Micro Mode Based On The Complexity Of The Dynamic Characteristics Of Three-phase Flow Theory Of Traffic Flow

As a new interdiscipline research field, traffic flow theory has attracted many scientists and engineers in recent years. The long term goal of traffic flow research is to construct a kind of model which can describe and reproduce the features of real traffic and can be used to bring the basic rule of traffic flow to light. Thereby, these research results could help the the traffic engineering department to plan, design, construct and improve the urban traffic network and traffic control system. At the same time, as a typical self-driven system, traffic flow theory can raise awareness and promote research work on statistical physics, fluid mechanics, nonlinear dynamics, applied mathematics, traffic engineering and so on. Therefore, the traffic flow research, not only has practical engineering value, but also far-reaching scientific importance.To create realistic traffic flow model, first of all, there must be a reasonable traffic flow theory as a basis. There are two dominant frameworks in traffic flow research area, which are fundamental diagram approach and three-phase theory. As the traditional theory of traffic flow, fundamental diagram approach is based on the relationship between the vehicle density and flow. It divided traffic flow into free flow and congested flow. And the three-phase theory, which is the most advanced traffic theory, further divided the congested flow into synchronized flow and wide moving jams. Because the models based on the three-phase theory could better simulate the real traffic flow, especially the congested patterns in traffic bottleneck, this theory is gradually being accepted by the scientists in recent years.Besides an reasonable theory, we also need an appropriate modeling tool to construct traffic models. Since the birth of the modern traffic flow, more than 100 theoretical models have been proposed by traffic scientists and physicists, from probabilistic theory models in the 1930s, to kinematic models and car-following models in the 1950s, and then to fluid-dynamical models in the 1970s-1980s, all of which play a vital role in describing the complex phenomena in traffic. Since the f990s, cellular automaton (CA) models, with simple rules, clear definition, good expansibility and high computational efficiency, have risen in traffic flow theory as a new force and gradually win the good graces of the traffic scholars and engineers. In this paper, we use CA to construct more realistic traffic flow model and reveal the various traffic flow characteristics through theoretical analysis and numerical simulation. Besides, we use time series analysis method to give a clear, mathematical characteristics to each phase in three-phase theory. The contents of the paper are as follows.1. We investigates the effect of acceleration a and randomization D in a refined Nagel-Schreckenberg traffic flow model, in which each cell corresponds to 1.5m. It is found that when a≥D, there is no qualitative change of the fundamental diagram and spatiotemporal pattern. In contrast, when a < D, qualitative changes of traffic flow characteristics appear, and the congested traffic flow exhibits the characteristics of synchronized flow. Based on this finding, we study spatiotemporal patterns induced by a speed limit bottleneck by using the refined model with a < D and considering slow-to-start effect. It is shown the patterns are consistent with the results of three phase traffic theory. Our work is expected to shed some light on the mechanism of synchronized flow, which might be related to the value of a/D.2. We proposes anticipation effect as another possible mechanism that might be origin of synchronized flow. The simulation of this effect is performed based on a new cellular automaton model. Since the overreaction in deceleration is re- moved due to the introduction of anticipation effect, qualitative changes of traffic flow characteristics appear compared with the results of NaSch model, and the congested traffic flow exhibits the features of synchronized flow. We also study the spatiotemporal patterns induced by an on-ramp. It is shown that the patterns are consistent with the three phase traffic theory. Our work is expected to shed some light on the mechanism of synchronized flow.3. We use the detrended fluctuation analysis(DFA) method to examine the long-range power law correlations of each phase and address the question of whether each scaling exponent corresponds to specific phase. We analyze the traffic flow time series of two important cellular automaton(CA) models, NaSch model and Kerner-Klenov-Wolf(KKW) model. In particular, KKW model is based on three-phase theory. We make comparison between the analysis results of two traffic data sets. Application of the DFA method shows evidence for crossover phenomena associated with a change in short and long-range scaling exponents. This method may be of use in distinguishing the different traffic states from traffic data sets based on differences in these scaling properties.