基于Q学习算法的摘挂列车调车作业计划优化

针对摘挂列车编组调车作业计划编制问题,基于强化学习技术和Q学习算法,提出1种调车作业计划优化方法。在表格调车法的基础上,将调车作业计划分为下落和重组2个部分。通过动作、状态和奖励3要素构建调车作业问题的强化学习模型,以调车机车为智能体,以车组下落的股道编号为动作,以待编车列的下落情况为状态,形成车组挂车、摘车具体条件和车辆重组流程,并依据车组下落的连接状态和车辆重组后产生的总调车程设计奖励函数。改进Q学习算法求解模型,以最小化调车程为目标,建立待编车列与最优调车作业计划之间的映射关系,智能体学习充分后即可求解得到最优的调车作业计划。通过3组算例对比验证本方法效果,结果表明:相较于统筹对口法和排序二叉树法,本方法使用的股道数量更少、调车作业计划更优;相较于分支定界法,本方法可在更短时间内求解质量近似的调车作业计划。因而,本方法有助于提高车站调车作业计划编制的智能化决策水平。     

How Do Governments Support the Development of Public Private Partnerships? Measuring and Comparing PPP Governmental Support in 20 European Countries

Abstract

Taking an institutional perspective, in this article we develop an index of the governmental support for public private partnership (PPP) — a ‘PPP Governmental Support Index’ (GSI) — which aims to measure the extent to which national governments provide an institutional framework that is either conducive or preventive for the introduction and diffusion of PPPs within transport infrastructure and other sectors. First, based on a substantive review of the literature, we define the elements of the PPP GSI, including the policy and political commitment regarding PPPs, the legal and regulatory framework, and the presence/absence of dedicated PPP-supporting arrangements. Second, we calculate the PPP GSI for 20 European countries, cluster them and compare similarities and differences in national governmental support of infrastructure PPPs. Third, we explore the potential link between national institutional index scores and infrastructure PPP activity in the 20 countries. Lastly, we discuss the potential and usefulness of the presented PPP GSI, as well as methodological limitations, and elaborate on how this index might be utilised to strengthen future comparative research on PPP in transport and other sectors.     

The role of volume-delay functions in forecasting and evaluating congestion charging schemes: the Stockholm case

This paper uses observations from before and during the Stockkholm congestion charging trial in order to validate and improve a transportation model for Stockholm. The model overestimates the impact of the charges on traffic volumes while at the same time it substantially underestimates the impact on travel times. These forecast errors lead to considerable underestimation of economic benefits which are dominated by travel time savings. The source of error lies in the static assignment that is used in the model. Making the volume-delay functions (VDFs) steeper only marginally improves the quality of forecast but strongly impacts the result of benefit calculations. We therefore conclude that the dynamic assignment is crucial for an informed decision on introducing measures aimed at relieving congestion. However, in the absence of such a calibrated dynamic model for a city, we recommend that at least a sensitivity analysis with respect to the slope of VDFs is performed.     

青藏铁路综合视频监控系统室外摄像机抖动频率特征分析

青藏铁路沿线多处区段常年多风,导致安装在铁塔高处的长焦摄像机产生抖动引起图像模糊,因此使用的长焦摄像机应具有防抖功能.而多个摄像机、镜头等生产厂家,均无法提供其电子防抖性能的具体参数指标,或无法给出光学/机械防抖性能参数指标来源.通过提出采用SURF算法提取匹配图像的特征点,介绍算法原理和算法实现过程,并开展算法验证和...     

山钢莱芜分公司专用铁路编组站改造方案研究

专用铁路编组站改造是山东钢铁股份有限公司莱芜分公司新旧动能转换项目的控制工程之一,对于铁路运输组织、站场设施管理工作的提升具有重要意义.在阐述既有莱钢编组站站场设施和运输组织现状的基础上,结合钢铁新旧动能转换工程规划,分析莱钢编组站目前存在的无贯通正线、到发线有效长不足、卸车线有效长不足、站坪坡度超限等问题,提出莱钢编...     

重载组合式列车卸车站平面布置方案研究

重载运输是现代货运发展的方向,卸车站作为重载铁路运输通道上的关键节点,对通道运输能力具有重要影响.从货物运输组织模式、站内主要技术作业内容和卸车站能力3个方面阐述重载组合式列车卸车站平面布置方案的影响因素,提出重载组合式列车车场布置形式,以及卸车站基本平面布置可以采用尽头式和环形式2种布置图型.结合实例,在调机、本务机...     

通苏嘉甬高速铁路跨杭州湾通道选择及桥位论证

通苏嘉甬高速铁路在嘉兴和宁波间需跨越杭州湾,杭州湾海域及两岸建设条件复杂、控制因素众多.开展跨杭州湾通道选择及桥位论证,对项目决策、建设和运营至关重要.在以往工程经验的基础上,从区域社会经济特征、区域交通现状及规划、气象水文地质等自然条件、港口航道和海事管控要求、秦山核电站及生态环境敏感区、防洪和涌潮影响规定等方面,详...     

大型铁路客运枢纽换乘服务研究

站在运输服务设计的角度,以增强旅客换乘出行体验为目的,提出“人、行李分离”的换乘服务理念,并基于该服务理念设计铁路客运枢纽内同站和异站换乘的方案。异站换乘方案设计时,提出铁路专用车的概念,打造一种全新的换乘模式,满足旅客换乘出行的多元化需求,吸引旅客主动换乘。最后,从心理和生理舒适性两个角度对换乘服务理念进行评价,结果表明“人、行李分离”服务创造了旅客换乘出行附加价值,提高了旅客的换乘出行体验感。     

铁路货运市场化定价机制探讨

为发挥价格政策对资源配置的引导和调节作用,以提高效益为目的,综合运用各种价格调节手段,构建合理的运价体系和结构,理顺不同环节间的价格关系,实现科学合理定价,在分析货运价格体系现状的基础上,提出市场化定价的基本原则和定价依据,从运作机制、定价关系、定价手段等方面对铁路货运市场化定价进行研究,制定实施符合铁路运输行业特征、...