Applied Mechanics and Materials Vols. 97-98

Abstract: This paper puts forward a new design of intelligent navigational simulator due to the fact that the target ships in the current existing navigational simulator at home and aboard which does not have the capacity to automatically give way to other ships to avoid collision. The Designed navigational simulators in which the both target ships and own ships have the ability of anti-collision which makes a big improve on the current navigational training effectiveness. The new designed simulator ,on one hand, can instruct the trainer to take correct action to avoid collision, on the other hand, the target ships in the training exercise can intelligently navigate according to the seaman ordinary practice or《COLREG 1972》at the current traffic situation which is much closer to the real situation on the sea. On the base of the research of algorithm of automatic anti-collision, the raw system frame and basic principle of the intelligent navigational simulator and some typical experiment and analysis is also given.

Abstract: To provide continuous guiding information is one of the basic functions in road guide sign system. The representation model of guiding information based on road network topology is presented first in this paper; the meaning of guiding path in guide sign system is introduced then; the analysis model for finding continuous guiding path and the assessment index for guiding continuity is constructed. As an example, the road guide sign system in Guangzhou Zhujiang New Town is analyzed by the present model. It shows that the analysis model is feasible.

Abstract: The urban road traffic congestion has not only brought many inconvenient for people's routine work and life, but also will restrict the growth of the economical, to accelerate the urban environment worsening and serious influence the city sustainable development. This paper studies based on the dynamic detection of urban road traffic congestion condition recognition technology can fast and accurate discover in the road network which already had the traffic congestion or soon occurs, then estimated the crowded proliferation scope and duration, which are advantageous to carry on the transportation induction and the traffic control promptly. And according to the different target client to the different emphasis point to the distinguish algorithm, has designed the urban road traffic congestion recognition grading warning system.

Abstract: Nowadays, the total volume of passenger trip demand has increased due to population and economic growth. In this situation , government policy encourages people to use public transportation for inter-city trips. In the meantime, buses are the most widely used in transit technology today. The most important issue in buses service is timely arrival. Due to the limited capacity of the streets and increasing car production, we cannot devote a specific lane to bus operation to separate their operation from other traffic. Generally, actual arrival time of buses in comparison to planned arrival time occurs in three scenarios : sooner, on-time, and later. This article represents different scheduling model to achieve timely arrival. For this purpose we tested two different scenarios by actual public bus operation. The data was obtained in the city of Tehran, since Iran was used in this study.

Abstract: Given the necessity for monitoring security situation in operations of expressway service area, the feasibility of building intelligent monitoring system based on neural network expert system was explored. The study has revealed that: the security situation in service-area operations consists of safety situations of every service subsystem in the area; and the safety situations of every service subsystem were, in turn, determined by the busy degree of every service, the higher level of busyness means the worse security situation in operations. According to this, the variables to measure the states of safety situations in every service subsystem were constructed; Based on these variables, the neural network expert system for evaluating and monitoring the general safety states in service area operations was built; Such a neural network expert system has been successfully trained with the samples and data given by the field experts, which accordingly makes it clear the system is feasible.

Abstract: The influence of different traffic information on drivers’ day-to-day route choice behavior based on microscopic simulation is investigated. Firstly, it is assumed that drivers select routes in terms of drivers’ perceived travel time on routes. Consequently, the route choice model is developed. Then, updating the drivers’ perceived travel time on routes is modeled in three kinds of traffic information conditions respectively, which no information, releasing historical information and releasing predictive information. Finally, by setting a simple road network with two parallel paths, the drivers’ day-to-day route choice is simulated. The statistical characteristics of drivers’ behavior are computed. Considering user equilibrium as a yardstick, the effects of three kinds of traffic information are compared. The results show that the impacts of traffic information on drivers are related to the random level of driver’s route choice and reliance on the information. In addition, the road network cannot reach user equilibrium in three kinds of information. This research results can provide a useful reference for the application of traffic information system.

Abstract: A new technique is present in this paper that transforms the railway track irregularity power spectral density of left and right rails into the excitation power spectral density of wheelset of train, compared with the railway track irregularity spectrums of left and right rails which are not the direct inputs of simulation dynamic model of trail. A parameter model is chosen as the model of excitation spectrum and parameters fitting result shows that this model is suitable for the excitation spectrums of traversing, floating and head shaking, and the method present in this paper is effective.

Abstract: As a kind of typical bottleneck influencing the transportation seriously, the lane reduction has seldomly been investigated with the optimal velocity model. In this paper, we study this issue using the optimal velocity model, in which two kinds of vehicles (fast and slow) are introduced. The asymmetric lane changing rules in the slowdown section and the lane squeezing behaviors at the bottleneck are taken into account. Under the periodic boundary condition, the numerical simulations are performed. The fundamental diagram is obtained. Then, the influences of speed limit and the length of the slowdown section on traffic are discussed in detail from the view of traffic control. The currents are dependent on the speed limit, but independent of the length.

Abstract: In this paper, to overcome the limitations of the weighted combination and single objective optimization methods, we presented a multi-objective optimization and simulation methodology for network-wide traffic signal control. A multi-objective genetic algorithm based on Non-dominated Sorting Genetic Algorithm II was given to solve the model directly to obtain Pareto optimal solution set. The objectives were evaluated by Enhanced Cell Transmission Model used to describe traffic dynamics on signalized urban road network. The results showed that the single objective optimization method made some of the objectives worsen when the objective to be optimized reaching optimal, and that the weighted combination optimization method gained a compromised solution, but the multi-objective optimization method gave consideration to more objectives, making the number of optimal or suboptimal ones is more than that of worse ones.

Abstract: The most important and critical step to improve road traffic safety is prediction and identification of traffic accident black spot. A new prediction model of traffic accident black spots is proposed based on GA-BP neural network algorithm and rough set theory. First of all, the traffic accident statistics of Jinwei Road in Tianjin are analyzed. With consideration of static road conditions, the samples of road accident black spots are obtained by the GA-BP neural network algorithm. Furthermore, an effective road traffic accident black spot prediction model is established by utilizing rough set theory with consideration of the impact of real time dynamic conditions. Finally, a numerical example is illustrated. Experimental results show that the proposed model with the combination of these two theories can reduce the hybrid and burdensome amount of data, lower the false alarm rate and improve the forecasting accuracy of accident black spots.