Research On Quantum Particle Swarm Optimization For Coordination Signal Timing Optimization Of Multi-intersection

With the increasing deteriorative of the road congestion, traffic signal control research as an important means to alleviate the city traffic conditions are concerned more frequently. This paper established a single-intersection signal timing optimization model using some indicators such as the least number of vehicles queuing,average stop delay,parking rates and so on. For the case of multiple intersection control, a traffic control model of multi-intersection was designed, an improved Quantum Particle Swarm Optimization (QPSO) was proposed to solve this problem.1. When the time period was fixed, a single intersection control model was established which using the number of vehicles queuing as the performance indicator, a Quantum Particle Swarm Optimization was designed to solve this kind of problem. Compared with the fixed timing situation, this algorithm has more advantages.2. A single Intersection signal timing optimization TSP model was established which using the average stop delay as the performance indicator, a QPSO-based selection was designed to solve this kind of problem. Simulation result shows that this algorithm could improve the intersection capacity, reducing the vehicle delay, achieve real-time optimization of traffic signal control.3. A multi-objective optimization model was established which using the number of vehicles queuing,average stop delay of vehicles,parking rates as the performance indicators, Cauchy mutation-QPSO was used to solve this kind of problem, it overcome the premature search successfully. The simulation result shows the effectiveness of this algorithm.4. With the multi-intersection problem, the characteristics of road traffic flow was studied,2×2 of intersection traffic control model was established which using the least number of vehicles as the target. The quantum particle swarm algorithm was improved using the random chaos and unique traverse properties, the global search capability of the algorithm was enhanced, the premature convergence was overcomed. The simulation result shows that the model and the corresponding algorithm were correct and effective.