Wheel condition assessment of high-speed trains under various operational conditions using semi-supervised adversarial domain adaptation

Chen, Si‐Xin; Zhou, Lu; Ni, Yiqing

doi:10.1016/j.ymssp.2022.108853

Cited by 11 publications

(3 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another way of dealing with such lack-of-data problems is transfer learning, which attempts to solve these problems by exploiting knowledge about models trained on some domain by transferring it to another domain. This knowledge is usually in the form of some feature extractor [29], or by seeking a domain, onto which data from different tasks can be mapped and a common model can be used [19,30].…”

Section: Model-agnostic Meta-learningmentioning

confidence: 99%

Towards a population-informed approach to the definition of data-driven models for structural dynamics

Tsialiamanis

Dervilis

Wagg

et al. 2023

Mechanical Systems and Signal Processing

View full text Add to dashboard Cite

Section: Model-agnostic Meta-learningmentioning

confidence: 99%

Towards a population-informed approach to the definition of data-driven models for structural dynamics

Tsialiamanis

Dervilis

Wagg

et al. 2023

Mechanical Systems and Signal Processing

View full text Add to dashboard Cite

“…5 In the recent years, deep learning algorithms that are powered by big data have found their application in the field of brake and railway status monitoring. [6][7][8] These algorithms have the capability to handle the nonlinear and chaotic signals generated by braking systems and railway, enabling the learning of underlying status features for effective monitoring. Especially in the field of brake systems, Jegadeeshwaran et al 9 investigated several hydraulic brake faults and compared the diagnostic effects of different machine learning methods in the braking system.…”

Section: Introductionmentioning

confidence: 99%

Health status monitoring of high-speed train brake pads considering noise under variable working conditions

Kang

Zhang

et al. 2023

Structural Health Monitoring

View full text Add to dashboard Cite

The brake pads of high-speed trains operate under complex and variable conditions, and the collected brake signals are easily affected by noise, making monitoring the health status of brake pads more difficult. A multi-representation adaptation network for online monitoring the health status of high-speed train brake pads, which are affected by noise under variable working conditions, is proposed in this study. First, a parameter-sharing deep residual network is used to extract the friction block features of the source and target domain data. Then, the features are mapped to different low-dimensional feature spaces through the inception adaptation module, and multiple representations are obtained. The network applies conditional maximum mean discrepancy to align representations of the source and target domains, thus learning multiple domain-invariant representations. Hence, the network acquires more knowledge of the friction block status and attenuates the interference of noise signals on the status monitoring. The friction block vibration data were collected from various brake disc speeds, and variable working condition-transfer experiments under the influence of noise were performed on the brake friction and bearing datasets. The results show that the proposed network outperforms other transfer methods, which can better extract and identify the status features of the friction block under the noise interference.

show abstract

“…Qin et al developed a stepwise adaptive CNN to classify the faults of a high-speed train bogie with a continuously varying vehicle speed [26]. Chen et al established a semi-supervised adversarial DA to assess the condition of high-speed train wheels under different surrounding environments [27].…”

Section: Introductionmentioning

confidence: 99%

Unsupervised Discrepancy-Based Domain Adaptation Network to Detect Rail Joint Condition

Jiang,

Wang,

et al. 2023

IEEE Trans. Instrum. Meas.

Self Cite

View full text Add to dashboard Cite

Damage to maglev rail joints, which connect adjacent rail segments, threatens the safety and comfort of railway systems. Machine learning methods have been used in combination with online monitoring data to assess the health conditions of maglev rail joints. However, most of the existing methods rely on the data collected in controlled scenarios, such as those involving constant train operation speeds. Given the diversity of operational conditions, a model learned from one known case (source domain) cannot be directly applied to the case of interest (target domain). Therefore, this paper proposes a domain adaptation approach to diagnose the health conditions of maglev rail joints in complex operational conditions. The domain adaptation is unsupervised because the source and target domains are characterized by labeled and unlabeled samples, respectively. Domain adaptation is implemented by integrating the sample moments with different orders into the transfer loss of a neural network. By minimizing the transfer loss, the domain shift caused by the difference in the operational conditions can be reduced, and the knowledge of features learned from the neural network is transferred from the source domain to the target domain. The proposed approach is validated over a dataset of time-frequency spectrograms derived from the experimental acceleration data of maglev rail joints in two operation modes: stable passing and braking. The proposed approach can successfully identify the conditions of the maglev rail joints, i.e., bolt-looseness-caused rail step, misalignment-caused lateral dislocation, and normal condition, even when the operation mode of the maglev train changes.

show abstract

Wheel condition assessment of high-speed trains under various operational conditions using semi-supervised adversarial domain adaptation

Cited by 11 publications

References 41 publications

Towards a population-informed approach to the definition of data-driven models for structural dynamics

Towards a population-informed approach to the definition of data-driven models for structural dynamics

Health status monitoring of high-speed train brake pads considering noise under variable working conditions

Unsupervised Discrepancy-Based Domain Adaptation Network to Detect Rail Joint Condition

Contact Info

Product

Resources

About