自适应遗传算法支持下的既有铁路曲线整正方法  

Realignment method for existing railway curve based on adaptive genetic algorithm

作　　者：明祖涛 李颖喆 李玮 赖祖龙 MING Zutao;LI Yingzhe;LI Wei;LAI Zulong(School of Geography and Information Engineering,China University of Geosciences,Wuhan 430074,China;Wuhan Geotechnical Engineering and Surveying Co.,Ltd.,Wuhan 430022,China)

机构地区：[1]中国地质大学(武汉)地理与信息工程学院,武汉430074 [2]武汉市勘察设计有限公司,武汉430022

出　　处：《测绘工程》2023年第1期37-41,共5页Engineering of Surveying and Mapping

基　　金：国家自然科学基金资助项目(41804033);中央高校基本科研业务费专项资金(CUGL180831)。

摘　　要：针对拟合与最优化方法在铁路曲线整正问题中难以加入约束条件与计算精确里程的不足,提出一种使用自适应遗传算法求拨距总量最小值的方法。首先利用遗传算法限制性低的特性,在个体的并行计算中加入多种逻辑、比较运算,从而能够顾及各种约束条件,用黄金分割法搜索测点的拨距量和精确里程,最后累加测点拨距量并计算所有个体的适应度。种群繁衍过程中采用保留父代最优个体与自适应的交叉变异算子的改进策略,使种群在最优方向持续进化。通过大量仿真实验和实际工程数据验证,该方法具有较高的精确性和适用性。The application of some fitting and optimization methods to the railway curve realignment problem makes it difficult to incorporate constraints and to precisely calculate the mileage. Thus, the adaptive genetic algorithm is proposed to find the minimum cumulative move distance. First, the low restrictiveness of the genetic algorithm is exploited by adding logic and comparison operations to individual parallel computing to incorporate constraints. Then, the golden section method is used to search for the move distance and determine the precise mileage of measurement points. Finally, the cumulative move distance of the measurement points is determined, and the fitness of all the individuals is calculated. The improved strategy is applied to population reproduction by retaining the optimal individual of the parents and the adaptive cross and mutation operator to ensure that the population evolves in the optimal direction. The results of a large number of simulation experiments using actual engineering data verify the high accuracy and applicability of the proposed method.

关 键 词：铁路曲线整正 坐标法拨距计算 自适应遗传算法 最优化方法 网格法 

分 类 号：P258[天文地球—测绘科学与技术]