Optimization Model Of Signal Timing At Intersections Of Stochastic Chance Constrained Programming

With the development of intelligent traffic control and urbanization,the problem of urban road traffic congestion becomes more and more serious an important node affecting the efficiency of urban traffic operation,the signal control scheme plays a vital role in traffic operation.The traditional signal control method has not adapted to the randomness and uncertainty of traffic flow,according to the stochastic characteristics of traffic flow to optimize the signal distribution time is the attention of many scholars at home and abroad,the urban intersection signal control optimization problem research is of great significance.This paper takes the signal control optimization problem of urban Road intersection as the research object,considers the randomness of traffic flow and combines the principle of opportunity constraint programming,sets up a single target and multi-objective signal control optimization model,uses the adaptive particle swarm algorithm to solve the model,and obtains the improved time matching scheme.Based on the existing research results,the following work has been done:(1)The development history of Signalized intersection is reviewed,the basic concept of traffic signal control theory is introduced,several evaluation indexes affecting the operation efficiency of intersection are expounded in detail,the stochastic distribution of traffic flow is sorted,and the specific distributing function and adaptive range are given.(2)The principle and algorithm flow of elementary particle swarm algorithm(PSO)are sorted out,an adaptive particle swarm algorithm(APSO)is proposed to improve the algorithm,which is based on the randomness and uncertainty of traffic flow,and it is processed by penalty function method in order to get the fast convergence speed and high precision.At the same time,it overcomes the shortcoming of getting into the local optimum and improves the global searching ability of the algorithm.(3)Based on the minimum absolute value of each phase green time and the required green light difference,a timing optimization model for random chance constrained programming is established.By combining the adaptive particle swarm algorithm with the stochastic simulation method,the calculation results show that the stochastic characteristic of arrival rate has a significant effect on the setting of signal time parameters of intersections,and the correctness of the model and algorithm is validated.(4)With the least number of stops,the average delay is the minimum and the maximum capacity is the goal,the Multi-objective signal control optimization model based on stochastic chance constraint programming is established,and the APSO algorithm and PSO algorithm are used to solve the model under different saturation states.The optimization results show that: in three cases of high saturation,medium and low,the delay relative to PSO algorithm is reduced by 8%-15%,the number of parking times is reduced by 8%-13% compared with thePSO algorithm,and the capacity of APSO is improved by 6%-11%,which verifies the correctness of the algorithm and model.