基于遗传算法的铁路纵断面优化系统地研究.pdfVIP

第18卷第10期 铁路 计 算 机 应 用 研究与开发 Vol.18 No.10 RAILWAYCOMPUTERAPPLICATION RESEARCH AND DEVELOPMENT 文章编号：1005-8451（2009）10-0001-05 基于遗传算法的铁路纵断面优化系统的研究 谢春玲，蒲 浩 （中南大学 土木建筑学院，长沙 410075） 摘 要：提出一种基于全局最优的改进式遗传算法。采用改进的启发式交叉算子和多种算子混合操作， 加快算法向最优解收敛的速度。给出相应的纵断面优化模型，并改进初始种群生成的方法，将一个二维优 化问题转化为两个一维优化问题进行求解，简化了算法的执行过程。引入惩罚函数的概念处理方案的约束 问题，设计算法实现的流程图，并开发相应的优化系统。利用实例证明遗传算法在纵断面优化设计中的可 行性及全局寻优的性能。 关键词：铁路；纵断面优化；遗传算法；优化系统 中图分类号：U212∶34-TP39文献标识码：A Study on Railway Profile Optimizing System based on genetic algorithms XIE Chun-ling, PU Hao (Central South University, Civil Engineering, Changsha 4 10075, China) Abstract: It was put forward the modified genetic algorithm . A mixed operation between improved heuristic crossover operator and multi-operator was adopted to accelerate the convergence of the algorithm to the optimal solution . The corresponding model of vertical section optimization was put forward. The method to improve the method of generating initial population, by transforming a two-dimensional optimization problem into two one-dimensional optimization problems, was also improved. The constraint problem of program was dealt with by utilizing the concept of penalty function in this paper. A flow chart of the algorithm was designed. The Corresponding Optimization System was developed as well. The example adopted in this paper approved the feasibility and the performance of global optimization of genetic algorithms to optimize the design of vertical section. Key words: railway; profile optimization; genetic algorithm; Optimization Sy stem 铁路纵断面优化设计是一个反复迭代、不断 1 纵断面优化的数学模型 改善、逐步逼近最佳方案的过程。目前国内传统的 纵断面优化设计一般都是由人工给出或由计算机 1.1设计变量 自动生成一个