Detecting anomalies in the Brake Operating Unit (BOU) braking system of metro trains is very important for trains’ reliability and safety. However, current periodic maintenance and inspection cannot detect anomalies at an early stage. In addition, constructing a stable and accurate anomaly detection system is a very challenging task. Hence, in this work, we propose a method for detecting anomalies of BOU on metro vehicles using a one-class long short-term memory (LSTM) autoencoder. First, we extracted brake cylinder (BC) pressure data from the BOU data since one of the anomaly cases of metro trains is that BC pressure relief time is delayed by 4 s. After that, extracted BC pressure data is split into subsequences which are fed into our proposed one-class LSTM autoencoder which consists of two LSTM blocks (encoder and decoder). The one-class LSTM autoencoder is trained using training data which only consists of normal subsequences. To detect anomalies from test data that contain abnormal subsequences, the mean absolute error (MAE) for each subsequence is calculated. When the error is larger than a predefined threshold which was set to the maximum value of MAE in the training (normal) dataset, we can declare that example an anomaly. We conducted the experiments with the BOU data of metro trains in Korea. Experimental results show that our proposed method can detect anomalies of the BOU data well.

Light maintenance training for electric multiple-unit components of the organization of railway operations is generally conducted using maintenance manuals and work videos, following the guidelines of each organization. These manuals are in the form of booklets, complicated and inconvenient for maintenance operators to carry. Therefore, training content that visualizes maintenance procedures in a three-dimensions (3D) space is necessary to overcome the drawbacks of booklet-type training. In this study, we developed augmented reality (AR)-based training content for railway vehicle maintenance to increase training efficiency. Providing warning signs for risky procedures reduces human error, and transparency control makes trainees check the product hierarchy. A virtual experience based on the maintenance manual is provided to improve maintenance proficiency. An axle-mounted disc brake system maintenance manual is implemented in AR to reflect the requirements of maintenance operators. The convenience of this tool is improved by loading the AR content on a mobile device. Two methods of verification were used: the system usability scale (SUS) survey and training efficiency evaluation. The resulting SUS grade was B (excellent), and the training efficiency improved by 34%.

The stiffness in various directions of air springs have a great influence on the stability of high-speed trains. Due to the change in the bogie structure, the required functions of the air springs have also diversified, and the damping of various loading modes such as vertical, lateral, and torsional movements is emerging as important. The stiffnesses are affected by the shape of the air bag, the material, and the ply angles. In this paper, the relationship between the ply angle and the radial position is proposed by equations from deformation modes. The variable angles in the plies were compared and demonstrated in air-spring analysis using the rebar elements. The vertical, lateral, and torsional stiffness of air springs without auxiliary springs were compared between constant ply angles and variable ply angles with respect to the positions. When mainly dealing with the vertical stiffness, the effect of the angle variation was small; however, it was found that it had a large effect of more than 24% and 30% on the stiffness in the lateral direction and torsional direction, respectively.

The Brake operating unit (BOU) of a railway vehicle is one of the important systems for controlling the braking of the train. Because this system uses compressed air, it is difficult to understand and train the system. The existing education method involves learning the pneumatic flow of various control air in a 2D pneumatic circuit diagram based on a maintenance manual. However, in the actual braking system, it was difficult to learn effectively because the air flows in 3D. In order to solve these problems, the improvement of the training technique using the new 3D augmented reality (AR) was performed. In this study, to increase the learning effect of air brake flow, a technique for simultaneously displaying the pneumatic flow in 2D circuit diagram and 3D model was proposed. First, the distance ratio for simultaneous display can be determined using the proposed streamline matching variable calculation algorithm (SLMVC) that uses position and animation duration as input variables. Second, to avoid the complexity of using the 24 variables of the Particle System module in Unity, an existing universal 3D platform, a continuously emission property correction algorithm (CEPC) that can output particle objects as a streamline using only 4 properties (e.g., start lifetime, start speed, emission rate over time, start delay). As a result, the following 6 different types of BOU air pressure could be simultaneously displayed in 2D and 3D (e.g., AC, BC, SR, SBR, AS1, AS2). Therefore, maintenance staff can effectively learn complex pneumatic flow. To verify the usability of the developed content, a survey using the NASA-TLX technique was conducted targeting 60 maintenance staff. As a result of the comparison between Group A using the existing maintenance manual and Group B using the developed AR content, the perceived workload decreased by 28%. In particular, the frustration part decreased by 64% and the performance part decreased by 62%, indicating that the usability of AR content was very good.

The air spring for railway vehicles uses the air pressure inside the bellows to absorb vibration and shock to improve ride comfort and adjust the height of the underframe with a leveling valve to control stable driving of the train. This study developed augmented reality content that proposes a novel visual technology to effectively support the training of air spring maintenance tasks. In this study, a special effect algorithm that displays the dispersion and diffusion of fluid, and an algorithm that allows objects to be rotated at various angles, were proposed to increase the visual learning effect of fluid flow for maintenance. The FDG algorithm can increase the training effect by visualizing the leakage of air at a specific location when the air spring is damaged. In addition, the OAR algorithm allows an axisymmetric model, which is difficult to rotate by gestures, to be rotated at various angles, using a touch cube. Using these algorithms, maintenance personnel can effectively learn complex maintenance tasks. The UMUX and CSUQ surveys were conducted with 40 railway maintenance workers to evaluate the effectiveness of the developed educational content. The results showed that the UMUX, across 4 items, averaged as score of 81.56. Likewise, the CSUQ survey score, consisting of 19 questions in 4 categories, was very high, at 80.83. These results show that this AR content is usable for air spring maintenance and field training support.