基于黄金分割双种群遗传算法的区域交通信号控制

Regional Traffic Signals Control with Golden Ratio Dual Population Genetic Algorithm

摘要:

针对非均匀交通流的城市区域信号配时优化问题,以区域总通行能力和总延误为优化目标,构建基于目标相对占优策略的城市区域交通信号优化模型。在采用遗传算法对优化模型进行求解时,由于遗传算法易早熟收敛会导致寻优效果不佳,因此引入黄金分割法对双种群遗传算法进行改进,同时在双种群之间进行个体迁移增加种群多样性,减小算法陷入局部最优的可能性,利用4个标准测试函数进行验证,实验结果表明改进的算法能够快速搜索到全局最优解。最后采用数值计算对提出的模型及算法进行效用评价,结果表明,所建模型符合实际交通控制目标并且计算简单,验证了模型的有效性；所改进的算法在城市区域路网中能够有效地获得良好的信号配时方案。

Abstract:

Aiming at the problem of urban regional signal timing optimization with non-uniform traffic flow, the regional total capacity and total delay are taken as the optimization goals, and the urban regional traffic signal optimization model based on the target relative dominance strategy is constructed. When the genetic algorithm is used to solve the optimization model, because the genetic algorithm tends to converge prematurely, the optimization effect is not good. Therefore, the golden ratio method is introduced to improve the dual population genetic algorithm, and at the same time, individual migration between the dual populations increases the diversity of the population. To reduce the possibility of the algorithm falling into the local optimum, four standard test functions are used for verification. The experimental results show that the improved algorithm can quickly search for the global optimum solution. Finally, numerical calculations are used to evaluate the effectiveness of the proposed model and algorithm. The results show that the built model meets the actual traffic control goals and the calculation is simple, which verifies the effectiveness of the model; the improved algorithm can be effectively obtained in the urban area road network Good signal timing scheme.