| With the increasing number of motor vehicles,traffic congestion has become a global problem which can affect urban function and sustainable development,and ITS is recognized as the best way to effectively alleviate the traffic jam.Along with the development of advanced technologies such as Sensor networks,Internet of Things,Cyber-Physical Systems,Cloud Computing and Big Data,the intelligent traffic information service model has some new change,which makes full control of the traffic network possible.Now most researches about traditional traffic information supply model are based on rough and incomplete control of the road network conditions,which is different from the actual application environment.In the controlled road network,in order to achieve full controlling,the traffic information supply model should first support accurate traffic state description.Therefore,this paper aims to establish the information model based on traffic state description,and obtain the complete information condition which can describe the traffic condition to guide the construction of traffic information collection system.The main contents of this paper include:Based on the characteristics of traffic information of the controlled road network,this paper analyzes the information requirement of traffic state recognition and ’congestion management,and compares and analyzes the existing traffic information collecting technology,which can be set as the basis of the construction of information collection system.Taking into account the information gathering needs of the road sections,the network structure characteristics of the road sections and their influence on the traffic state description of the road network,this paper defines the evaluation index of road information service level and uses the multi-attribute decision theory to quantify the information service level.The theory of rough set is introduced in this paper,in which the basic concepts are given traffic connotation,and the reasonableness of information extraction is proved.Aiming at traffic status identification,the data expression system is established based on the attributes set of space-time,traffic flow characteristics and state attribute.After analyzing and comparing the effects of various algorithms,we choose the genetic algorithm to carry out data discretization and attribute reduction.According to the results of the reduction,this paper extracts the complete information condition based on traffic state identification.Through analysis of the influencing factors of traffic detectors layout,this paper builds a multi-objective optimization model,taking minimum cost of system,maximum reliability of data,and maximum system information service level as the objective function,and the principle of OD coverage and information integrity as constraints.In addition,the tolerance and hierarchical sequence method is applied in this paper,and we adjust the tolerance coefficient to ensure the feasibility and effectiveness of the multi-objective optimization model.Finally,the genetic algorithm is used to solve this model.In the case study,data of the Nguyen-Dupuis road network is acquired through the VISSIM secondary development technology,and the complete information condition of road netwoks can be achieved by Rosetta software.The complete information condition sets include(speed,traffic flow),(road number,travel time,speed),(travel time,road number,occupancy),(road number,speed,occupancy),(road number,speed,queue length),and the extracted principles can be used to verify the accuracy of the proposed complete information condition.Finally,the optimized layout of detectors is designed by MATLAB programming,which proves the validity and practicability of this model. |
