Fault Diagnosis of High-speed Train Bogie Based on Deep Neural Network

Zhang, Yuanjie; Qin, Na; Huang, Deqing; Liang, Kaiwei

doi:10.1016/j.ifacol.2019.12.395

Cited by 9 publications

(7 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As an essential part of TCN, the communication transmission medium occupies a large proportion of the TCN system. Transmission medium fault mainly includes permanent fault and minor fault [12]. Permanent fault: Permanent fault mainly includes open circuit and short circuit faults of transmission medium.…”

Section: Physical Layer Fault Typementioning

confidence: 99%

“…[11], Tian proposed to use the Fluke Testing Instrument to test several electrical parameters of the train bus cable like MVB (Multifunction Vehicle Bus) and WTB (Wire Train Bus), such as signal attenuation, near-end crosstalk, and return loss, to judge whether the cable is faulty or not. To some extent, Zhang [12] proposed a diagnosis method based on a fuzzy neural network that solves the problem of low fault diagnosis accuracy caused by randomness and uncertainty of the TCN. Li [13] proposed a view to improving the kernel function of the support vector machine (SVM) to obtain the physical waveform parameters of TCN and established a fault diagnosis model based on MKSVM.…”

Section: Introductionmentioning

confidence: 99%

“…Zhou et al [20] applied TDR technology to realize fault detection of twisted pairs. Zhang [12] studied the diagnosis in field bus network fault of CRH5 train, simulated the fault of TCN cable, used a cable tester to collect data, and realized the fault diagnosis of the physical layer cable using the residual neural network method.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Novel Fault Diagnosis Method for Train Real‐Time Ethernet Network Based on Physical Layer Electrical Signal Features

Zhu,

Nie,

et al. 2023

IEEJ Transactions Elec Engng

View full text Add to dashboard Cite

Train real‐time ethernet network (TREN) is responsible for the transmission of train control commands and equipment status information. Due to the harsh and changeable climatic environments, complex electromagnetic environments, and long service, the TREN has a high incidence of physical layer failures, which affects train operation safety. This paper presents a novel fault diagnosis method in which waveform electrical signal features are extracted from the physical layer signal of TREN. Random forest is used for identifying typical failure modes of actual TREN, and the model parameters are optimized to improve diagnostic performance. Finally, the physical layer signals of actual TREN are collected. The results show that the performance of the proposed fault diagnosis method is better than other advanced methods. © 2023 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.

show abstract

Section: Physical Layer Fault Typementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Novel Fault Diagnosis Method for Train Real‐Time Ethernet Network Based on Physical Layer Electrical Signal Features

Zhu,

Nie,

et al. 2023

IEEJ Transactions Elec Engng

View full text Add to dashboard Cite

show abstract

“…Moreover, to further evaluate the performance of the developed research, the CNN-LSTM algorithm was compared with other hybrid deep learning architectures, which are CNN-CNN [9], LSTM-LSTM [10] and DNN-DNN [42]. To better compare Fig.…”

Section: Performance Evaluation On the Hurst Exponent And The Cnn-lstm Algorithmmentioning

confidence: 99%

Prediction of the remaining useful life of cutting tool using the Hurst exponent and CNN-LSTM

Zhang

et al. 2021

Int J Adv Manuf Technol

View full text Add to dashboard Cite

To enhance production quality, productivity and energy consumption, it is paramount to predict the remaining useful life (RUL) of a cutting tool accurately and efficiently. Deep learning algorithm-driven approaches have been actively explored in the research field though there are still potential areas to further enhance the performance of the approaches. In this research, to improve accuracy and expedite computational efficiency for predicting the RUL of cutting tools, a novel systematic methodology is designed to integrate strategies of signal partition and deep learning for effectively processing and analysing multi-sourced sensor signals collected throughout the lifecycle of a cutting tool. In more detail, the methodology consists of two sub-systems: (i) a Hurst exponentbased method is developed to effectively partition complex and multi-sourced signals along the tool wear evolution, and (ii) a hybrid CNN-LSTM algorithm is designed to combine feature extraction, fusion and regression in a systematic means to facilitate the prediction based on segmented signals. The system was validated using a case study with a large set of databases with multiple cutting tools and multi-sourced signals. Comprehensive comparisons between the proposed methodology and some other mainstream algorithms, such as CNN, LSTM, DNN and PCA, were carried out under the conditions of partitioned and unpartitioned signals. Benchmarks showed that, based on the case study in this research, the prediction accuracy of the proposed methodology reached 87.3%, which is significantly better than those of the comparative algorithms.

show abstract

“…High-speed trains are a type of equipment utilizing advanced technology and complex structures. Safety is the top priority in high-speed railway operation [1], which can be achieved through train fault diagnosis and prediction [9]. High-speed train fault diagnosis methods can be divided into manual diagnosis, automatic test equipment (ATE) diagnosis and built-in test equipment (BITE) diagnosis.…”

Section: Related Workmentioning

confidence: 99%

A Fault Diagnosis and Visualization Method for High-Speed Train Based on Edge and Cloud Collaboration

et al. 2021

View full text Add to dashboard Cite

Safety is the most important aspect of railway transportation. To ensure the safety of high-speed trains, various train components are equipped with sensor devices for real-time monitoring. Sensor monitoring data can be used for fast intelligent diagnosis and accurate positioning of train faults. However, existing train fault diagnosis technology based on cloud computing has disadvantages of long processing times and high consumption of computing resources, which conflict with the real-time response requirements of fault diagnosis. Aiming at the problems of train fault diagnosis in the cloud environment, this paper proposes a train fault diagnosis model based on edge and cloud collaboration. The model first utilizes a SAES-DNN (stacked auto-encoders deep neural network) fault recognition method, which can integrate automatic feature extraction and type recognition and complete fault classification over deep hidden features in high-dimensional data, so as to quickly locate faults. Next, to adapt to the characteristics of edge computing, the model applies a SAES-DNN model trained in the cloud and deployed in the edge via the transfer learning strategy and carries out real-time fault diagnosis on the vehicle sensor monitoring data. Using a motor fault as an example, when compared with a similar intelligent learning model, the proposed intelligent fault diagnosis model can greatly improve diagnosis accuracy and significantly reduce training time. Through the transfer learning approach, adaptability of the fault diagnosis algorithm for personalized applications and real-time performance of the fault diagnosis is enhanced. This paper also proposes a visual analysis method of train fault data based on knowledge graphs, which can effectively analyze fault causes and fault correlation.

show abstract

Fault Diagnosis of High-speed Train Bogie Based on Deep Neural Network

Cited by 9 publications

References 5 publications

A Novel Fault Diagnosis Method for Train Real‐Time Ethernet Network Based on Physical Layer Electrical Signal Features

A Novel Fault Diagnosis Method for Train Real‐Time Ethernet Network Based on Physical Layer Electrical Signal Features

Prediction of the remaining useful life of cutting tool using the Hurst exponent and CNN-LSTM

A Fault Diagnosis and Visualization Method for High-Speed Train Based on Edge and Cloud Collaboration

Contact Info

Product

Resources

About