| Under the "Internet plus" environment,the prediction ability based on big data and the ability of evaluating and analyzing traffic operation conditions and active problem finding based on real-time data can accurately formulate,optimize and evaluate the scheme.Facing the acquisition equipment that may be popular in the future and the traffic big data formed,this paper selects the traffic big data formed by the electronic license plate system as the historical and real-time data source,and studies the dynamic signal control of single intersection.The specific contents are as follows:(1)This paper firstly introduces the constitutions and operation mechanism of the cooperative environment between the electronic license plate and the traffic sensing network,clarifies the types of the electronic license plate sensors to be used in the signal control,and formulates the sensors’ layout scheme.By analyzing the data constitutions and layout of the traffic big data generated in this environment,the data content that the dynamic signal control needs to obtain from the base data center is clarified,and the data information extraction algorithm is designed.(2)In order to control the intersection traffic running state,and queuing changes of intersection in real time,the optimal installation position of queue detection sensor is determined by analyzing the queue dissipation process of inlet channel,and a model and algorithm for designing and evaluating the required parameters are established by using sensors’ acquisition data to obtain the number of queuing vehicles in real time,queuing assembly wave velocity and queuing dissipation wave velocity.(3)Based on the traffic big data produced in the cooperative environment of electronic license plate and traffic induction network,a single intersection is taken as the research object.Firstly,the current timing scheme is optimized by using the historical data extracted from the base data center and the iterative calculation method,to obtain the basic timing scheme based on minimum delay.Secondly,the real-time data collected by the sensors is used to obtain the indicators of stopping times and the number of vehicles in queuing,in order to identifying the single-cycle operation states of the intersection.According to the idea of queuing equalization control,the end of the red light and the end of the green light are selected as the decision points of queuing equalization control.Then,with the four-phase signal control as an example,based on the real-time data acquisition of sensors,the adjustment model of the timing scheme for the undersaturated state of the intersection is constructed,in addition,the green time adjustment model the new phase algorithm for the supersaturated state of the intersection is constructed,a traffic congestion dredging strategy mapped by data acquisition is proposed.Finally,the performance of the dynamic timing scheme is evaluated by the data collected by sensors.(4)In the case study,Anylogic software was used to simulate the real-time data and real-time feedback control process of the dynamic timing scheme under the cooperative environment of the future electronic license plate and traffic sensing network.The logical and spatial layers of the model are built by using the road traffic library and process library of the software,then,the parameters and variables and functions of timing calculation and evaluation are introduced into the logic layer,and the timing scheme is formulated through simulation.By analyzing the change of the queueing vehicles with time,the average stopping times and the average total delay in the left and the direct traffic flow,it is concluded that the dynamic timing scheme can be more suitable for the time-varying traffic flow than the basic timing scheme,the dynamic timing scheme can better adapt to the time change of the traffic flow,which makes the directional traffic flows change steadily periodically and equalize each phase queuing. |
PDF Full Text Request