The Railway Technical Research Institute has conducted various types of research and development aimed at ensuring safety in the event of a train collision. These studies encompass two types of collision, namely, primary collisions and secondary collisions. Primary collisions refer to the impact between an obstacle and a carbody. Secondary collisions occur between interior fixtures and passengers (and drivers). For Primary collisions, an analytical method was developed to evaluate deformation behavior and amount energy absorbed by the carbody, following several fatal accidents. In the case of secondary collisions, a quantitative analysis was made of injuries to the driver, which was used to draft a set of guiding principles for carbody-structure design, geared to improve crashworthiness. Finally, this paper presents a method devised to accurately evaluate passenger injury using a FE model.
Ballasted track is widely as a main track structure and requires regular maintenance because track geometry deteriorates under the influence of dynamic loads. Therefore, several pieces of research into prediction and prevention of ballasted track deterioration, and maintenance of deteriorated ballasted track have been conducted by RTRI or other organizations around the world. This paper introduces some of the results obtained from the recent research into ballasted track carried out by RTRI and suggests possible future developments in this field.
Human Science in the railway field is a scientific discipline aimed at solving problems and proposing the countermeasures which will improve the safety, ease of use and comfort of railways. This is achieved through insights acquired by measuring, evaluating and analyzing the psychological, physical, physiological and behavioral characteristics of people, such as passengers using railways and workers operating them. This paper outlines recent topics examined in the area of human science research in education and training, and improvements to the on-board and station environment, before finally outlining possible future prospects.
A rapid prototyping bogie (RPB) has multiple actuators to emulate missing bogie components. In the original RPB hybrid control system, deterioration in the control performance, which was caused by dynamic coupling between the multiple actuators, was observed during the RPB performance tests. To solve this issue, a new controller was developed, based upon the dynamically substructured systems (DSS) method, and trialled on an existing proof-of-concept test rig. Through random excitation tests, it was confirmed that the actuators were well controlled. As a consequence, the DSS approach was determined as a viable framework for future research into the RPB system.
Generally braking performance of train sets is evaluated by stopping distance or deceleration. However, stopping distance and deceleration are performance indices established for a whole train set, and do not represent performance of each car individually. For more detailed analysis of detail braking performance, it is useful to have the braking force in each car. This paper describes a method designed to estimate the braking force in each car based on the acceleration and coupler force, by using coupling devices as force sensors. Running tests results on an 18-car freight train using the devised method showed that it was possible to estimate the braking force of each car and that the estimated value was close to the theoretical value.
Traction machines are essential parts for a train to run. Therefore, a condition monitoring system (CMS) is being developed, that detects machine failure in the early stages to prevent traffic disruption. The CMS observes the vibrations of a machine and detects abnormal vibrations with a machine learning algorithm. In the CMS, octave-band analysis is performed to extract feature vectors from vibration data. Running tests were conducted to verify the performance of the CMS. Test results showed that simulated abnormal vibrations were clearly distinguishable from normal ones with the CMS.
In Japan, four welding processes have been generally adopted for producing continuous welded rail (CWR): flash welding (FW), gas pressure welding (GPW), enclosed arc welding (EAW), and thermit welding (TW). Thermit welding method is the most popular rail welding process in Japan. The reliability of the thermit welds is high, because the failure rate has only been about 0.004 % over the last 10 years in Japan. However, recently there have been many cases where thermit welds are judged to be defective due to surface defects. Surface defects are often observed when welding new and worn rails. Welding test results show that surface defects are caused by gas generated from the contact of luting sand with molten steel which enters gaps between the welding mold and rail surface.
Rail corrugation not only generates noise and vibrations but also causes track irregularity and deterioration of track material. It is therefore necessary to manage rail corrugation appropriately. Rail surface roughness due to rail corrugation is managed either through visual inspection or with a simple device. As such, this paper describes the development of a portable "Trolley for the Continuous Measurement of Rail Surface Roughness," designed to gather insight into the mechanisms leading to rail corrugation and developing measures to manage it efficiently. This trolley adopts the asymmetrical chord offset method for measuring the rail surface roughness continuously. A software was also built for controlling this trolley and processing the measured data. This paper presents the details of this development and describes some cases where the trolley could be used.
In order to evaluate the axial force of a continuous welded rail quantitatively in a simple manner, a method was developed to measure axial force based on the natural frequency. However, the accuracy of this method is insufficient because of certain variations in track condition. This study extracted factors which influence measurement accuracy and proposed an error correction method using track finite-element analysis for the purpose of improving the accuracy of this measuring method. Furthermore, measurements were taken of the natural frequency and axial force of a rail on a real track in order to validate the proposed method.
Preventing accidents and incidents depends on gaining a maximum amount of information about the situations in which these events occurred and about what actions were taken by the people involved in these events. Therefore, we developed a hearing investigation technique. And as a result of having tried this technique, we confirmed that we could collect more the data about the background factors than before. Moreover, we developed an educational program that consists of a five-step of exercise and explanation. And we confirmed that the participant's investigative attitude becomes proper.
It is necessary for train dispatchers to share risk-related information with their team members or superiors in order to deal with abnormal train traffic control situations. A training method was therefore developed to give train dispatchers communication skills. The training included risk scenarios, points to remember in terms of communication skills and a training program. A list of 45 key communication points to remember was also developed, which was designed to promote understanding about communication skills through regular use. This training method can easily be introduced into actual train traffic control fields. Trial experiments were conducted with actual train dispatchers which confirmed the effectiveness of the method.
Dummy models for impact simulation have only been able to assess impact from one direction. Some models are able to evaluate frontal impact, while others can evaluate side impact or rear impact. When evaluating collision safety, separate dummy models are employed depending on whether the dummy models are parallel to the direction of travel or perpendicular. In addition, it is difficult to evaluate the interior fixtures of Japanese railway vehicles because the physiques of dummy models are based on Westerners. Therefore, a human model with a Japanese physique was developed to evaluate various situations in railway vehicles.
The analysis of the whole model of train-induced vibration, which consists of the moving train, the track, the supporting infrastructure, the ground, and the building, is currently too large to solve. We thus proposed a numerical simulation method by combining two separate dynamic analysis models. One is an analysis model of the dynamic interaction between the moving train and the track-structure system for calculating excitation force. The other is a three dimensional dynamic analysis model of the supporting structure, the ground, and the building for calculating the propagation of vibration.