Microscopic Simulation Modeling And Evacuation Guidance Optimization Of Pedestrian Evacuation In Different Scenarios

With the development of Urbanization, the large public activities such as parade and assembly are increasing fastly today. In large public places, pedestrians are always complex and in high density. If emergency occurred, a small turbulence may lead to an unstable state, or even serious stamped. Pedestrian evacuations have been widely concerned by scholars. Making scientific evacuation strategies and effective controlling policies are the keys to re-duce the accidents and guarantee the safety of pedestrians. On the basis of the pedestrian microscopic simulation theory, this dissertation analyses pedestrian movement behaviors, evacuation plans and guidance in different scenarios based on simulation models and opera-tion models. The main research works of the dissertation are as follows:(1) According to the counter pedestrian behavior characteristic in the channel, a im-proved random partial walk lattice gas model is proposed. By introducing the concept of the pedestrian view field, the effect of pedestrians’movement preference feature of open area in front of the view field are taken into account. Three dynamic evolution processes under dif-ferent pedestrian density are reproduced. The randomness of lane formation under different pedestrian density is found, and the probability of lane formation is given. Numerical simu-lation of the relationship diagrams between the probability of lane formation and parameters of system geometry size, the proportion of right walker flow, the strength of the drift, view field size are investigated. The result show that extended model cannot form the lane forma-tion under low pedestrian density, which is corresponding with real pedestrian traffic. It is found that the density of pedestrian counter flow could divide into five intervals, and there are differences in the dynamic evolution processes between five intervals. This model and result is useful to study the dynamic evolution process, and helpful for raising efficiency of pedestrian counter flow in channel.(2) An improved social force model combined with discrete potential field is proposed to determine the local pedestrian veloctiy direction for complex evacuation scene which have internal barriers. A new method of calculating the discrete potential field is put forward in the model, thus the desired speed along the gradient descent direction calculation formula is given. Considering to the channel scenes and indoor evacuation scenarios, this paper compare the desire velocity direction calculated by this model to by the original social force model, and discusses the differences between this model and the original model of social forces. The scope of applicability of the model is also obtained. Based on comparative analysis of time efficiency, it is show that the model can greatly improve the simulation of complex scenes insiders evacuation efficiency in ensuring the accuracy of the premise. Taking a supermarket as an example, this paper applied the improved social forces model to analyze the emergency evacuation procedure of pedestrian in supermarket. The results show that there is an safety number capacity of supermarket, and while the evacuee number exceeds the safety capac-ity, evacuation time increases linearly rapid and the risk degree also greatly increased. It could optimize the evacuation process while train 70% pedestrian to konw the evacuation exit information, but not all.(3) In order to solve the problem of exit using imbalance caused by a blind exit choice of evacuee during multi-exit room evacuation, an optimization model of exit choice is pro-posed under the multi-exit room evacuation based on point queue model. While, a hybrid algorithm based on genetic and simulated annealing algorithm is designed to solve the op-timization model. For demonstrating the effective, a cellular automate model with certainty exit choice decision is presented to investigate the influence of different multi-exit selection schemes on evacuation time, including the multi-exit selection scheme obtained by our op-timization model. The comparative results show that the exit selection scheme obtained by our optimization model based on the point queue model can balance the utilization of the multi-exit and decrease evacuation time. The more evacuees there are inside the room, the more evacuation time saved by the model. Thus, the optimization model is helpful to make pedestrian evacuation scheme.(4) Considering the limitations of tranditional evacuation signs on the evacution guid-ance, a bi-level programming model is established to describe the dynamic evacuation signs guidance strategy in complexity scenarios. The upper level model aimed the minimum of the total evacuation route impedance. The solution of upper level model is the evacuation sign direction, whcih can be used to guide evacuees walk direction; The lower level model is an instantaneous dynamic user optimal model(DUO). With the DUO model pedestrian flow can be allocated according to the route impedance and queuing time of exit, and make an balance between each evacuation routes and exit. Finally, a simulation model framework, which combined bi-level programming model and social force simulation model, is estab-lished to describe the pedestrian evacuation behavior. The result shows that the evacuation signs can be divided into critical signs and non-critical signs. Critical signs play an important role in inducing the evacuation time. The bi-level model can induce the evacuation time by the extention of partly pedestrian time. And it’s more effective guide the evacuees in the late stages of evacuation prograss while in a limited number of dynamic evacuation guidance.