摘要
由于在交叉口进行信号控制使得系统的总延误增加,所以信号控制的优化目标是使系统的总延误最小,同时要使得增加的延误在用户之间尽可能均衡分配。为解决这一问题,设计了一个平衡约束条件下信号控制与平衡网络设计组合问题的多目标规划模型,然后使用效用函数法将有双目标的上层优化问题转化为一个单目标优化问题,此时,该模型变成一个标准的双层规划模型,并利用全局优化方法模拟退火算法和惩罚函数法相结合求解。
Because optimal signal control can lead to an increase of delayed cost of system, the goal of signal control is to minimize the total delayed cost of system. At the same time, assignment of the increased delay should be balanced among users. In this paper, a multiobjective programming model under equity constraint that combines signal control and equilibrium network design is proposed and it uses a utility function method to transform the two objectives in the upper level optimization into a single optimization problem. In this case, the model becomes a standard bilevel programming model. A penalty function approach by embodying a global optimization method such as simulated annealing algorism is used to solve the model.
出处
《长沙交通学院学报》
2003年第3期73-78,共6页
Journal of Changsha Communications University
关键词
网络设计
信号控制
多目标规划
模拟退火算法
利益分配
效用函数
network design
singal control
multiobjective programming
simulated annealing
benefit balance
utility function