centrifugal pumps are nonlinear, non-smooth, and possess implied trend terms, which makes it difficult for traditional signal processing methods to accurately extract their fault characteristics and details. With a view to rectifying this, we introduced empirical mode decomposition (EMD) to [...] Read more.
Vibration signals from centrifugal pumps are nonlinear, non-smooth, and possess implied trend terms, which makes it difficult for traditional signal processing methods to accurately extract their fault characteristics and details. With a view to rectifying this, we introduced empirical mode decomposition (EMD) to extract the trend term signals. These were then refit using the least squares (LS) method. The result (EMD-LS) was then combined with multi-fractal theory to form a new signal identification method (EMD-LS-MFDFA), whose accuracy was verified with a binomial multi-fractal sequence (BMS). Then, based on the centrifugal pump test platform, the vibration signals of shell failures under different degrees of cavitation and separate states of loosened foot bolts were collected. The signals’ multi-fractal spectra parameters were analyzed using the EMD-LS-MFDFA method, from which five spectral parameters (Δα, Δf, α0, αmax, and αmin) were extracted for comparison and analysis. The results showed EMD-LS-MFDFA’s performance was closer to the BMS theoretical value than that of MFDFA, displayed high accuracy, and was fully capable of revealing the multiple fractal characteristics of the centrifugal pump fault vibration signal. Additionally, the mean values of the five types of multi-fractal spectral characteristic parameters it extracted were much greater than the normal state values. This indicates that the parameters could effectively distinguish the normal state and fault state of the centrifugal pump. Moreover, α0 and αmax had a smaller mean square than Δα, Δf and αmin, and their stability was higher. Thus, compared to the feature parameters extracted by MFDFA, our method could better realize the separation between the normal state, cavitation (whether slight, moderate, or severe), and when the anchor bolt was loose. This can be used to characterize centrifugal pump failure, quantify and characterize a pump’s different working states, and provide a meaningful reference for the diagnosis and study of pump faults. Full article
(This article belongs to the Special Issue Design and Optimization Method of Pumps)
Show Figures

Figure 1

Figure 1
<p>The relationship between Hurst index and quality index and theoretical value: (<b>a</b>) <math display="inline"><semantics> <mrow> <mo>Δ</mo> <mi>h</mi> <mo stretchy="false">(</mo> <mi>q</mi> <mo stretchy="false">)</mo> <mo>~</mo> <mi>q</mi> </mrow> </semantics></math>, (<b>b</b>) <math display="inline"><semantics> <mrow> <mo>Δ</mo> <mi>τ</mi> <mo stretchy="false">(</mo> <mi>q</mi> <mo stretchy="false">)</mo> <mo>~</mo> <mi>q</mi> </mrow> </semantics></math>.</p>
Full article ">Figure 2
<p>Laboratory centrifugal circulating water pump device.</p>
Full article ">Figure 3
<p>Energy spectrum of centrifugal pump vibration sensor signals.</p>
Full article ">Figure 4
<p>Quality index of centrifugal pump vibration signals.</p>
Full article ">Figure 5
<p>Generalized Hurst exponent of centrifugal pump vibration signals.</p>
Full article ">Figure 6
<p>Multi-fractal spectrum of centrifugal pump vibration signals.</p>
Full article ">Figure 7
<p>Stability of the characteristic parameters of the centrifugal pump vibration signals: (<b>a</b>) <math display="inline"><semantics> <mrow> <mo>Δ</mo> <mi>α</mi> </mrow> </semantics></math>, (<b>b</b>) <math display="inline"><semantics> <mrow> <mo>Δ</mo> <mi>f</mi> </mrow> </semantics></math>, (<b>c</b>) <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mn>0</mn> </msub> </mrow> </semantics></math>, (<b>d</b>) <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mrow> <mi>max</mi> </mrow> </msub> </mrow> </semantics></math>, (<b>e</b>) <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mrow> <mi>min</mi> </mrow> </msub> </mrow> </semantics></math>.</p>
Full article ">Figure 8
<p>Classification and comparison of centrifugal pump vibration signals by <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mn>0</mn> </msub> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mrow> <mi>max</mi> </mrow> </msub> </mrow> </semantics></math> of different methods: (<b>a</b>) <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mn>0</mn> </msub> </mrow> </semantics></math> from MFDFA, (<b>b</b>) <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mrow> <mi>max</mi> </mrow> </msub> </mrow> </semantics></math> from MFDFA, (<b>c</b>) <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mn>0</mn> </msub> </mrow> </semantics></math> from EMD-LS-MFDFA, (<b>d</b>) <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mrow> <mi>max</mi> </mrow> </msub> </mrow> </semantics></math> from EMD-LS-MFDFA.</p>
Full article ">Figure 8 Cont.
<p>Classification and comparison of centrifugal pump vibration signals by <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mn>0</mn> </msub> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mrow> <mi>max</mi> </mrow> </msub> </mrow> </semantics></math> of different methods: (<b>a</b>) <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mn>0</mn> </msub> </mrow> </semantics></math> from MFDFA, (<b>b</b>) <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mrow> <mi>max</mi> </mrow> </msub> </mrow> </semantics></math> from MFDFA, (<b>c</b>) <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mn>0</mn> </msub> </mrow> </semantics></math> from EMD-LS-MFDFA, (<b>d</b>) <math display="inline"><semantics> <mrow> <msub> <mi>α</mi> <mrow> <mi>max</mi> </mrow> </msub> </mrow> </semantics></math> from EMD-LS-MFDFA.</p>
Full article ">
14 pages, 4505 KiB  
Article
Vibration-Based Loosening Detection of a Multi-Bolt Structure Using Machine Learning Algorithms
by Oybek Eraliev, Kwang-Hee Lee and Chul-Hee Lee
Sensors 2022, 22(3), 1210; https://doi.org/10.3390/s22031210 - 5 Feb 2022
Cited by 19 | Viewed by 4214
Abstract
Since artificial intelligence (AI) was introduced into engineering fields, it has made many breakthroughs. Machine learning (ML) algorithms have been very commonly used in structural health monitoring (SHM) systems in the last decade. In this study, a vibration-based early stage of bolt loosening [...] Read more.
Since artificial intelligence (AI) was introduced into engineering fields, it has made many breakthroughs. Machine learning (ML) algorithms have been very commonly used in structural health monitoring (SHM) systems in the last decade. In this study, a vibration-based early stage of bolt loosening detection and identification technique is proposed using ML algorithms, for a motor fastened with four bolts (M8 × 1.5) to a stationary support. First, several cases with fastened and loosened bolts were established, and the motor was operated in three different types of working condition (800 rpm, 1000 rpm, and 1200 rpm), in order to obtain enough vibration data. Second, for feature extraction of the dataset, the short-time Fourier transform (STFT) method was performed. Third, different types of classifier of ML were trained, and a new test dataset was applied to evaluate the performance of the classifiers. Finally, the classifier with the greatest accuracy was identified. The test results showed that the capability of the classifier was satisfactory for detecting bolt loosening and identifying which bolt or bolts started to lose their preload in each working condition. The identified classifier will be implemented for online monitoring of the early stage of bolt loosening of a multi-bolt structure in future works. Full article
Show Figures

Figure 1

Figure 1
<p>The flowchart of the proposed method for bolt-loosening detection.</p>
Full article ">Figure 2
<p>(<b>a</b>) Experimental setup; (<b>b</b>) scheme of top view of motor and position of bolt specimens and sensors; (<b>c</b>) torque wrench.</p>
Full article ">Figure 3
<p>Rows of vibration signals of the dataset corresponding to all cases.</p>
Full article ">Figure 4
<p>The spectrograms of short-time Fourier transform coefficients for all cases.</p>
Full article ">Figure 5
<p>Accuracy of classifiers for both sensors with 513 features.</p>
Full article ">Figure 6
<p>Training time of classifiers with 513 features (sensor 2).</p>
Full article ">Figure 7
<p>Representation of the feature reduction method for all cases.</p>
Full article ">Figure 8
<p>Accuracy of classifiers with 25 features (sensor 2).</p>
Full article ">Figure 9
<p>Training time of classifiers with 25 features (sensor 2).</p>
Full article ">Figure 10
<p>Confusion matrix of RF classifier.</p>
Full article ">
19 pages, 7292 KiB  
Article
Detection Method for Bolted Connection Looseness at Small Angles of Timber Structures based on Deep Learning
by Yabin Yu, Ying Liu, Jiawei Chen, Dong Jiang, Zilong Zhuang and Xiaoli Wu
Sensors 2021, 21(9), 3106; https://doi.org/10.3390/s21093106 - 29 Apr 2021
Cited by 27 | Viewed by 2616
Abstract
Bolted connections are widely used in timber structures. Bolt looseness is one of the most important factors leading to structural failure. At present, most of the detection methods for bolt looseness do not achieve a good balance between cost and accuracy. In this [...] Read more.
Bolted connections are widely used in timber structures. Bolt looseness is one of the most important factors leading to structural failure. At present, most of the detection methods for bolt looseness do not achieve a good balance between cost and accuracy. In this paper, the detection method of small angle of bolt loosening in a timber structure is studied using deep learning and machine vision technology. Firstly, three schemes are designed, and the recognition targets are the nut’s own specification number, rectangular mark, and circular mark, respectively. The Single Shot MultiBox Detector (SSD) algorithm is adopted to train the image datasets. The scheme with the smallest identification angle error is the one identifying round objects, of which the identification angle error is 0.38°. Then, the identification accuracy was further improved, and the minimum recognition angle reached 1°. Finally, the looseness in a four-bolted connection and an eight-bolted connection are tested, confirming the feasibility of this method when applied on multi-bolted connection, and realizing a low operating costing and high accuracy. Full article
(This article belongs to the Section Fault Diagnosis & Sensors)
Show Figures

Figure 1

Figure 1
<p>Experimental equipment of (<b>a</b>) bolt loosening and (<b>b</b>) curve of bolt looseness.</p>
Full article ">Figure 2
<p>Setup of image collection.</p>
Full article ">Figure 3
<p>Collected images of the three schemes (<b>A</b>–<b>C</b>).</p>
Full article ">Figure 4
<p>Rotation angle calculation process.</p>
Full article ">Figure 5
<p>Single Shot MultiBox Detector (SSD) network structure.</p>
Full article ">Figure 6
<p>Features in multi-scales.</p>
Full article ">Figure 7
<p>Prior boxes in a feature image.</p>
Full article ">Figure 8
<p>Setup of angle measurement.</p>
Full article ">Figure 9
<p>Best identification of rotation angles: (<b>a</b>) 50° and 150°; (<b>b</b>) 250° and 350°.</p>
Full article ">Figure 9 Cont.
<p>Best identification of rotation angles: (<b>a</b>) 50° and 150°; (<b>b</b>) 250° and 350°.</p>
Full article ">Figure 10
<p>Identification results of scheme A: (<b>a</b>) initial state of a nut; (<b>b</b>) state of a nut after rotation.</p>
Full article ">Figure 11
<p>Verification results of lighting condition and shadow influence: (<b>a</b>) objects shot under weak lighting condition; (<b>b</b>) identification of the objects; (<b>c</b>) object shot under dark condition; (<b>d</b>) identification result of the object shot under camera flash.</p>
Full article ">Figure 12
<p>Loss decline curve.</p>
Full article ">Figure 13
<p>Best identification results at a tiny angle.</p>
Full article ">Figure 14
<p>Four-bolted connection.</p>
Full article ">Figure 15
<p>Identification results of a four-bolted connection.</p>
Full article ">Figure 16
<p>Eight-bolted connection.</p>
Full article ">Figure 17
<p>Identification results of an eight-bolted connection.</p>
Full article ">Figure 18
<p>Secondary identification results of a multi-bolted connection.</p>
Full article ">Figure 19
<p>Identification results of an eight-bolted connection after cropping.</p>
Full article ">
Back to TopTop