线路重叠交叉段下部隧道联络通道冻结法施工对上部隧道的影响

以某地铁线路隧道重叠交叉段联络通道施工为工程背景,针对下部隧道联络通道冻结法施工,建立Midas/GTS建立三维有限元模型,分析联络通道冻结法施工过程对上部隧道的影响。分析结果表明:下部隧道联络通道施工完成后,上部隧道最大沉降量为-0.322 mm,最大隆起量为0.211 mm,最大水平位移为-0.053 mm,均在安全可控范围内;上部隧道结构最大拉压应力也均满足强度要求。     

基于深度学习的高铁接触网顶紧螺栓的缺陷检测

针对高铁接触网顶紧螺栓目标小、缺陷检测困难的问题,提出一种基于深度学习的顶紧螺栓缺陷检测方法。根据4C检测车拍摄的高铁接触网图片(大小为6600*4400 pixels),首先对特征信息更多的斜撑套筒进行定位,采用TDM模块与SSD相结合的算法提升算法对小目标的检测精度,并通过改变默认框的尺寸以得到更好的检测精度和速度;然后利用DeepLab v3 plus算法对顶紧螺栓部分进行语义分割;最后提出一种阈值法对顶紧螺栓的缺陷情况进行判别。为满足实际工程的速度需求,对训练好的模型进行优化。实验结果表明:相较于经典的SSD,本文改进的SSD方法对斜撑套筒的定位精度和速度都有提升;对于6600*4400 pixels的原始样本,本文提出的顶紧螺栓缺陷检测方法精度上达到95.9%,速度达到17.9 fps。     

Identification of communities in urban mobility networks using multi-layer graphs of network traffic

This paper proposes a novel approach to identify the pockets of activity or the community structure in a city network using multi-layer graphs that represent the movement of disparate entities (i.e. private cars, buses and passengers) in the network. First, we process the trip data corresponding to each entity through a Voronoi segmentation procedure which provides a natural null model to compare multiple layers in a real world network. Second, given nodes that represent Voronoi cells and link weights that define the strength of connection between them, we apply a community detection algorithm and partition the network into smaller areas independently at each layer. The partitioning algorithm returns geographically well connected regions in all layers and reveal significant characteristics underlying the spatial structure of our city. Third, we test an algorithm that reveals the unified community structure of multi-layer networks, which are combinations of single-layer networks coupled through links between each node in one network layer to itself in other layers. This approach allows us to directly compare the resulting communities in multiple layers where connection types are categorically different.     

Soft Radial Basis Cellular Neural Network (SRB-CNN) based robust low-cost truck detection using a single presence detection sensor

This paper does present a comprehensive concept for a robust and reliable truck detection involving solely one single presence sensor (e.g. an inductive loop, but also any other presence sensor) at a signalized traffic junction. Hereby, two operations modes are distinguished: (a) during green traffic light phases, and (b) a much challenging case, during red traffic light phases. First, it is shown how difficult the underlying classification task is, this mainly due to strongly overlapped classes, which cannot be easily separated by simple hyper-planes. Then, a novel soft radial basis cellular neural/nonlinear network (SRB-CNN) based concept is developed, validated and extensively benchmarked with a selection of the best representatives of the current related state-of-the-art classification concepts (namely the following: support vector machines with radial basis function, artificial neural network, naive Bayes, and decision trees). For benchmarking purposes, all selected competing classifiers do use the same features and the superiority of the novel CNN based classifier is thereby underscored, as it strongly outperforms the other ones. This novel SRB-CNN based concept does satisfactorily fulfill the hard industrial requirements regarding robustness, low-cost, high processing speed, low memory consumption, and the capability to be deployed in low cost embedded systems.     

铁路道岔防跳结构优化建议

为提升铁路道岔防跳结构的可维护性、完善检测方法和标准,提出以下优化方案:(1)改变防跳顶铁顶头截面形状,减小修配面积;(2)改变防跳限位装置螺栓头部厚度,选择更高强度的材质,取消开口销安装孔;(3)增加防跳间隔铁安装面内侧避让空间。通过对比分析或受力仿真分析,确认优化方案的可行性,提升道岔防跳结构零部件的维修便利性的同时,增强其防跳功能,提高行车安全性。最后,给出防跳间隙设置的建议,并提出防跳卡铁的检测方法,为铁路道岔防跳结构设计、检测提供参考。     

地铁车辆段轮对在线检测设备对出入段能力的影响分析

轮对在线检测设备在车辆段咽喉区的设置方式不同,引起信号转换轨位置的变化,从而影响列车出入段能力。为研究设备的合理设置方式,减少对列车出入段能力的影响,通过分解列车出段作业过程,计算出理想条件下的列车出段时间,再以几种典型的设备设置方式为例,分别计算出段作业时间差,对比分析后提出,在同等条件下,设备的最优位置应设于咽喉区第一付道岔和信号转换段之间,同时应尽量避免将设备设置在咽喉区第一付道岔内的岔线区。     

An inference engine for smartphones to preprocess data and detect stationary and transportation modes

A smartphone can be utilized as a cost-effective device for the purposes of intelligent transportation system. To detect the movement and the stationary statuses in the motorized and non-motorized modes, this study develops a new inference engine, including two sets of rules. The first sets of rules are defined by the related thresholds on the features of smartphone sensors while the second sets are extracted from the human knowledge to improve the results of the first rules. The experimental results reveal that by utilizing Inertial Measurement Unit (IMU) sensors in the proposed inference engine, it is possible to save 40% energy in comparison with the previous research. Moreover, this engine increases the accuracy of the motorized mode detection to 95.2% and determines the stationary states in motorized mode with 97.1% accuracy.     

Assessing carbon emissions from road transport through traffic flow estimators

Carbon emissions from road transport are one of the main issues related to modern transport planning. To address them adequately, the acquisition of reliable data about traffic flow is an essential prerequisite. However, the large quantity and the heterogeneity of available information often cause problems; missing or incomplete data are one of the most critical aspects. This paper discusses how technology handles imperfect information in order to obtain more accurate quantification of CO₂ emissions. First, an analysis of single estimators and combination models is provided, highlighting their main characteristics. Then, the TANINO model (Tool for the Analysis of Non-conservative Carbon Emissions In TraNspOrt) is presented, jointly developed at the University of Seville and at the IUAV University of Venice. It consists of two different modules: the first is a combination model that optimizes the results of three traffic flow single estimators, while the second is a macro-model of carbon evaluation, which takes into account road infrastructure, vehicle type and traffic conditions. TANINO is then tested to calculate CO₂ emissions along the ring road of the Spanish city of Seville, showing its more efficient performance, compared to the single estimators normally adopted for such aims. Transport planning can benefit from the adequate knowledge of traffic flows and related CO₂ emissions, since it allows a more reliable monitoring of the progresses granted by specific carbon policies.     

基于时间卷积自编码网络的实时交通事件自动检测方法

为了提高道路异常交通事件检测效率并降低误报率,提出了一种基于时间卷积自编码网络的实时交通事件自动检测方法。首先设计了基于波动相似性度量的交通模式搜索算法用来筛选具有相同交通规律的样本数据;并构造了交通流模式矩阵作为网络模型输入,以避免样本不均衡与单一样本数据随机性对交通模式学习的干扰;同时设计了新的时间卷积自编码网络对交通模式特征进行无监督提取并对未来交通参数进行合理预测;为了降低交通流参数随机波动性带来的事件判别的干扰,设计了异常状态评估方法,通过对模型预测误差分布的学习,结合当前检测数据给出最终的事件判定结果。采用美国西雅图I90公路与I405公路2015年全年的交通流检测数据与历史事故数据进行实证研究,并与6种典型交通事件检测算法进行性能对比。研究结果表明:基于时间卷积自编码网络的实时交通事件自动检测算法具有较高的检测率、较低的误报率以及更快的平均检测时间;综合各种交通运行情况下,可接受误检率分别为5%、10%时,平均检测率可分别达到93%、98%;同时算法能够自适应学习交通状态的动态变化,对不同交通运行环境具有较强适应性与稳定性。     

基于ADS1202单数据链路的变流器电气量检测

针对多相大功率变流器需要采集并处理的温度、电流及电压等电气量数量多、空间分布分散的特点,本文设计了一种基于ADS1202的单数据链路的电气量检测方法。采用曼彻斯特码完成单数据链路信号传输,通过FPGA编程实现曼彻斯特码自同步解码,并使用Sink3数字滤波器对电气量进行处理。实验验证了信号调理、转换和后处理的有效性和可靠性,并实际应用于多相大功率变流器装置中电气量的分布检测。