Modeling analysis of T-shaped crowd flow based on artificial neural network

Inhalt:
In recent years, stampede incidents occur frequently in indoor public places under emergencies, posing a great threat to the safety of people's lives and property. Therefore, we believe that it is necessary to conduct in-depth research on related issues. In previous studies, although there have been algorithm studies on the population in the T-shaped channel, there is no relevant study on fitting with neural networks. Therefore, we use a fully connected network to perform fitting learning on the crowd walking under T-shaped intersections and find that the learning effect is good. We believe that the results of this research can be put into practical simulation to improve the design of interior buildings and to prevent safety accidents in advance.