Simulation model of speed-density characteristics for mixed bicycle flow-Comparison between cellular automata model and gas dynamics model

The mixed bicycle flow refers to the bicycle flow containing electric bicycles. The traffic characteristics data of the mixed bicycle flow was collected by the virtual coil method in Nanjing and Ningbo, China. And the speed-density characteristics of the mixed bicycle flow with different proportions of electric bicycles were obtained. The results show that the overall speed of the mixed bicycle flow containing electric bicycles is higher than that of pure bicycle flow when the density is relatively low. The speed decreases when the density is higher than 0.08 bic/m2; the speed-density characteristics of the bicycles and the electric bicycles tend to be the same when the density is higher than 0.25 bic/m2. And when the density reaches 0.58 bic/m2, the mixed bicycle flow becomes blocked and the speed is zero. The cellular automata model and gas dynamics model were also adopted to simulate the speed-density characteristics of the mixed bicycle flow. The simulation results of the cellular automata model are effectively consistent with the actual survey data when the density is lower than 0.225 bic/m2; the simulation results of the gas dynamics model are effectively consistent with the actual survey data when the density is higher than 0.300 bic/m2; but both of the two types of simulation models are inapplicable when the density is between 0.225 and 0.300 bic/m2. These results will be used in the management of mixed bicycles and the research of vehicle-bicycle conflict and so on.