An efficient robotic-assisted bolt-ball joint looseness monitoring approach using CBAM-enhanced lightweight ResNet

Li, Li; Yuan, Rui; Lv, Yong; Xu, Shijie; Hu, Huangxing; Song, Gangbing

doi:10.1088/1361-665x/ad06e0

Cited by 5 publications

(4 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Markov process modeling of channel pruning State s k represents the preservation of the kth channel, where the transition pk represents the probability of retaining the channels, and is 1 − p k the probability of terminating this process. State represents k − 1 the terminal state, and Cout is the maximum number of channels in each layer [40][41][42].…”

Section: Target Detection Algorithm Network Pruningmentioning

confidence: 99%

A novel robotic-assisted deep learning-enabled computer vision approach for nondestructive diagnosis of railway bolt faults

Hua,

Wang,

Han

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

Railways play a vital role in the inland transportation system worldwide, and abnormal bolt components at the track joints are the main cause of train accidents. The detection and identification of faults in rail bolt components are of considerable research importance. To address this problem, numerous researchers have opted for computer vision-based methods to accomplish the detection and identification of the target, but the existing methods have poor detection performance diminished detection capabilities when the target position changes or some feature information is occluded, and the detection speed and accuracy are far from meeting the requirements of practical applications. Therefore, based on the construction of a dedicated dataset for bolt components, this paper uses the K-means dimensional clustering algorithm to re-cluster the dataset according to the target size characteristics, with the aim of reduce the bounding box regression error. At the same time, a novel loss function iteration method is proposed by incorporating an adaptive optimization algorithm, in order to improve the detection speed and ensure good convergence, and the model complexity is reduced based on deep model pruning. Finally, the optimized detection model is implemented on the robotic-assisted platform for testing, and the experimental results verify that the algorithm can quickly and accurately complete various fault diagnosis tasks of bolt components in practical applications. The main achievements of this study include the construction of a large-scale image dataset for novel rail bolt components and propelled the application of deep learning methods in vision-based rail bolt fault diagnosis.

show abstract

Section: Target Detection Algorithm Network Pruningmentioning

confidence: 99%

A novel robotic-assisted deep learning-enabled computer vision approach for nondestructive diagnosis of railway bolt faults

Hua,

Wang,

Han

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…These characteristics can be attributed to various factors, including friction, damping, stiffness, and coupling [5]. In recent years, there has been notable development in nonlinear dynamics, and related methods such as wavelet transform (WT) [6], empirical mode decomposition (EMD) [7], and variational mode decomposition (VMD) [8] have been extensively utilized in bearing fault diagnosis [9][10][11][12][13][14]. Although WT is effective in extracting bearing fault features from vibration signal, it heavily relies on human engineering experience for choosing wavelet basis functions, making it less adaptable [15].…”

Section: Introductionmentioning

confidence: 99%

Improved multiscale coded dispersion entropy: a novel quadratic-coded health indicator of rolling bearings

Fan,

Yuan,

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

Dispersion entropy (DE) is widely used to quantify the complexity of nonlinear time series. In order to improve the ability to capture fault characteristics, a novel approach called coded dispersion entropy (CDE) has been introduced in recent years. CDE aims to expand the number of possible dispersion patterns and enhance the encoding of similar dispersion patterns. However, the coding method of CDE ignores the amplitude differences between dispersion elements and average elements, resulting in the inaccurate assignment of samples. Additionally, CDE is unable to extract effective information from other time scales. These limitations hinder the ability of CDE to effectively characterize faults. This paper proposes an improved multiscale coded dispersion entropy (IMCDE) to solve these limitations. The method introduces the interval scaling factor "R" and utilizes the sum and difference between the mean element and R as new coding boundaries. This approach addresses the sensitivity issue of the CDE coding mode towards smaller amplitude differences by the reasonable quadratic coding mode. Additionally, a composite coarse-graining process is introduced to rearrange the first coarse-grained point in sequence, resulting in multiple sets of sequences. The average probability of the same dispersion pattern at each scale is calculated to correct the entropy error. The experimental results from two sets of bearing faults demonstrate the effectiveness of this method in extracting critical features associated with faulty bearings. Furthermore, the method showed higher classification accuracy compared to multiscale dispersion entropy (MDE) and multiscale coded dispersion entropy (MCDE). Additionally, it exhibited smaller classification error than MDE and MCDE.

show abstract

“…For instance, Wang et al [36] adopted a robotic arm to obtain percussion-induced audio data from a spatial bolt-ball joint and a memory-augmented neural network for its looseness classification. Similarly, Li et al [37] extracted audio signals from a spatial joint using a robot and classified its bolt looseness using a lightweight convolutional block attention module-based CNN model. On the other hand, He et al [38] extracted audio signals from an underwater flange using a hydrophone and applied a k-nearest neighbor algorithm for its looseness recognition.…”

Section: Introductionmentioning

confidence: 99%

Addressing data scarcity using audio signal augmentation and deep learning for bolt looseness prediction

Chelimilla,

Chinthapenta,

Korla

2024

Smart Mater. Struct.

View full text Add to dashboard Cite

Deep learning models such as convolutional neural networks (CNNs) encounter challenges, including instability and overfitting, while predicting bolt looseness in data-scarce scenarios. In this study, we proposed a novel audio signal augmentation approach to classify bolt looseness in the event of data deficiency using CNN models. Audio signals at varied bolt torque conditions were extracted using the percussion method. Audio signal augmentation was performed using signal shifting and scaling strategies after segmenting the extracted audio signals. The unaugmented and augmented audio signals were transformed into scalograms using the continuous wavelet transform approach to train the CNN models. Upon training with augmented datasets, a promising improvement in the loss and accuracy of the CNN models in recognizing bolt looseness was noticed. One of the significant observations from the current study is that the implementation of audio signal augmentation improved the extrinsic generalization ability of the CNN models to classify bolt looseness. A maximum increase of 73.5% to identify bolt looseness in novel data was exhibited as compared to without augmentation. Overall, a maximum accuracy of 94.5% to classify bolt looseness in unseen data was demonstrated upon audio signal augmentation. In summary, the results affirm that the audio signal augmentation approach empowered the CNN models to predict bolt looseness in data-deficient scenarios accurately.

show abstract

An efficient robotic-assisted bolt-ball joint looseness monitoring approach using CBAM-enhanced lightweight ResNet

Cited by 5 publications

References 52 publications

A novel robotic-assisted deep learning-enabled computer vision approach for nondestructive diagnosis of railway bolt faults

A novel robotic-assisted deep learning-enabled computer vision approach for nondestructive diagnosis of railway bolt faults

Improved multiscale coded dispersion entropy: a novel quadratic-coded health indicator of rolling bearings

Addressing data scarcity using audio signal augmentation and deep learning for bolt looseness prediction

Contact Info

Product

Resources

About