Research On Risk Identification And Early Warning Of Stampede In Closed Place Considering Crowd Pressure

As a carrier of entertainment activities and sports events,closed places play an important role in daily life.However,the safety problems caused by dense people in closed places are becoming more and more serious.How to recognize and quantify the trampling risk of dense people in closed places and monitor and warn them in real time to prevent trampling events has become the research content of many scholars,Based on this,this rearsch focuses on the following contents.Firstly,the statistical trample accident is analyzed and the risk analysis and identification of crowd trampling are completed,and the cause,mechanism and characteristics of the accident are defined.The early warning stage is determined by the accident occurrence mechanism,and based on the accident cause,the crowd crowded trample accident tree is constructed and the accident tree is qualitatively analyzed,the key risk factors are identified and preventive measures are put forward for the cause of the accident.Secondly,according to the identified key risk factors,dynamic indicators are selected,namely,number,density and speed.The risk of dense population is measured by the internal pressure of the crowd,and the maximum extrusion force is transformed into quantitative relationship among the number,density and speed.The process of quantification of maximum extrusion pressure in dense population is:adjusting the social force model and combining with particle flow theory,setting up scene in PFC2 D,and first,building the crowd evacuation experiment scene like Helbing scholars to verify the effectiveness of the model and software,The model and software are effective by observing the curve of different expected speed and evacuation time,and the distribution and stress state of the crowd in the simulation process.Secondly,the scene is set up to carry out numerical experiments under different conditions,and the maximum crowd extrusion pressure time-varying graph is output,and the quantitative data is completed.The fitting results show that the quantitative relationship between the maximum extrusion pressure and population density and speed is 0.972706.At the same time,the research on the movement state of different places of the population is completed.It is found that the symmetrical distribution of the population has similar movement properties during the evacuation process,and the evacuation time is the longest,the most stressed and the most dangerous evacuation time near the wall.Finally,according to the current stress threshold and duration relationship,the trampling risk of dense population is divided into four levels.Combined with the quantitative formula of maximum extrusion pressure,the FNSD early warning model is constructed.FNSD model can alert and carry out evacuation work when the medium risk is reached.Finally,the validity of the early warning model is verified by using the data of Beijing Huilongguan metro station and the wavelet neural network.FNSD warning models are used to predict the trampling risk seven hundred times at 5 minutes interval;The number of times of correct prediction risk level of FNSD early warning model is 623 times,and the accuracy index of the model prediction risk level reaches89%.In practice,it can be used to monitor and early-warning the trampling risk of dense people in real time,which plays an important role in preventing stampede events in traffic stations and venues.