Compared with the ordinary expressway, the ring expressway has more ramps, and most of ramps are connected with the urban roads. Due to the great influence of urban traffic, its traffic flow is fluctuant and appears mutations frequently. The classical on-ramp control model is difficult to deal with the traffic flow mutations, the controller will be failure when the mutations happen. In order to solve this problem, the paper designed the structure of a real-time extension controller for the ring expressway on-ramp with the extension control theory and the grey prediction method. According to the deviation and differential of the actual and the expected traffic flow density of the main line, the designed controller divides the main line traffic station into three measure modes, and presents different control algorithms to each mode. Finally, takes a section of Lanzhou North Ring Expressway as an example to valuate and analyze the designed controller, the Simulink software is used to simulate the control process. The results show through the designed controller, the traffic density of main line converges to the expected density, the fluctuation of on-ramp metering rate is reduced, and the average queue length of the on-ramp vehicles is shortened.

CICTP