基于无人机数据和ADP算法的铁路线路多目标优化方法  

Multi-Objective Optimization Approach for Railway Alignment Based on Unmanned Aerial Vehicle Data and Approximate Dynamic Programming Algorithm

在线阅读下载全文

作  者:洪英杰 高岩 杨书生[1,2,3] 刘托 王平 何庆[1,2] HONG Yingjie;GAO Yan;YANG Shusheng;LIU Tuo;WANG Ping;HE Qing(School of Civil Engineering,Southwest Jiaotong University,Chengdu 610031,Sichuan,China;Key Laboratory of High-speed Railway Engineering,Ministry of Education,Southwest Jiaotong University,Chengdu 610031,Sichuan,China;Engineering Management Department,Jiqing High-Speed Railway Co.,Ltd.,Ji’nan 250013,Shandong,China)

机构地区:[1]西南交通大学土木工程学院,四川成都610031 [2]西南交通大学高速铁路线路工程教育部重点实验室,四川成都610031 [3]济青高速铁路有限公司工程管理部,山东济南250013

出  处:《铁道运输与经济》2025年第4期186-195,204,共11页Railway Transport and Economy

基  金:四川省自然科学基金创新研究群体项目(2023NSFSC1975);四川省青年基金项目(2025ZNSFC1317)-。

摘  要:铁路线路方案的规划与评价为多目标决策,影响工程经济、环境等多方面。为探讨铁路线路多目标优化方法,提出了基于工程造价、生态指标和碳排放的多目标线形优化方法。基于无人机采集的高精度地理信息数据,通过监督分类进行建(构)造物边界和生态特征的智能识别,建立包含周边复杂环境的耦合约束集。基于自适应动态规划(Approximate dynamic programming,ADP)算法,引入深度神经网络模型实现线形的智能精细化调整,运用帕累托(Pareto)最优原理处理不同目标之间的冲突关系,将帕累托最优解在三维空间中构建出来,给予决策者更多的决策空间。本方法在华东地区某高速铁路连接线项目中得到应用,结果表明:该方法较人工选线方案降低建设经济费用2.28%,生态优化和碳排放优化也分别达到2.67%和1.59%。该智能选线方法可以为设计人员提供不同优化目标的多种线路方案,实现铁路线路经济效益、环境影响的平衡。The planning and evaluation of railway alignment schemes is a multi-objective decisionmaking problem,affecting engineering economics and environments.To explore the multiobjective optimization approach for railway alignment,this paper introduced a multi-objective alignment optimization approach based on engineering costs,ecological indicators,and carbon emissions.According to the high-precision geographic information data collected by unmanned aerial vehicles(UAVs),the intelligent identification of construction boundaries and ecological features was carried out through supervised classification,and the coupled constraint set including the surrounding complex environment was established.Based on an approximate dynamic programming(ADP)algorithm,the deep neural network model was introduced to realize the intelligent and fine adjustment of the alignment.The Pareto optimality principle was utilized to manage conflicts between different objectives,constructing the Pareto optimal solution in three-dimensional space to provide more decision space for decision makers.This approach has been applied to a high speed railway connection project in East China,and the results demonstrate that compared with the artificial alignment selection scheme,this approach can reduce the economic cost of railway construction by 2.28%and achieve ecological optimization and carbon emission optimization of 2.67%and 1.59%,respectively.This intelligent railway alignment selection approach can provide designers with multiple alignment options tailored to different optimization objectives,achieving a balance between economic benefits and environmental impacts of railway alignment.

关 键 词:铁路选线 铁路线形优化 多目标动态规划 无人机数据 方案比选 

分 类 号:U212.3[交通运输工程—道路与铁道工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象