Estimating the influence of common disruptions on urban rail transit networks

With the continuous expansion of urban rapid transit networks, disruptive incidents—such as station closures, train delays, and mechanical problems—have become more common, causing such problems as threats to passenger safety, delays in service, and so on. More importantly, these disruptions often have ripple effects that spread to other stations and lines. In order to provide better management and plan for emergencies, it has become important to identify such disruptions and evaluate their influence on travel times and delays. This paper proposes a novel approach to achieve these goals. It employs the tap-in and tap-out data on the distribution of passengers from smart cards collected by automated fare collection (AFC) facilities as well as past disruptions within networks. Three characteristic types of abnormal passenger flow are divided and analyzed, comprising (1) “missed” passengers who have left the system, (2) passengers who took detours, and (3) passengers who were delayed but continued their journeys. In addition, the suggested computing method, serving to estimate total delay times, was used to manage these disruptions. Finally, a real-world case study based on the Beijing metro network with the real tap-in and tap-out passenger data is presented.     

How the environmental impact affects the design of logistics networks based on cost minimization

The degree of centralization of the logistics network is one of the important strategic decisions to take. This issue has often been studied but mainly from a cost minimization focus. This research studies the influence of incorporating the environmental impact in logistics network design models, comparing the results with those obtained considering cost minimization only. A fuzzy biobjective optimization model is used to carry out a number of experiments. By weighting the relative importance of the two objectives, the differences with respect to the base cost minimization solution can be ascertained. Our results show that in general more material is moved through warehouses when the environmental impact is considered and more decentralized facilities are opened.     

A hybrid deep learning based traffic flow prediction method and its understanding

Deep neural networks (DNNs) have recently demonstrated the capability to predict traffic flow with big data. While existing DNN models can provide better performance than shallow models, it is still an open issue of making full use of spatial-temporal characteristics of the traffic flow to improve their performance. In addition, our understanding of them on traffic data remains limited. This paper proposes a DNN based traffic flow prediction model (DNN-BTF) to improve the prediction accuracy. The DNN-BTF model makes full use of weekly/daily periodicity and spatial-temporal characteristics of traffic flow. Inspired by recent work in machine learning, an attention based model was introduced that automatically learns to determine the importance of past traffic flow. The convolutional neural network was also used to mine the spatial features and the recurrent neural network to mine the temporal features of traffic flow. We also showed through visualization how DNN-BTF model understands traffic flow data and presents a challenge to conventional thinking about neural networks in the transportation field that neural networks is purely a “black-box” model. Data from open-access database PeMS was used to validate the proposed DNN-BTF model on a long-term horizon prediction task. Experimental results demonstrated that our method outperforms the state-of-the-art approaches.     

基于BP神经网络的列车制动控制仿真研究

针对列车制动过程存在的复杂性、非线性、时变性、不确定性等因素,通过分析影响建立BP神经网络模型的主要因素,建立了用于列车制动控制的BP神经网络模型。以货物列车为仿真对象,在Matlab环境中进行了仿真研究。仿真结果表明,该方法控制安全性好、停车误差小,基于BP神经网络的智能算法运用于列车制动控制是可行的。     

城市轨道交通各发展阶段的运行特征及在公交系统中的作用

城市轨道交通网络的形成分为三个发展阶段:成网前阶段,轨道骨架网阶段和稳定的轨道网络阶段.分析了不同发展阶段轨道交通运行特征指标的取值范围,采用客运量和客运周转量指标研究了轨道交通不同发展阶段在公共交通系统中所起的作用.     

浅谈铁路应急通信系统的接入技术

介绍了用于我国铁路应急通信系统的各种接入技术，并对各种接入技术的使用条件提出了一些建议。     

复突触神经网络模型及其稳定性研究

传统的Hopfield网络在联想记忆、优化计算方面发挥了重要作用,但该网络存在一个弊端,即它只能处理线性或二次的最优化问题。因此,本文首先对Hopfield神经网络进行了扩展,将其称为复突触神经网络(或模型),扩展的网络不仅可以解决目标函数或约束中含有高阶的最优化问题,而且还可以处理目标函数含有对数、三角形式等较复杂函数的最优化问题;然后给出了该网络稳定的充分条件并给出了证明,同时对网络中的权矩阵是否对称也进行了讨论;最后结合典型的模糊c均值聚类问题验证了该模型的有效性。     

水网地区航道网技术评价指标与方法研究

在对过去区域航道网技术评价体系进行初步探讨的基础上,结合多个市域航道网规划实践,基于对区域航道网规划特点及技术评价特性分析,借鉴公路的成熟做法,提出区域航道网技术评价指标体系及方法,并以苏州、无锡、常州市航道作为水网地区实例,进行江苏省水网地区规划航道网技术评价与分析比较,进而提出各技术评价指标建议值。     

基于TCP/IP拥塞控制的算法研究

在WPAN上使用TCP的关键在于拥塞控制，根据TCP的拥塞控制机制，通过分析拥塞控制的闭环控制原理、TCP／IP拥塞控制所使用的典型技术，研究TCP／IP拥塞控制在无线网上应用的算法。