A feature engineering framework for online fault diagnosis of freight train air brakes

Wang, Qihang; Gao, Tianci; Tang, Haichuan; Wang, Yifeng; Chen, Zhengxing; Wang, Jianhui; Wang, Ping; He, Qing

doi:10.1016/j.measurement.2021.109672

Cited by 17 publications

(15 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Locomotive vehicles are equipped with a CAS to actuate system components, such as the braking system and the pneumatically operated doors [36]. In this case, both Wang et al [37] and Liu et al [38] carried out an investigation for anomaly detection of the air brakes subsystem. Similarly, Sun et al [39] implemented an anomaly detection technique based on intelligent approaches; however, this research focused on actual air leakage detection for the pneumatically actuated train doors.…”

Section: Rq1-area Of Applicationmentioning

confidence: 99%

“…Experimental Setup Approach in a Lab-Based Environment Experimental lab-based case studies comprised 49% of the papers. Two of the locomotive publications involved this approach, such as Wang et al [37] collected data from a three-cab in-lab experimental platform, while Sun et al [39] generated a dataset from a jig that replicated the actual pneumatic train door subsystem. Wang et al's [37] work involved the use of 15 pressure sensors strategically placed within the CA brake system.…”

mentioning

confidence: 99%

“…Two of the locomotive publications involved this approach, such as Wang et al [37] collected data from a three-cab in-lab experimental platform, while Sun et al [39] generated a dataset from a jig that replicated the actual pneumatic train door subsystem. Wang et al's [37] work involved the use of 15 pressure sensors strategically placed within the CA brake system. Thus, the experimental platform was capable of simulating different operational conditions while inducing typical brake faults, such as brake pipe leakage [37].…”

mentioning

confidence: 99%

See 2 more Smart Citations

Intelligent Approaches for Anomaly Detection in Compressed Air Systems: A Systematic Review

et al. 2023

View full text Add to dashboard Cite

Inefficiencies within compressed air systems (CASs) call for the integration of Industry 4.0 technologies for financially viable and sustainable operations. A systematic literature review of intelligent approaches within CASs was carried out, in which the research methodology was based on the PRISMA guidelines. The search was carried out on 1 November 2022 within two databases: Scopus and Web of Science. The research methodology resulted in 37 papers eligible for a qualitative and bibliometric analysis based on a set of research questions. These aimed to identify specific characteristics of the selected publications. Thus, the review performed a comprehensive analysis on mathematical approaches, multiple machine learning (ML) methods, the implementation of neural networks (NNs), the development of time-series techniques, comparative analysis, and hybrid techniques. This systematic literature review allowed the comparison of these approaches, while widening the perspective on how such methods can be implemented within CASs for a more intelligent approach. Any limitations or challenges faced were mitigated through an unbiased procedure of involving multiple databases, search terms, and researchers. Therefore, this systematic review resulted in discussions and implications for the definition of future implementations of intelligent approaches that could result in sustainable CASs.

show abstract

Section: Rq1-area Of Applicationmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Intelligent Approaches for Anomaly Detection in Compressed Air Systems: A Systematic Review

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Some approaches mainly Analyzing Figure 17a, it is possible to carry out some considerations regarding possible applications of a predictive maintenance approach [1]. Some approaches mainly based on machine learning [34,35] are already available in the literature, however the innovative solution proposed in this paper was based on the smart measurement of pressure values in some peculiar points of the systems to identify possible faults in the main components. This way it could be possible to run simple and efficient algorithm directly on the on-board gateway in real time to perform these diagnostic activities.…”

Section: Diagnostic Analysismentioning

confidence: 99%

Energy Autonomous Wireless Sensor Nodes for Freight Train Braking Systems Monitoring

Zanelli

Mauri

Dezza

et al. 2022

Sensors

View full text Add to dashboard Cite

Nowadays, railway freight transportation is becoming more and more crucial since it represents the best alternative to road transport in terms of sustainability, pollution, and impact on the environment and on public health. Upgrading the potentiality of this kind of transportation, it would be possible to avoid delays in goods deliveries due to road accidents, traffic jams, and other situation occurring on roads. A key factor in this framework is therefore represented by monitoring and maintenance of the train components. Implementing a real time monitoring of the main components and a predictive maintenance approach, it would be possible to avoid unexpected breakdowns and consequently unavailability of wagons for unscheduled repair activities. As highlighted in recent statistical analysis, one of the elements more critical in case of failure is represented by the brake system. In this view, a real time monitoring of pressure values in some specific points of the system would provide significant information on its health status. In addition, since the braking actions are related to the load present on the convoy, thanks to this kind of monitoring, it would be possible to appreciate the different behavior of the system in case of loaded and unloaded trains. This paper presented an innovative wireless monitoring system to perform brake system diagnostics. A low-power system architecture, in terms of energy harvesting and wireless communication, was developed due to the difficulty in applying a wired monitoring system to a freight convoy. The developed system allows acquiring brake pressure data in critical points in order to verify the correct behavior of the brake system. Experimental results collected during a five-month field test were provided to validate the approach.

show abstract

“…In recent years, with the rapid increase in coverage and lines, fault diagnosis and security maintenance for railways gradually become one of the top concerns in China [ 1 , 2 , 3 ]. The intelligence fault diagnosis of key railway equipment, such as turnout [ 4 ], rolling bearings [ 5 ], gearboxes [ 6 ], and brakes [ 7 ], etc., has been achieving wide application due to the rapid development of artificial intelligence techniques, sensor technology, and computer technology. The railway turnout system (RTS) is one of the significant structures of the railway infrastructure, which is essential to control the running direction of the train and ensure running safety.…”

Section: Introductionmentioning

confidence: 99%

A Fault-Diagnosis Method for Railway Turnout Systems Based on Improved Autoencoder and Data Augmentation

Hei

et al. 2022

Sensors

View full text Add to dashboard Cite

In recent years, with the rapid increase in coverage and lines, security maintenance has become one of the top concerns with regard to railway transportation in China. As the key transportation infrastructure, the railway turnout system (RTS) plays a vital role in transportation, which will cause incalculable losses when accidents occur. The traditional fault-diagnosis and maintenance methods of the RTS are no longer applicable to the growing amount of data, so intelligent fault diagnosis has become a research hotspot. However, the key challenge of RTS intelligent fault diagnosis is to effectively extract the deep features in the signal and accurately identify failure modes in the face of unbalanced datasets. To solve the above two problems, this paper focuses on unbalanced data and proposes a fault-diagnosis method based on an improved autoencoder and data augmentation, which realizes deep feature extraction and fault identification of unbalanced data. An improved autoencoder is proposed to smooth the noise and extract the deep features to overcome the noise fluctuation caused by the physical characteristics of the data. Then, synthetic minority oversampling technology (SMOTE) is utilized to effectively expand the fault types and solve the problem of unbalanced datasets. Furthermore, the health state is identified by the Softmax regression model that is trained with the balanced characteristics data, which improves the diagnosis precision and generalization ability. Finally, different experiments are conducted on a real dataset based on a railway station in China, and the average diagnostic accuracy reaches 99.13% superior to other methods, which indicates the effectiveness and feasibility of the proposed method.

show abstract

A feature engineering framework for online fault diagnosis of freight train air brakes

Cited by 17 publications

References 25 publications

Intelligent Approaches for Anomaly Detection in Compressed Air Systems: A Systematic Review

Intelligent Approaches for Anomaly Detection in Compressed Air Systems: A Systematic Review

Energy Autonomous Wireless Sensor Nodes for Freight Train Braking Systems Monitoring

A Fault-Diagnosis Method for Railway Turnout Systems Based on Improved Autoencoder and Data Augmentation

Contact Info

Product

Resources

About