System Identification of Dynamic Control Design for Diversion Railings During Evacuations in Subway Stations

The diversion railing of subway stations can orderly divert passenger flow, playing an important role in guiding passengers to move along specific paths, maintaining order, and improving station operation efficiency and safety. The Massmotion software embedded with the social force model and the minimum cost model is used to simulate and analyze the impact of different railing lengths and desired crowd velocities on the evacuation process of passenger flow at the platform stair under the same passenger flow background. It is found that the flow control effect of the diversion railing with a certain internal inclination angle is better than that under the right angle by comparing the passenger flow density under different working conditions. The dynamic characteristics of evacuees characterized by the length of railings are investigated, and input and output data are obtained. Different system recognition models are adopted for data identification and validation. The results indicate that the presence of railings effectively reduces the density of personnel at the bottleneck of the stairs, but the increase in the desired passenger speed is more pronounced than the optimization degree brought about by the increase in railing length. In addition, the dynamic length change of the railing can affect the crowd density at the bottleneck of the stair in real time, and an accurate polynomial ARX model is obtained, which can provide a strong reference for the dynamic control system of the diversion railings.