Study On Phase Transition Phenomenon Of Multiparticles Nonequilibrium System

Numourous multi-particles systems exsit widely in various fields of nature and human society. As systems far from equilibrium, self-propelled particles or driven diffusive system appears many non-equilibrium physical phenomena, such as the phase transition, metastable state, hysteresis, self-organized criticality, phase separation. However, the relationships among the particles of the whole system are so complicated that. In recent years, the physicists have developed many methods to explore the features of the multi-partciles systems which are far from equilibrium, such as cellular automata, complex networks and so on. These methods have promoted remarkably the study of many non-equilibrium multi-particles systems in nature and society. However, the big difference presents betweent simulation results and realistic situations because some classic models are designed relatively simple. In this paper, we have researched a few multi-particles systems from the physical significance, developed some new models and carried out numerical simulations and theoretical analysis. The contents of the paper are as follows:1. We have applied the realistic traffic characteristics in the BML model, studied the three extened models, simulated and analyzed the non-equilibrium physical phenomena.●BML model with car-length:We consider the car with ω length can occupy co cells and study the influence of car length on the average velocity and the on the structure of the intermediate stable phase. The simulation results show that all the three phases could coexist in a density range whenω>1. More interesting, the structure of the intermediate stable phase qualitatively changes. There appear more than two stripes in each direction. However, the average velocity of the phase essentially does not change.●BML model with two-level crossings:We investigate the effect of two-level crossings on the traffic jam in BML model. We found that the first-order phase transition occurs from free flow to global jams when the density of the two-level crossings is low. However, as the density increases, the average velocity appears linear relationship with the density of cars in free flow phase, the system occurs more intermediate stable phase simultaneously. The parameter cf presents the probability of the two-level crossing is occupied by northbound car and eastbound car. In global jams phase, the relationship between cf and the density of cars follows linear or power-law functions with different density of crossings.●Manhattan urban model with single road:We study a cellular automaton model of vehicular traffic in Manhattan-like urban model with single road. The origin-destination trips and traffic lights have been considered. We have found the first-order phase transition occurs from free flow state to global deadlock state. With a grid coarsening method, vehicle distribution in the moving state has been studied. The effect of boundary optimization strategy (BOS), the traffic light period, the length of road and the count of the cross have also been investigated.2. We study the motion of self-propelled particles model in a two-dimensional lattice model. We have investigated effect of alignment sensitivity g, as well as number of orientations and neighbors, the leadership particle on the collective behaviors of particles. It is found that the mean orientation <m> changes non-monotonically with g. The number of orientations and neighbors has only quantitative influence on the non-monotonic behavior. It is also shown that the steady state depends on the initial configuration at large value of g. We have found the similar non-monotonic phenomenon in one-dimensional and three dimensional lattices. Interestingly, the particles couldn’t trend to the orientation of the leader particle, the leadership hardly play a role in self-organized motion.3. Drive diffusion systems are a far from equilibrium system that have shown a steady and time-independent state. We study a periodic one-dimensional exclusion process composed of a driven and biased diffusive part in mesoscopic limit. It’s shown that depending on the biased diffusion parameter δ, several different phase diagram topologies occur, in which a rich phase behavior has been exhibited and the density profile in the diffusive part is qualitatively different. Our analytical results are in good agreement with Mote Carlo simulations.