入口匝道控制是解决城市快速路拥堵的一条重要途径，采用有效的控制方法提高快速路系统的使用效率，加强快速路系统的交通控制，对缓解城市快速路的交通拥堵有着重要作用。提出了基于神经网络预测对快速路主路交通流量进行预测，进而根据每分钟流量与期望流量差值和两分钟内流量差值来控制匝道调解率，依据匝道控制率调节每分钟绿信比，使得每分钟入匝流量得以控制。文章中模糊变量选用三角型隶属度函数，引入了49条模糊规则，用MATLAB进行系统仿真，得出匝道控制率，通过对红绿灯时长进行调节对匝道上的车流量进行放行，经验证实验结果有效可靠。%This article proposes a neural network prediction method to predict traffic flow based on the mainstream,then according to each minute difference between flow and expected flow and the flow difference of two adjacent minutes to control the ramp rate,then uses ramp metering rate to adjust green time of per minute,in this paper uses the triangle mem-bership function of fuzzy variables,and introduces 49 fuzzy rules,and uses MATLAB to make system simulation to get the ramp control rate,based on the red and green time to adjust the traffic flow on the ramp.
展开▼
