燃料电池公共汽车示范运行评价分析

燃料电池汽车技术是国际上比较活跃的研究领域之一。在很多国家,如美国、日本和欧洲等,政府和企业都积极支持燃料电池汽车开发,相继开展了车辆示范项目。文中介绍了各国政府开展的燃料电池公共汽车示范项目和中国燃料电池公共汽车示范项目的车辆运行情况,并对运行进行了评价分析,这些结果可以为将来开展类似的示范项目提供一些经验。     

Design and implementation of a UPnP-can gateway for automotive environments

In this paper, we propose a universal plug and play (UPnP) — controller area network (CAN) gateway system using UPnP middleware for interoperability between external smart devices and an in-vehicle network. The proposed gateway consists of a UPnP communication device, a CAN communication device, and a device translator layer. In-vehicle devices are not usually IP-based, so we implemented an in-vehicle device manager in the UPnP communication device which is in the gateway. We developed a vehicle simulator to produce real vehicular data for performance analysis. The CAN communication device transmits and receives real-time vehicle data between the real vehicular simulator and external devices through the UPnP. The device translator layer configures a message frame for enabling seamless data input and output between the CAN and UPnP protocols. After implementation, we generated an internal-external service request and tested the result. Finally, we confirmed the service request and operation between external devices and the internal vehicular device. Additionally, for a variety of external device numbers and communication environments, we demonstrated the gateway performance by measuring the round trip time (RTT) for overall service implementation.     

Conceptual design of economic hybrid vehicle system using clutchless geared smart transmission

Most hybrid vehicles employ the continuously variable transmission (CVT or eCVT) currently as their choice of the transmissions. Recently, an automated geared transmission (AGT) or dual clutch transmission (DCT) is being tried for some hybrid vehicles for the better fuel economy than the CVT hybrid. However, this AGT or DCT is using automated clutches which require the hydraulic power in addition to the slippage in the clutch plate invoking some energy loss as well as wear. Also, they require a motor with significant power to match to the engine power. The clutchless geared smart transmission (CGST) has no clutch and the clutch function is performed by a planetary gear system controlled by a motor-generator. The hybrid vehicles proposed here using CGST may have some merits in durability, fuel efficiency, and cost since they do not have clutches. The motor used for the clutch function can be also working for power merge with the engine in propelling the vehicle. The proposed hybrid system can be either mild hybrid or full hybrid by adopting a different capacity of battery with much smaller motor-generator due to the planetary gear system compared to the other type hybrid vehicles. In this study, the prospects of newly proposed CGST hybrid system are examined in practical aspects compared with AGT hybrid or DCT hybrid systems.     

Operating speed pattern optimization of railway vehicles with differential evolution algorithm

The railway transportation system has much advantage in eco-friendliness, punctuality and safety compared to any other transportation system. Most of the railway system administrators have to control and operate under limited resources such as trains and facilities. It is necessary to control traveling time and energy consumption for efficient operation in the railway systems, because the board rate of passenger is inconstant with time variance. It is common that the shorter traveling time causes the greater energy consumption. In this study, a new optimization method considering operation time or energy consumption is proposed by using differential evolution algorithm and some cases are reviewed. The total energy change due to operation time variation are investigated by using the proposed optimization method for tangent and gradient track conditions. Both cases, the total energy decreases exponentially. However, because of gradient the total energy are saturated after a certain time for gradient track.     

基于3D CNN的道路视频交通状态自动识别

为了从视频直接有效地提取交通信息,提出了基于三维卷积神经网络 （3D convolutional neural networks,3D CNN）的交通状态识别方法.首先,以C3D （convolutional 3D）深度卷积网络为3D CNN原型,对卷积层数量与位置、平面卷积尺寸及三维卷积深度进行优化调整,形成了37个备选模型;其次,建立了视频数据集,对备选模型进行系统的训练测试,提出了交通状态识别模型C3D*;然后,对C3D* 和现有三维卷积网络模型进行视频交通状态识别测试分析;最后,对比测试了C3D* 及常用二维卷积网络的交通状态识别效果. 对比结果显示:针对视频交通状态识别,C3D* 的F均值为91.32%,比C3D、R3D （region convolutional 3D network）、R (2+1) D （resnets adopting 2D spatial convolution and a 1D temporal convolution）分别高12.24%、26.72%、28.02%;与LeNet、AlexNet、GoogleNet、VGG16的图像识别结果相比,C3D* 的F均值分别高32.61%、69.91%、50.11%、69.17%.      

90mm铸钢件GS-20Mn5+E级钢的焊接工艺

船舶承载能力的上升对船舶选用材料提出了更高的要求,焊接工艺不当将会导致焊接裂纹、未熔合等缺陷,直接影响船舶航行过程中的财产和人生安全。文章详细介绍了90mm厚的铸钢件GS-20Mn5与E级钢的焊接工艺。     

电力系统故障限流器的研究现状与应用前景

规模迅速发展的现代电力系统,正面临着短路故障电流随着系统容量的持续增加而不断增长的挑战。一种带故障限流器(FCL)的新型系统保护方案因其显著的优越性,正在逐步进入现代电力系统。本文介绍了目前几种最具发展潜力的FCL,总结了国内外的研究状况,最后对FCL的应用前景进行了展望。     

Full scale ice sea trials of Korean ice breaking research vessel ‘Araon’ on the big floes near Antarctica

A speed sea trial was performed in the Amundsen Sea in February and March of 2012. The Korean ice breaking research vessel “Araon” was used to check speed performance on the big floes. Two of ice trial cases were carried out. We describe the time history of location, the engine power, and the revolutions per minute during the ice trial, and the trajectory of the ship. Additionally, the measured ice properties are considered and discussed. The ice trial results were analyzed according to variation in ice thicknesses, ice strength, propulsion power, and the speed level of the ship. Our analysis results are compared to model test results and ice trial results. A correction to the target ice thickness was used to compare the power and speed relation in the same ice thickness because it is easy to know the relative speed performance of the ship. The Hamburgische Schiffbau-Versuchs Anstalt method was applied for the correction. The speed of Araon in big floes was higher than the speed in level ice. The speed after the correction at 10 MW of power and 103 cm of ice thickness was 5.4 knots based on analysis results.