摘要
公路路线设计是一个多目标的问题,传统将多目标转换为单目标优化方法存在目标统一准确性差的缺点,考虑多目标间的冲突性,提出采用多目标遗传算法进行路线优化。主要思想是通过建立多目标模型和路线线形模型,进行平面线形设计,经过纵断面优化后计算各目标函数值。采用快速非支配排序法对目标值进行排序,满足分布性的要求下构造非支配集,反复迭代计算后,输出满足终止条件的非劣解集。利用此算法进行实例验证,结果表明能在兼顾安全、经济、舒适、环保的目标下,给出一序列最优解,为路线方案决策提供直接依据。
Highway alignment design belongs to multi-objective problems.The accuracy is poor for unify the objective with the traditional optimization methods that multi-objective is converted to single-objective.Considering the many conflicts between the objectives,the optimize method of multi-objective genetic algorithm is advanced in this text.The main idea is that the multi-objective model and the route model for the design of horizontal alignment are established firstly.Vertical alignment is optimized and then the objective function values are calculated.The values are sorted by quick non-dominated sorting method.The non-dominated set is structured under the requirement of distribution.And then iterative calculations are repeated.A pareto optimal solution set is output to meet the terms for terminating.This algorithm is verified by the examples.Results show that the set of optimal solutions are presented considering the multi-objective of security,economic,comfortable and environmental protection.The solutions provide a basis for highway route decision-making.
出处
《公路》
北大核心
2012年第10期101-107,共7页
Highway
基金
山东省交通科技项目
项目编号2009Y019
关键词
公路路线
多目标遗传算法
目标函数
非劣解
非支配集
highway alignment
multi-objective genetic algorithm
objective function
pareto optimal solution
non-dominated set
