A Convolutional Autoencoder Based Fault Detection Method for Metro Railway Turnout

Chen, Chen; Li, Xingqiu; Huang, Kai; Xu, Zhiai; Meng, Ming

doi:10.32604/cmes.2023.024033

Cited by 6 publications

(6 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where a (1) mp , a (2) mp , and a (3) mp represent the modes of tensor samples with CP decomposition. R is an integer, and tensor A is a rank-one tensor when R = 1.…”

Section: Tensor Learningmentioning

confidence: 99%

“…The computational complexity of CNN is similar to that of CAE. For the dataset {χ +i ∈ R I 1 ×I 2 ×...×I N } l + i=1 and {χ −i ∈ R I 1 ×I 2 ×...×I N } l − i=1 , the computational complexity of STM is O (l + + l − ) 2 NT N n=1 I n , T represents the number of iterations for the alternating projection method. FFCHTM's computational complexity is O (l + + l − ) 2 R 2 N n=1 I n .…”

Section: Complexity Comparisonmentioning

confidence: 99%

“…1 illustrates the relationship between the railway switch machine and the switch & crossing system. The primary function of the railway switch machine is to control track switching and alter the train's direction [2,3]. However, the railway switch machine is susceptible to damage since it is exposed to harsh weather or wet tunnel conditions.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Fault Diagnosis Scheme for Railway Switch Machine Using Multi-Sensor Fusion Tensor Machine

Chen,

Xu,

Mei

et al. 2024

CMC

View full text Add to dashboard Cite

Railway switch machine is essential for maintaining the safety and punctuality of train operations. A data-driven fault diagnosis scheme for railway switch machine using tensor machine and multi-representation monitoring data is developed herein. Unlike existing methods, this approach takes into account the spatial information of the time series monitoring data, aligning with the domain expertise of on-site manual monitoring. Besides, a multisensor fusion tensor machine is designed to improve single signal data's limitations in insufficient information. First, one-dimensional signal data is preprocessed and transformed into two-dimensional images. Afterward, the fusion feature tensor is created by utilizing the images of the three-phase current and employing the CANDE-COMP/PARAFAC (CP) decomposition method. Then, the tensor learning-based model is built using the extracted fusion feature tensor. The developed fault diagnosis scheme is valid with the field three-phase current dataset. The experiment indicates an enhanced performance of the developed fault diagnosis scheme over the current approach, particularly in terms of recall, precision, and F1-score.

show abstract

“…where a (1) mp , a (2) mp , and a (3) mp represent the modes of tensor samples with CP decomposition. R is an integer, and tensor A is a rank-one tensor when R = 1.…”

Section: Tensor Learningmentioning

confidence: 99%

Section: Complexity Comparisonmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fault Diagnosis Scheme for Railway Switch Machine Using Multi-Sensor Fusion Tensor Machine

Chen,

Xu,

Mei

et al. 2024

CMC

View full text Add to dashboard Cite

show abstract

“…The intelligent maintenance of railway equipment has garnered increasing attention as a way to enhance sustainable transportation and manufacturing [1,2]. As an essential topic in prognostics and health management (PHM), fault diagnosis can help reduce the workload for inspectors and enhance the efficiency of traditional regular inspections [3,4].…”

Section: Introductionmentioning

confidence: 99%

“…The primary approach to diagnosing faults in railway equipment relies on the use of train monitoring data [9]. Turnout fault diagnosis methods generally involve data acquisition, feature extraction, and pattern recognition [10,11]. From a feature construction perspective, these methods can be divided into three categories: pattern recognition, distance measurement, and deep learning methods.…”

Section: Introductionmentioning

confidence: 99%

Fault Diagnosis Method for Railway Turnout with Pinball Loss-Based Multiclass Support Matrix Machine

Geng,

Xu,

Mei

2023

Applied Sciences

View full text Add to dashboard Cite

The intelligent maintenance of railway equipment plays a pivotal role in advancing the sustainability of transportation and manufacturing. Railway turnouts, being an essential component of railway infrastructure, often encounter various faults, which present operational challenges. Existing fault diagnosis methods for railway turnouts primarily utilize vectorized monitoring data, interpreted either through vector-based models or distance-based measurements. However, these methods exhibit limited interpretability or are heavily reliant on standard curves, which impairs their performance or restricts their generalizability. To address these limitations, a railway turnouts fault diagnosis method with monitoring signal images and support matrix machine is proposed herein. In addition, a pinball loss-based multiclass support matrix machine (PL-MSMM) is designed to address the noise sensitivity limitations of the multiclass support matrix machine (MSMM). First, the time-series monitoring signals in one dimension are transformed into images in two dimensions. Subsequently, the image-based feature matrix is constructed. Then, the PL-MSMM model is trained using the feature matrix to facilitate the fault diagnosis. The proposed method is evaluated using a real-world operational current dataset, achieving a fault identification accuracy rate of 98.67%. This method outperforms the existing method in terms of accuracy, precision, and F1-score, demonstrating its superiority.

show abstract

Dual-student knowledge distillation for visual anomaly detection

Hao,

Huang,

Chen

et al. 2024

Complex Intell. Syst.

View full text Add to dashboard Cite

Anomaly detection poses a significant challenge in the industry and knowledge distillation constructed using a frozen teacher network and a trainable student network is the prevailing approach for detecting suspicious regions. Forward and reverse distillation are the main ways to achieve anomaly detection. To design an effective model and aggregate detection results, we propose a dual-student knowledge distillation (DSKD) based on forward and reverse distillation. Taking advantage of the priority of reverse distillation to obtain high-level representation, we combine a skip connection and an attention module to build a reverse distillation student network that simultaneously focuses on high-level representation and low-level features. DSKD uses a forward distillation network as an auxiliary to allow the student network to preferentially obtain the query image. For different anomaly score maps obtained by the dual-student network, we use synthetic noise enhancement in combination with image segmentation loss to adaptively learn the weight scores of individual maps. Empirical experiments conducted on the MVTec dataset show that the proposed DSKD method achieves good performance on texture images as well as competitive results on object images compared with other state-of-the-art methods. Meanwhile, ablation experiments and a visualization analysis validate the contributions of each of the model’s components.

show abstract

A Convolutional Autoencoder Based Fault Detection Method for Metro Railway Turnout

Cited by 6 publications

References 34 publications

Fault Diagnosis Scheme for Railway Switch Machine Using Multi-Sensor Fusion Tensor Machine

Fault Diagnosis Scheme for Railway Switch Machine Using Multi-Sensor Fusion Tensor Machine

Fault Diagnosis Method for Railway Turnout with Pinball Loss-Based Multiclass Support Matrix Machine

Dual-student knowledge distillation for visual anomaly detection

Contact Info

Product

Resources

About