A fault diagnosis of a train door system is carried out using the motor current signal that operates the door. A test rig is prepared, in which various fault modes are examined by applying extreme conditions, as well as the natural and artificial wears of critical components. Two approaches are undertaken toward the fault classification for comparative purposes: one is the traditional feature-based method that requires several steps for the processing features such as signal segmentation, the extraction of time-domain features, selection by Fisher’s discrimination, and K-nearest neighbor. The other is the deep learning approach by employing the convolutional neural network (CNN) to skip the hand-crafted features extraction process. In the traditional approach, good accuracy is found only after the current signal is segmented into the three velocity regimes, which enhances the discrimination capability. In the CNN, superior accuracy is obtained even by the original raw signal, which is more convenient in terms of implementation. However, in view of practical applications, the traditional approach is more useful in that the features processing can be easily applied to assess the health state of each fault and monitor the progression over time in the real operation, which is not enabled by the deep learning approach.
The reliability and safety of the train system is a critical issue, as it transports many passengers in its daily operation. Most studies focus on fault diagnosis methods to determine the cause of faults in the train system. Aside from fault diagnosis, it is also vital to perceive a fault even before it occurs. In this study, a fault occurrence prediction based on a machine learning model is developed. The fault occurrence prediction method aims to predict the remaining useful life (RUL) of a train subsystem. RUL refers to the remaining amount of time before a fault occurs on a train subsystem. The prediction method developed in this study can be used to clear a fault even before it occurs. In case of inevitable faults, the output from the prediction method can be used to alert the personnel in charge by imposing an alarm. Therefore, the fault occurrence prediction method is expected to increase the reliability of the train system. The deep neural-network-based model is tested on an actual device. Deep neural network is used because of its feature extraction capability, especially in handling big amount of data. The testing results in 90.08% accuracy. In addition, a graphical user interface is developed as an interface between a user and the actual device containing the fault occurrence prediction model.
As railway is considered one of the most significant transports, sudden malfunction of train components or delayed maintenance may considerably disrupt societal activities. To prevent this issue, various railway maintenance frameworks, from “periodic time-based and distance-based traditional maintenance frameworks” to “monitoring/conditional-based maintenance systems,” have been proposed and developed. However, these maintenance frameworks depend on the current status and situations of trains and cars. To overcome these issues, several predictive frameworks have been proposed. This study proposes a new and effective remaining useful life (RUL) estimation framework using big data from a train control and monitoring system (TCMS). TCMS data is classified into two types: operation data and alarm data. Alarm or RUL information is extracted from the alarm data. Subsequently, a deep learning model achieves the mapping relationship between operation data and the extracted RUL. However, a number of TCMS data have missing values due to malfunction of embedded sensors and/or low life of monitoring modules. This issue is addressed in the proposed generative adversarial network (GAN) framework. Both deep neural network (DNN) models for a generator and a predictor estimate missing values and predict train fault, simultaneously. To prove the effectiveness of the proposed GAN-based predictive maintenance framework, TCMS data-based case studies and comparisons with other methods were carried out.
To cope with the increased demand on the intra-city transportation by urban rails, the introduction of the light rail transit (LRT) systems has been expedited in Korea due to the possible reduction of both the development and operation costs from adopting LRT systems. The LRT systems have so far been designed, constructed and operated based on the corresponding law and regulations. It has been conceived that fully complying with the existing guidelines may incur some extra costs on LRT. In addition, the present design of LRT stations seems to require unnecessarily long flow of passengers traffic, particularly for disabled people. In this paper, as an approach to solving the aforementioned issues, an introduction of 'LRT stops' has been studied where the stops are similar in concept to bus stops and are intended to replace the stations of a bigger scale in general. Specifically, necessary guidelines for design have been developed by modifying the existing ones to be fit with LRT stops. A design case was also presented to evaluate them. The effective use of the results reported here will provide an opportunity of cost reduction in connection with the construction and operation, and also let people benefit from convenient use of rails, thereby resulting in enhanced transportation welfare.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.