多目标演化算法在公交车辆发车间隔优化中的应用  

Application of multi-objective evolutionary algorithm in bus departing interval optimization

作　　者：曹莲英[1] 侯琳[2] 李文勇[2] 

机构地区：[1]山东交通学院交通与物流工程系,济南250023 [2]桂林电子科技大学机电工程学院,桂林541004

出　　处：《东南大学学报（自然科学版）》2009年第S1期260-265,共6页Journal of Southeast University：Natural Science Edition

基　　金：国家自然科学基金资助项目(50808050)

摘　　要：为了使公交车辆的发车间隔得到优化,根据客流量的变化,建立了以乘客和公交企业运营费用最小为目标的公交车辆发车间隔优化模型,并采用一种多目标演化算法(MOPEA)来求解模型.该算法通过粒子系统从非平衡状态达到平衡状态的理论来定义Rank函数,从而使得所有个体在每次迭代过程中均能参与杂交、变异等演化操作,最终求得发车间隔的全局最优解,从而避免传统演化算法中出现的陷入问题的局部解的现象.同时,保留了目标函数的多样性,使相向的多目标优化问题得到了一个"折中"的最优解,即Pareto最优解.最后通过实例验证了该算法比传统演化算法更具优越性.According to the alteration rule of the passenger movement,the optimal model of bus departing interval aiming at the minimal fee of the passengers and bus companies is built,and it is solved by a multi-objective optimization problems evolutionary algorithm(MOPEA).In the algorithm,the theory of particle system changing from non-equilibrium to equilibrium is used to define the Rank function,so all the individuals in the population have chance to participate in the evolving operation such as crossover and mutation to solve the global Pareto optimal solutions of bus departing interval optimization problems.This algorithm can avoid premature phenomenon of problems.And the diversity of objective functions is also reserved.The algorithm can gain compromise optimal results of conflicting multi-objective optimization problems—Pareto optimal front.Finally,an example is given and the result shows that the proposed algorithm has more advantages than traditional evolutionary algorithms.

关 键 词：公交 发车间隔 多目标演化算法 PARETO最优前沿 

分 类 号：U492[交通运输工程—交通运输规划与管理]