Research On Modelling And Characteristics Analysis Of Traffic Flow Based On Cellular Automaton

In this paper, cellular automaton theory is used to model and simulate some complex phenomena (such as car following, car overtaking, train following operation and delay propagation of trains) in road traffic and railway traffic. The inherent law and forming mechanism of these traffic phenomena are discussed. The main ideas and innovations of this paper are composed of two parts: the modelling and character analysis of traffic flow in road traffic system and railway traffic system.There are two aspects for modelling and character analysis of traffic flow in road traffic system.1. A new single-lane cellular automaton model considering the influence of the nearest neighbor vehicle and the next-nearest neighbor vehicle is given. A type of the driving strategies used in the double look-ahead model has been employed to instead the acceleration rule of NaSch traffic model. Compared with NaSch model, the characters of the proposed model are as follows: (1) The proposed model can reproduce some complex traffic flow states that have been observed in the real traffic, such as go-and-stop wave, and both the maximum flux and the critical density are closer to the real measurement than that of NaSch model. (2) The influence of sensitivity coefficients of the nearest neighbor vehicle and the next-nearest neighbor vehicle on maximum flux and critical density is discussed respectively.2. The single-lane cellular automaton model considering the influence of the nearest neighbor vehicle and the next-nearest neighbor vehicle is extended to the two-lane system. The attributes of the traffic system composed of fast-lane and slow-lane are investigated by the new traffic model. The simulation results are as follows. (1) The proposed two-lane traffic model can reproduce some traffic phenomena observed in the real traffic, such as go-and-stop wave and density invertion, and that maximum flux and critical density are close to the field measurements. (2) The initial density distribution of fast-lane and slow-lane has much influence on the traffic flow states. With the ratio between the densities of slow-lane and fast-lane increasing, lane changing frequency increases, but maximum flux decreases. (3) The influence of the sensitivity coefficients is discussed.There are two aspects for modelling and character analysis of traffic flow in railway traffic. 3. A new cellular automaton model in railway traffic with moving block system is proposed to study the characters of the mixed railway traffic flow with different maximum velocity and acceleration. The simulation results show as follows. (1) The proposed model can describe some phenomena of traffic flow in railway traffic, such as train following operation and the delay propagation of trains. (2) Different from former cellular automaton model of railway traffic, the time-space diagram without delay of the proposed model is asymmetric, and the more the difference of maximum velocity between slow train and quick train is, the more the difference of the curves of the diagram is. (3) The space headway is related to the following sequence of slow train and quick train. (4) The influence factors of train following operation are discussed, such as train time interval, station dwell time and safe section behind station, etc. (5) It is pointed out that the minimum train time interval is related to train following distance.4. A railway cellular automaton model in which the station is composed of several tracks is proposed to study the influence of station for train following operation and the delay propagation of trains. The simulation results show as follows. (1) The delay of 'pass by main track, start and stop by siding track' scheme is more than that of the 'three tracks play the same role' scheme. (2) The increase of the track numbers can reduce some delays before the station, other than improve the passing ability of lines. (3) The main reason of the delays is that the trains dwell in the station for some time and compete for limited resources (such as tracks and safety section). The influencing factors of the delays are also discussed. (4) The sequence of different types of trains has great effect on train following operation, which shows the importance of the organization of train operation.