Sei Takahashi

Genetic algorithm approach for adaptive offset optimization for the fluctuation of traffic flow

This paper describes offset optimization for the fluctuations of traffic flow using a genetic algorithm (GA). An offset, which is the target of signal control parameters for this study, is difficult to optimize because of its variety of combinations. Traffic signal optimization using GAs has has been investigated in previous studies, most of which focused on signal control without considering the fluctuations of traffic flow. In a practical situation, the rate of flow changes as time passes, so that offset optimization considering these fluctuations of flow is required. As a case study, an urban traffic route in a city of the Chubu region in Japan, with twenty-one signalized intersections, was tested. To perform offset-optimization by a GA, offset values were represented in a chromosome having the same number of genes as the signals. Two different schemes are introduced into the GA-based program and evaluated in terms of average travel time. The results show that the offset optimization schemes used in this study are valuable for efficient signal control.

Distributed Interlocking System and Its Safety Verification

Computerized interlocking systems have been studied for many years. Available interlocking systems mainly are centralized control systems. The great progress of distributed control technology and intelligent terminals make it possible developing distribute interlocking system. In this research, a distributed interlocking system and its control logic are put forward. As the most important property of interlocking system, system logic safety has been verified by using Petri net. The result shows that the new interlocking system is safe and its logic is correct

Toward developing a Decentralized Railway Signalling System Using Petri Nets

Railway interlocking systems ensure the safe operation of trains in stations by controlling sets of devices and equipment. Component-based decentralized railway interlocking system (CBDRIS) is a recently presented railway signalling in which component-based technology is applied. For such a safety-critical real-time system, fail-safe and fault-tolerance have to be ensured. In this paper, a Petri net-based development strategy of CBDRIS is proposed. The development strategy separates the development process into dynamical process (Standardizing hardware as well as control flow of interlocking device components) and static process (Converting current interlocking table of a specific station to a unified format for CBDRIS). In addition, hardware and software fault-tolerance measures adopted in CBDRIS are described respectively.

Safety and Reliability Estimation of Automatic Train Protection and Block System

Considering the high maintenance costs of trackside equipment and its vulnerability to natural disasters, a novel railway system named Automatic Train Protection and Block (ATPB) is proposed by the authors to aid in improving the efficiency and reducing the cost of regional train lines. It is a railway radio system based on onboard equipment. In order to ensure its system safety, the paper gives a formal analysis on the functional requirements specification for the system. Specifically, after analyzing the actual requirements, a UML model of the ATPB system is created first to check the functional completeness and structural reasonability. Second, a formal and unambiguous specification of the ATPB system is established by VDM++, i.e. VDM++ model, based on the UML class diagram. Third, the internal consistency and satisfiability of the formal specification is verified and validated. Finally, a simulation is conducted strictly according to the formal specification. Without any runtime errors, collisions or derailments, the results demonstrate the high quality of the simulation and the safety of the specification. Copyright

Modeling a Distributed Railway Interlocking System with Object-Oriented Petri Nets

This paper describes how great progress of distributed technology and intelligent terminals makes it possible to develop a distributed railway interlocking system (DRIS). This paper presents a modeling method of DRIS by using G-nets, which are Petri nets extended with object-oriented concepts. The modeling method improves maintenance and reusability remarkably. Based on the models, the DRIS can be implemented with an object-oriented language such as C++ or JAVA.

Study on Formal Specification of Automatic Train Protection and Block System for Local Line

This paper presents a formal specification of an Automatic Train Protection and Block (ATPB) model for local line railway system in Japan proposed by the author [12], and validates the model by internal consistency proving and systematic testing. The system consists of two parts, the on-board subsystem and ground subsystem. The former is to detect the basic state of train, such as position, speed and integrity, monitor the speed, communicate with ground equipment and record the relative events. And the latter is responsible for communicating with train, controlling the route and interlocking, and decision-making for train operation adjustment. The main purpose of this project is to improve the efficiency and guarantee that there is no collision, no derailment and no over speeding at the same. The formal language used in this project is VDM++. And the state and specification of operation are all checked and validated using VDMTools. The results confirm the correctness of this system and the model throws new light on practical system design.

Multichannel temporal data classification of motor imagination using fNIRS

We describe classification of functional near-infrared spectroscopy (fNIRS) data acquired during finger tapping imagery tasks performed by a human subject, using an artificial neural network designed for image sequence recognition. Our goal is to develop a brain-computer interface that can handle various intentions of users. We used an fNIRS system to collect neural information from brain activity. For discrimination of the fNIRS data, we used our previously proposed neural network model called the Neocognitron-type Image Sequence Recognition Model (Neo-ISRM), which is suitable for analyzing multichannel temporal patterns. Finger tapping imagery of both left and right hands was used as the mental tasks to be discriminated with Neo-ISRM. The model gave good discrimination results for each category of tasks from data for the motor area, the prefrontal area, and the frontal lobe. In all experiments, we confirmed that the discrimination results for the frontal lobe had fewer discrimination errors compared with the results for both the motor and prefrontal areas.

Onboard Measurement Method for Signaling Equipment on Probe Trains

This paper describes how an automatic train protection system (ATS) and a track circuit is a fundamental and essential element of a railway signaling system. First, the paper mentions maintenance technology for ATS. For the stable operation of an ATS, the management of the quality factor (Q) of the ground beacon is essential and the Q must be kept its value within a specified level through maintenance. The newly developed measurement method is based on the current ratio between two different frequency levels of a pick-up coil current when the train passes the beacon. And it can provide the correct Q data against the distance-fluctuation. Secondly we develop the onboard method of measuring track circuit. The current of track circuit which is received through a pair of onboard signal receivers is continuously recorded and analyzed. From the analyzed data signal engineers judge the necessity of maintenance of track circuits. These measuring technologies are installed onto the probe train which is one of revenue-earning train with measurement functions. A condition based monitoring system is realized through the result of statistic processing of monitored data which is collected from two or more probe trains.

Adaptive Traffic Signal Control For The Fluctuations Of The Flow Using A Genetic Algorithm

This paper describes adaptive traflic signal control for the fluctuations of flow using a genetic algorithm (GA). The target of signal control parameters for this study is offset that is difficult to optimize because of its variety of combinations. Offset optimization using GAs has been investigated in previous studies. Most of them however, focused on signal control in a condition where the traf%c flow was fixed. In a practical scene, the rate of flow changes as time passes, so that offset-optimization considering these fluctuations of flow is required. As a case study, an urban traflic route in a city of the Chubu region in Japan, with twenty-one signalized intersections, was tested. To perform offset-optimization by a GA, offset values were represented in a chromosome having the same number of genes as the signals. Two different conditions of traffic flow, slow monotonous and rapid increase, were chosen for the simulation. The results show that the offset optimization technique used in this study was a valuable one for efficient signal control.

Data-Based Axle Temperature Prediction of High Speed Train by Multiple Regression Analysis

Axle is a key equipment of high speed train, and affects the safety of train operation. The fault of axle is commonly detected by comparing the current axle temperature with a fixed temperature threshold. Owing to the complex mechanism of axle temperature rising, the axle temperature has a wide variation range even working in normal condition. Therefore, the shortcomings of the method with a fixed temperature threshold are obvious: a high threshold may leads to missing alarm, existing potential safety risk, on the contrary, a low one can ensure the safety, but easily leads to failing alarm, causing unnecessary troubleshooting and maintenance. In view of the problem, a model for the calculation of dynamical temperature threshold is proposed in this paper by relational analysis of monitoring data. Specifically, after analyzing the characteristic of axle temperature changing, the temperature prediction process is divided into three stages according to its running modes, i.e. acceleration, stable running and deceleration. Then the temperature prediction model is established and validated, and the results denoted the effectiveness and practicability.