基于Elman神经网络的公共自行车单站点需求预测

公共自行车租赁系统目前存在的主要问题是高峰时段用户存取车困难,站点需要工作人员值守。为了提高城市公共自行车调度的科学性、缩短调度时间、降低调度成本,避免调度过程中调度员凭经验确定各个站点需求量的盲目性,从而更好地服务租赁者,满足其出行需求。建立了基于改进的Elman神经网络的公共自行车单站点需求量预测模型。通过仿真实验将改进模型和已有模型的预测结果与实际需求量进行对比,证明了提出的预测方法有效可行。     

城市公共自行车租赁数据的可视分析方法

自行车已经成为广大民众公共出行的一种常见交通工具,杭州市公共自行车系统分布范围已覆盖了整个市区.研究公共自行车系统的可视化及分析方法,需要查询每个站点的实时租赁数据及历史统计,包含上周、昨天和今天的统计数据,用户可以选取一些站点,在同一个视图内显示它们的统计数据,通过交互分析,预测每个站点的借用趋势,帮助管理人员充分理解和分析这些数据,并了解这个区域内的整体趋势.     

带有不确定性的公共租赁自行车的管理优化研究

研究城市公共租赁自行车的管理优化问题,将其分成两个子问题,即公共租赁自行车停放点的布局优化问题和公共租赁自行车的调度优化问题.设置合适的公共租赁自行车停放点的目的,一方面是方便管理人员的管理,另一方面是方便公众的出行.公共租赁自行车的调度需要保证调度过程中产生的总费用最少.在公共租赁自行车的实际管理中,每个停放点的需求量和调度车的行驶时间是不确定的,因此引入两个不确定变量,建立不确定0-1规划模型和不确定整数规划模型,并利用不确定理论将两个不确定性模型分别转化为确定性等价类模型.通过一个数值实验对所提出的模型进行验证,根据第1个优化模型求解得到公共租赁自行车的最优停放点,并基于最优停放点根据第2个优化模型求解生成公共租赁自行车的最优调度方案.     

成本约束下多机场群航班频率优化模型仿真北大核心

采用目前方法对多机场群航班频率进行优化时,没有考虑旅客的出行成本,对航班频率优化后存在延误时间长、延误航班数量多、出行成本高和日利用率低的问题。提出考虑出行成本的多机场群航班频率优化模型构建方法,在构建航班频率优化模型之前,对旅客的出行方式进行博弈分析,在Stackelberg博弈模型的基础上构建双层博弈模型,上层模型实现社会效益最大化,下层模型实现旅客出行费用最小化,完成多机场群航班频率优化模型的构建,并采用遗传算法对多机场群航班频率优化模型求解,完成航班频率的优化。仿真结果表明,所提方法的延误时间短、延误航班数量少、出行成本低、日利用率低。     

DES算法在自行车租赁系统中的应用

公共自行车租赁系统已逐步纳入城市公共交通体系,以解决公交出行“最后一公里”问题。本文针对自行车租赁系统中的安全认证问题,提出用软件模块实现DES算法代替传统所使用的PSAM(Purchase Secure Access Module)硬件卡方案,分别在自行车租赁系统内的售卡系统、消费终端和车辆信息几个模块中,实现DES加密。不仅提高现有系统智能卡的扩展性和安全性,而且减少硬件成本,提高了系统的效率。     

公共自行车系统的租赁点聚类与功能识别

租赁点功能识别对公共自行车系统的平衡调度和布局规划具有重要作用,而系统中积累的大量历史数据可反映用户在不同时间和地点的社会经济活动,并且与租赁点功能紧密联系。为此,对系统历史数据进行分析,构建公共自行车系统租赁点聚类模型。结合租赁点的时空属性,利用潜在狄利克雷分布模型挖掘租赁点的功能特征,使用K-means聚类算法进行特征聚类。通过集群模式特征分析并使用兴趣点数据和租赁点名称信息对聚类结果进行验证,结果表明,该模型可以有效地辅助系统管理者掌握公共自行车系统租赁点的功能分布。     

基于Anylogic的轨道交通换乘能力改进的研究

研究轨道交通换乘枢纽与市内其它交通方式的衔接与协调,是城市公共交通线网优化的主要内容之一。它能减少出行过程中的等待时间,缩短人们出行时间,提高公交服务质量,并保证客运交通的高效率,也能更好地促进城市轨道交通与其它交通方式的协调发展。以武汉地铁某换乘站研究对象,通过实地采集其基本设施的位置和数量、入站行人和出站行人的数量,并利用Anylogic仿真软件,建立站点的平面布置图空间模型和行人流程图。首先通过仿真软件仿真找出换乘客流瓶颈部位,其次分别对其提出改善建议并加以修改仿真模型。最终仿真结果表明,改善建议使得换乘站的换乘效率都得到提高。     

基于网络模型的城市公共自行车需求量预测研究

自行车共享系统逐渐出现在许多城市中,由于在不同时间和站点的自行车需求量(租/还量)的不平衡,系统中各站点的自行车需要人工频繁地使其不断达到平衡状态,然而实时监控并不能很好的解决这个问题。因此,提出了一个基于网络图的预测模型,可以预测未来时间段内的某个站点自行车的需求量,提前对自行车的重新分配。首先,我们通过分层聚类算法对预测站点进行聚类,得到与其相关的站点簇。其次,对站点簇构建网络模型。最后,使用纽约和华盛顿两个自行车共享系统的数据进行实验和结果分析。结果发现同一簇的站点具有相似的使用模式,模型预测误差率不高于0.45,且能够应用于不同城市的自行车共享系统。     

基于OD数据的群体行为可视分析

针对已有的公共自行车群体租车行为研究仅在站点尺度上进行分析,存在分析不完备等问题,提出了从城市—区域—站点的多尺度群体租车行为交互分析系统.首先,基于租/还站点和时间的一致性,从公共自行车数据中提取租车群体;然后结合站点的地理位置和群体租车行为的相似性,利用改进的迭代双聚类算法生成区域;再设计日历图、流量散点图和群体行为分布地图等视图,支持在宏观上分析群体行为随时间的变化和自流量与总流量的关系,在区域或站点尺度上分析群体行为的微观变化;最后结合多视图联动和交互分析,比较区域或站点的群体行为在工作日和周末的模式.通过4个案例,展示了群体租车行为的时空模式,证明了该系统在群体行为分析上的有用性和有效性.     

基于多Agent仿真的电动汽车分时租赁

“互联网$+$交通”模式产生出一系列新兴业态,对人们的城市交通出行方式带来颠覆性的改变,其中电动汽车分时租赁在互联网租车背景下得到新发展.鉴于此,从多Agent仿真的基本思路出发,以北京一嗨公司为例,针对电动汽车分时租赁系统中的主要对象租赁站、消费者和电动汽车车队进行交互模拟,通过分析车辆平均使用率、平均空闲率以及消费者满意度等对租赁商的运营能力进行评估,探讨车队规模和停车位数量对运营利润的影响.比较消费者还车和站点员工还车情景,同时在Anylogic仿真平台上实现该模型.     

A hub location inventory model for bicycle sharing system design: Formulation and solution

This study addresses a strategic design problem for bicycle sharing systems incorporating bicycle stock considerations. The problem is formulated as a hub location inventory model. The key design decisions considered are: the number and locations of bicycle stations in the system, the creation of bicycle lanes between bicycle stations, the selection of paths of users between origins and destinations, and the inventory levels of sharing bicycles to be held at the bicycle stations. The design decisions are made with consideration for both total cost and service levels (measured both by the availability rate for rental requests at the pick-up rental stations and coverage of the origins and destinations). The optimal design of this system requires an integrated view of the travel costs of users, bicycle inventory costs and facility costs of bicycle stations and bicycle lanes as well as service levels. The purpose of this study is to create a formal model that provides such an integrated view, and to develop methods for obtaining solutions for the design variables in practical situations. The complexity of the problem precludes the exact solution of the optimization problem for instances of realistic size, and so we propose a heuristic method for efficiently finding near-optimal solutions. In the test problem for which enumeration is possible, the heuristic solution is within 2% optimal. Finally, a numerical example is created to illustrate the model and proposed solution algorithm.     

Comparative analysis for periodical and random servicing systems considering different working circumstances: A textile application

As is in the textile industry, in some industries the selection of the operator servicing system to improve machines’ efficiencies is made intuitively or according to historical experiences generally. However, these preferences are not felicitous in any case. On the other hand, to estimate and analyze the machine efficiency in the case of multimachine assignment is a difficult issue due to machine interference problem. The solutions proposed such as the prepared tables, analytical formulas and queuing theory are generally for a limited number of machines and involve complex and tedious calculations and are generally insufficient for detailed analysis. In this paper, considering some working circumstances, the effect of random and periodical operator servicing systems on the machine efficiency is compared by using simulation methodology. Also, the algorithm of one of the simulation programs used is given. This study indicates that if the average stoppage rates of the machines assigned to the operator is small and the numbers of these machines are large then the random operator servicing system has superiority over the periodical operator servicing system.     

无线电监测站部署效率评估与优化方法研究

无线电监测网络规划中,以最少的监测站实现最大的监测区域或目标覆盖率,可达成频谱监测资源的最优化利用,提升无线电监测的全域化、精细化感知能力。建立监测传感器网络部署效率评估指标体系,以监测覆盖率和部署效率最大化为准则,通过空间二次聚类算法,得到监测站与地理位置的最优组合,达成部署优化的目的。仿真结果表明算法可获得更优化的部署方案且部署效率更高。     

TL-FCM: A hierarchical prediction model based on two-level fuzzy c-means clustering for bike-sharing system

In recent years, shared bikes have gradually emerged into public life as a new way to travel and helped solve the last-mile problem of residents’ travel. While this development has brought convenient travel to users, a series of problems exist, a prominent one is the uneven distribution of bikes at each shared bike station. Accurately predicting bike usage in a bike-sharing system can help solve this problem. In this paper, we investigate how to improve the accuracy of predicting the usage of bikes in bike-sharing system. First, considering both geographic location information of shared bike stations and the migration trend of bikes between stations, we design a two-level fuzzy c-means clustering algorithm to cluster shared bicycle stations into groups, which can better capture the connection between shared bicycle stations and improve the clustering accuracy of shared bicycle sites, then, we combine the two-level fuzzy c-means clustering algorithm with the multi-similarity reference model to predict the usage of bikes, which can significantly improve the accuracy of the forecast. To evaluate the performance of our model, we validate our model in the New York Bike-Sharing System. The results shows that our model obtained significantly better results than other models.

     

异构网中具有服务质量约束的高能效微基站部署方法

针对异构密集网络中基站密度增大带来的网络能耗过高问题,提出一种异构网络中高能效的微基站部署方法。首先,考虑微基站候选位置可行性以减轻环境条件的影响;其次,在不同的用户分布状态下对优化目依概率进行加权,增强了对不同用户分布场景的适应性;最后,通过对微基站部署数目、位置和功率配置的联合优化来提升系统能效,并提出了一种高能效的微基站部署算法。仿真表明,与仅优化微基站数量和部署位置的方法相比,所提方法提升能效最高达26%。实验结果表明,相对于不考虑功率的部署方法,所提出的联合优化方法能够有效提升系统能效,同时验证了微基站功率对异构网络能效的影响。     

GRASP with path relinking for the selective pickup and delivery problem

Bike sharing systems are very popular nowadays. One of the characteristics is that bikes are picked up from some surplus bike stations and transported to all deficit bike stations by a repositioning vehicle with limited capacity to satisfy the demand of deficit bike stations. Motivated by this real world bicycle repositioning problem, we study the selective pickup and delivery problem, where demand at every delivery node has to be satisfied by the supply collected from a subset of pickup nodes. The objective is to minimize the total travel cost incurred from visiting the nodes. We present a GRASP with path-relinking for solving the described problem. Experimental results show that this simple heuristic improves the existing results in the literature with an average improvement of 5.72% using small computing times. The proposed heuristic can contribute to the development of effective and efficient algorithms for real world bicycle reposition operations.     

A clustering deployment scheme for base stations and relay stations in multi-hop relay networks

Suitable deployment for the base stations and the relay stations is one of the most important features for the portable devices to obtain a high data transmission rate in multi-hop relay networks. In this paper, a novel clustering algorithm is proposed to select the appropriate deployment locations for the base stations and relay stations from the candidate positions by accounting for the traffic demands and using the uniform cluster concepts. Based on the different candidate positions, the proposed scheme makes an adaptive decision for selecting the deployment sites of the base stations and relay stations. A reasonable network throughput and coverage ratio can be obtained by balancing the network load among the clusters. Our simulation results indicate the superior performance of our proposed deployment scheme in achieving the appropriate performance for the network throughput and coverage ratio. Additionally, a suitable deployment budget can be utilized in multi-hop relay network systems.     

基于GIS的共享单车站点最优选择查询系统设计

随着智能手机的快速发展,将自行车系统加入到智能手机中就可以加快自行车的发展,提高自行车的利用率。在此基础上,利用地理信息系统更够开发出功能非常强大的软件。因为移动GIS是目前的发展趋势,在主流的移动操作系统上开发移动GIS能使软件更容易的让用户接受并使用,借助移动操作系统的界面友好性,使应用更加容易使用。本系统利用了ArcGIS for Android开发了自行车租赁系统的一个部分,自行车的站点查询。同时利用面向对象的程序设计方法进行开发移动GIS,系统具有良好的用户界面与灵活性,能够满足用户的需求。