We propose a new track circuit method using solitary waves to regenerate track circuits. With the new track circuit method, trains without cable and with low power can be located instantly, and five-aspect signals can be controled. Furthermore, this method is expected to improve the preventive maintenance of track circuits and to help realize inexpensive digital ATC (Automatic Train Control).

An obstruction warning signal is a type of wayside signal that is sent only in case of an emergency at level crossings. In this signal, a bright red flash is emitted, which notifies the driver via visual confirmation. However, the warning system primarily depends on the attention of the driver, and this leads to a risk of oversights and accidents. In this paper, we propose a blink detection method for obstruction warning signals using an in-vehicle camera to improve safety of train operation.

We use bit shift function in our method, which enables us to detect blinking within a second. An evaluation test was conducted under static and dynamic testing conditions, which confirmed that the proposed method can detect blinks as far as 600 meters with an accuracy of approximately 90 percent. In addition, zero false detection was confirmed on the data recorded form operating commercial lines.

Concrete cover depth largely depends on construction accuracy and is particularly affected by spacer installation accuracy. In areas where the cover depth is small, concrete spalling can occur locally due to corrosion of the reinforcing steel. Therefore, such areas should be identified in advance. In this study, the authors investigate a measuring method that can detect insufficient local coverage of concrete for middle slabs of RC rigid frame viaducts. Before examining the measurement method, the authors created a simulation, considering the variation of the concrete cover of the existing structure, to obtain the distribution of the variation of the concrete cover. Using this distribution, the authors compare the measurement accuracies of multiple cover measurement methods. The results show that it is generally possible to identify local lack of concrete cover, by conducting measurement with a rebar probe continuously in the line direction, where the distance between adjacent measuring lines is 1m or less.

When the rail head repair method is used to fix damaged rails, hot correcting work is required to prevent the repaired parts from dipping due to thermal contraction. The conventional straightening machine is applicable only to ballast tracks because the machine has a structure which lifts the bottom of the rail. Therefore, in this study, we developed new straightening machine that can also be applied to solid-bed tracks. Furthermore, we demonstrated the practicality of our proposed method using a field test.

The 2016 Kumamoto Earthquake caused the breakage of wiring fittings including pole bands that anchored a spring-type tensioning devices. To reduce the breakage risk of the pole bands, we proposed a pin-connected yoke as an alternative to the conventional triangle-shaped rigid yoke, which can reduce lateral load transferred from the tensioning device. A prototype of the pin-connected yoke demonstrated 20% to 50% load reduction compared to the conventional one in a static loading test.

This paper presents a method for measuring the contact positions between wheel and rail using strain gauges. Finite element method was conducted, and the results indicated that a strain near a hole in the wheel plate was small for a lateral force and changed linearly with respect to changes in the wheel load position. Therefore, strain gauges were placed near the hole, the full-scale wheel set was rotated on a roller rig, and the measured wheel/rail contact positions were compared with the real contact positions. As both sets of positons were almost the same, it was concluded that the proposed method could be effective for measuring wheel/rail contact positions.

In inertial load drive applications, such as railway traction system and electric vehicles, a feed-forward current control mode is applied, which employs a single pulse mode of inverters in a wide speed range. Tuning both the primary and secondary motor parameters is essential for sophisticated torque control in the feed-forward current control mode. In this study, an auto tuning method focusing on both the primary and secondary motor parameters is developed for more precise torque control in the feed-forward current control region. The proposed method is characterized by the primary angular frequency dependency of the PI compensator voltage. The proposed method is verified via simulation and a small-scale model experiment with a 0.75kW induction motor with an inertial load.

In railway train control, when the track is used as a communication channel for control information, it is necessary to transfer the control information while avoiding the interference of the noise caused by the train drive current flowing through the track. With regard to the characteristics of noise, in this study, we focused on the discrete harmonics, frequency variation of commercial power sources, and switching noise from the onboard inverter. The modulation of the control information adopts orthogonal frequency division multiplexing (OFDM), which employs multiple narrowband carriers simultaneously. Because the noise appears as a line spectrum, even if a communication error occurs in the noise carriers, the error is corrected, thereby increasing the communication speed. Our previous study demonstrated that a simulation using the Read-Solomon code yielded an error correction code (ECC). Recently, codes such as turbo codes and low-density parity-check (LDPC) codes, have been adopted as ECCs; however, thus far, no method has been proposed for the safe use of these codes. In this study, we consider this problem, and demonstrate the concept of eliminating the interference of the train-driven current noise due to the OFDM modulation using ECC when using a track circuit for information transmission in train control. To reduce the probability of miscorrection, the process that determines the number of corrected errors is introduced in the receiver, and the safe use of the ECC is demonstrated.

This paper reports that the core loss of a nanocrystalline motor composed of a nanocrystalline magnetic material increases due to a ringing phenomenon, i.e., a damping oscillation in the motor current waveform. The core losses of a nanocrystalline motor is evaluated, when it is driven by an Si-IGBT and by a GaN-FET inverters. The result is compared with that of an NO motor composed of non-oriented electrical steel sheets, which is also driven by the two types of inverters. The core losses of the NO motor decrease with an increase in the carrier frequency. While the core loss of the nanocrystalline motor driven by an Si-IGBT inverter is almost constant, that driven by a GaN-FET inverter significantly increases at higher carrier frequencies. The increase is caused by a ringing in the output current waveform of GaN-FET inverter.

There is growing demand for energy saving to prevent global warming. Thus, the railway industry is focusing on energy saving. Power consumption by running trains has a large variation of about 20% caused by drivers' operation. Therefore, improving the running profile will considerably reduce energy consumption. Some algorithms for generating energy saving running profile have been researched and developed. To extend the applicability of energy saving running profile to various railway, we have developed an algorithm for generating energy saving running profile which fulfills target passing time on line which has passing stations between departure and arrival station like limited express train, in addition to between only adjacent stations like local line. Conventional algorithms do not optimize the passing speed because it generates running profile without consideration after the passing station. As a result, there is inefficient acceleration on the next section, and the energy saving performance is degraded. To solve this issue, we propose an evaluation function that adds energy consumption and running time on the next section when generating an energy saving running profile. The proposed method was evaluated by a running simulation, and the results show that the energy saving is improved by at most 4.7%, compared with the conventional method.

Wireless power transfer (WPT) systems have receieved considerable interest for promoting the sale of electric vehicles (EVs). In order to increase the use of WPT systems, electromagnetic field (EMF) leakage should be evaluated at neighborhood and faraway points. Generally, these are measured in the shield room. However, systems that use many coils and cores require very large shield rooms, and accurate measurements are difficult to obtain. In this study, we investigate the coil scaling law to simplify EMF evaluation. Satisfied conditions that enable an equivalent evaluation between full-scale coils and mini-scale ones are derived. The effectiveness of this method is demonstrated using by simulations and experiments.