基于改进粒子群算法的交通控制算法研究

发布时间：2018-03-01 07:47

本文关键词： 粒子群算法遗传算法交通流量控制交叉操作变异操作　出处：《长安大学》2015年硕士论文　论文类型：学位论文

【摘要】：随着城市规模的扩大和汽车保有量的不断壮大,针对有限的城市交通资源和急剧增加的汽车数量,在保证交通流量合理有序的前提下,如何最大限度地发挥现有城市交通网络的通行能力,是当前交通控制研究的重点和难点。首先,本文针对粒子群算法存在局部最优的问题,将遗传算法的交叉操作和变异操作引入粒子群算法对其进行改进,详细阐述了改进粒子群算法的算法流程。四个标准测试函数收敛图对比发现,在收敛速度和稳定性方面,改进的粒子群算法优于遗传算法和粒子群算法。其次,为了提高交通流量控制和优化的精度,将混沌理论引入PSO对LS-SVM的核参数和惩罚系数进行优化选择,提出一种ECLS-SVM交通流量预测模型。通过基于ECLS-SVM算法的单步、3步、5步和7步预测结果和不同模型的预测时间和预测均方误差的对比结果可知,ECLS-SVM算法可以有效提高交通流量预测的精度和效率,对指导交通网络资源的合理分配和规划具有重要的理论意义和实际价值。在交通流量预测的基础上,运用粒子群算法实现城市单交叉路口和双交叉路口交通信号灯的优化控制,达到缓解城市交通拥堵的压力和提高城市交通效率的目的。针对交通信号控制的具体实例,建立单交叉路口和双交叉路口交通控制数学模型。再次,针对标准PSO算法存在局部最优和约束条件的问题,运用GA算法对标准PSO算法进行改进,之后将改进的粒子群算法GA-PSO应用于交通控制上。在单个交叉路口模型的基础上,结合以往交通控制模型,运用改进的粒子群算法对交通控制算法进行优化并与未改进的PSO算法进行对比,发现改进的粒子群算法更优。在此基础上,研究双交叉路口,建立交通协调优化模型,再运用改进的粒子群算法对该模型进行优化,并与标准PSO算法进行对比。最后,通过标准PSO和改进的GA-PSO算法的交通控制算法对比研究发现,引入交叉操作、变异操作的粒子群算法,可以增加全局搜索能力,同时可以避免陷入局部最优解。改进的PSO算法较标准PSO算法在解决交通流量控制问题的时候,完全避免了标准化误差、统计不完善、局部收敛等问题,能够很好地实现交通流量最优化控制。
[Abstract]:With the expansion of the city scale and the continuous expansion of the vehicle ownership, the limited urban traffic resources and the rapidly increasing number of vehicles, under the premise of ensuring a reasonable and orderly traffic flow, How to maximize the capacity of the existing urban traffic network is the focus and difficulty of current traffic control research. Firstly, the particle swarm optimization algorithm has the problem of local optimization. The crossover operation and mutation operation of genetic algorithm are introduced into the particle swarm optimization algorithm to improve it, and the algorithm flow of the improved particle swarm optimization algorithm is described in detail. By comparing the convergence diagrams of four standard test functions, it is found that the convergence speed and stability of the improved particle swarm optimization algorithm are analyzed. The improved particle swarm optimization algorithm is superior to genetic algorithm and particle swarm optimization algorithm. Secondly, in order to improve the accuracy of traffic flow control and optimization, chaotic theory is introduced into PSO to optimize the kernel parameters and penalty coefficients of LS-SVM. This paper presents a ECLS-SVM traffic flow forecasting model. By comparing the prediction results of three steps, five steps and seven steps based on ECLS-SVM algorithm with the prediction time and mean square error of different models, it can be seen that ECLS-SVM algorithm can effectively improve traffic performance. Accuracy and efficiency of flow forecasting, It is of great theoretical and practical value to guide the rational allocation and planning of traffic network resources. On the basis of traffic flow prediction, the particle swarm optimization algorithm is used to realize the optimal control of traffic lights at single and double intersections. To alleviate the pressure of urban traffic congestion and improve urban traffic efficiency. In view of the specific example of traffic signal control, the mathematical model of traffic control at single intersection and double intersection is established. In order to solve the problem of local optimization and constraint in standard PSO algorithm, GA algorithm is used to improve the standard PSO algorithm, and then the improved particle swarm optimization (GA-PSO) algorithm is applied to traffic control. On the basis of a single intersection model, the improved particle swarm optimization algorithm (GA-PSO) is applied to traffic control. Combined with the previous traffic control model, the improved particle swarm optimization algorithm is used to optimize the traffic control algorithm and compared with the unimproved PSO algorithm. It is found that the improved particle swarm optimization algorithm is better. The traffic coordination optimization model is established, and then the improved particle swarm optimization algorithm is used to optimize the model, and compared with the standard PSO algorithm. Finally, through the comparison of traffic control algorithm between standard PSO and improved GA-PSO algorithm, it is found that, The particle swarm optimization algorithm with crossover operation and mutation operation can increase the global search ability and avoid falling into the local optimal solution. The improved PSO algorithm is better than the standard PSO algorithm in solving the traffic flow control problem. The problems of standardization error, statistical imperfection and local convergence can be avoided completely, and the optimal traffic flow control can be realized well.
【学位授予单位】：长安大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U491.54;TP18

【参考文献】