Passenger flow prediction and control at peak hour for transfer stations of urban mass transit

Based on passenger classification, this paper proposes the model of passenger flow control at peak hour for transfer stations of urban mass transit, and provides a solution to passenger congestion on platform. In order to ensure operational safety at peak hour, the author takes the minimum passenger density on platform as objective function, predicts the “interference passenger flow” by wavelet neural network (WNN), and calculates the optimal control sequence of passenger flow by the model predictive control (MPC) theory. Finally, it is proved that the proposed model can control the passenger flow on platform close to the optimal number of gathering passengers by simulating Huixinxijienankou Station in AnyLogic.