Abstract:

In the signal optimization control process of the intersection group, a signal optimization control method for the associated intersection sub-area is proposed for the problem that the genetic algorithm prematurely converges and the relevance of adjacent intersections is considered less. Firstly, the soft-collection theory is used to divide the intersecting intersections into the same sub-area. Then, the niche technology based on shared function is used to adjust the fitness of individuals in the group and adaptively adjusts the crossover probability and mutation probability of the algorithm to improve the genetic algorithm. Finally, the improved genetic algorithm is used to optimize the average delay time of the associated intersection sub-area. The simulation experiments of the measured data of the road network show that the proposed method divides the intersection group into reasonable sub-areas, and the improved genetic algorithm reduces the number of iterations in the sub-area signal optimization control, which makes the average delay time of the intersection shorter.

Key words: signal optimization control, soft set theory, improved genetic algorithm, average delay time

摘要：

在交叉口群的信号优化控制过程中，针对遗传算法早熟收敛且对相邻交叉口关联性考虑较少的问题，提出一种关联交叉口子区的信号优化控制方法。利用软集合理论将关联性强的交叉口划分在同一个子区；采用基于共享函数的小生境技术调整群体中个体的适应度并自适应地调整算法的交叉概率[Pc]和变异概率[Pm]对遗传算法进行改进；使用改进的遗传算法对关联交叉口子区的平均延误时间[D]进行优化。路网实测数据的仿真实验表明本文方法对交叉口群进行了合理的子区划分，且改进的遗传算法在子区信号优化控制中迭代次数减少，使得交叉口的平均延误时间更短。

关键词: 信号优化控制, 软集合理论, 改进的遗传算法, 平均延误时间