HCO-ADE求解单点交叉口信号配时问题

HCO-ADE solves the signal timing problem of a single point intersection

为克服信号配时问题求解中单一约束优化方法的局限性,提出混合约束优化自适应差分进化算法(HCO-ADE)。以交叉口通行效率为目标建立信号配时优化模型,依据模型中约束条件被满足程度,采用外罚函数法和可行性准则法相结合的混合约束优化策略(HCOS)对约束条件进行处理,并提出饱和度偏好准则作为自适应差分进化算法(ADE)的选择策略,构建HCO-ADE。最后,以哈尔滨市四相位交叉口高峰时段5 min采集标段交通量数据为例,验证HCO-ADE。结果表明:提出的信号配时问题求解方法可行、有效,优于基于饱和度区间可能度的罚函数约束优化自适应差分进化算法(PFCO-ADE),所得信号配时方案较实测方案平均延误减少3.32%,通行能力提高2.16%,能够更好地提高信号配时方案稳定性。

In order to overcome the limitation of single constraint optimization method in signal timing problem solving,a hybrid constraint optimization adaptive differential evolution algorithm(HCO-ADE)is proposed.The signal timing optimization model is established with the intersection traffic efficiency as the objective.According to the degree of satisfaction of the constraints in the model,the hybrid constraint optimization strategy(HCOS),which combines the external penalty function method and the feasibility criterion method,is used to process the constraints.The saturation preference criterion is proposed as the selection strategy of adaptive differential evolution(ADE),and HCO-ADE is constructed.Finally,the HCO-ADE was verified by taking the traffic volume date collected 5 min during the peak hour data of four-phase intersections in Harbin City as an example.The results show that the proposed method for solving the signal timing problem is feasible and effective,and is superior to the penalty function constrained adaptive differential evolution algorithm(PFCO-ADE)based on the possibility of saturation interval.Compared with the measured scheme,the average delay of the obtained signal timing scheme is reduced by 3.32%,and the traffic capacity is increased by 2.16%,which can better improve the stability of the signal timing scheme.