基于全U网络的混凝土路面裂缝检测算法

针对现有的混凝土裂缝检测算法在各种复杂环境中检测精度不够、鲁棒性不强的问题,根据深度学习理论和U-net模型,提出一种全U型网络的裂缝检测算法。首先,依照原U-net模型路线构建网络;然后,在每个池化层后都进行一次上采样,恢复其在池化层之前的特征图规格,并将其与池化之前的卷积层进行融合,将融合之后的特征图作为新的融合层与原U-net网络上采样之后的网络层进行融合;最后,为了验证算法的有效性,在测试集中进行实验。结果表明,所提算法的平均精确率可达到83.48%,召回率为85.08%,F1为84.11%,相较于原U-net分别提升了1.48%,4.68%和3.29%,在复杂环境中也能提取完整裂缝,保证了裂缝检测的鲁棒性。     

多运营商高铁隧道移动网络覆盖方案

高铁移动网络覆盖是国内三大通信运营商的一个重点,而高铁隧道内移动网络覆盖更是运营商的一大难点痛点。文章根据我国中部省份某高铁线路覆盖规划实例,采用“设备+POI+泄漏电缆”模式,即三家运营商信号源设备通过同一POI(point of interface,多系统接入平台)接入,信号输出到泄漏电缆进行隧道覆盖,隧道口场坪站安装宽频切换天线对隧道外进行延伸覆盖,通过链路预算合理布置各运营商主设备信号源,从而实现隧道到室外的无缝覆盖。最后,根据已有成熟网络覆盖解决方案,对未来5G高铁隧道移动网络覆盖方案进行了探讨。     

伊朗德黑兰地铁轨道施工方法

依托伊朗德黑兰地铁一、二号线工程 ,针对地铁单洞单线采用的轨道施工轨排法存在的不足 ,根据单洞双线隧道的特点 ,提出了利用组装好的轨排直接铺设在隧道结构底板上作为轨道平板车临时轨道的施工新方法     

液晶显示驱动电路中Fibonacci型电荷泵单元

本文中所设计的电荷泵将用于手机液晶显示驱动模块中.通过对Fibonacci型电荷泵上升时间的估算,对减小上升时间和动态功耗进行折中考虑,本文提出优化开关频率的方法.用1.2μm CMOS双阱工艺参数对所设计的电荷泵进行模拟,结果表明这个电荷泵具有较快的上升速度和较高的效率.通过提高V_GS电平和保证开关管的衬底始终接在最高电位上,文中提出了一种新型Fibonacci电荷泵,它可以正常工作在从1.2V到5V变化的多种电源电压下.     

Homogeneity characterisation of sintered (U,Gd)O2 pellets by X-ray diffraction

(U,Gd)O₂ sintered pellets are fabricated by different methods. The homogeneity characterisation of the Gd content seems to be necessary for a production control to qualify the process and the final product obtained. In this paper, we propose an analysis of the X-ray diffraction powder patterns through the Rietveld method, in which the differences between the experimental and the calculated data proposed from a crystalline structure model are evaluated. This result allows us to determine the cell parameters, that can be correlated with the Gd concentration, and the existence of other phases with different Gd contents.     

可持续发展战略指导下的轨道交通规划与评价

提出了轨道交通规划与评价在实施可持续发展战略过程中需要研究的问题，研究重点是轨道交通规划与评价方法体系和环境对策，研究了轨道交通规划目标的变化、轨道交通规划优化决策理论与方法、轨道交通规划中的环境影响评价、城市轨道交通网络的合理规划、城际轨道交通——客运专线/高速铁路线路基本走向的规划、轨道交通规划可持续发展的评价指标体系概念框架、轨道交通系统可持续发展的能力建设问题等，并通过实例验证了所建立的理论方法体系和模型体系；不仅进行了有关理论和方法的探讨，而且以面向实际应用为目标，开发了相应的软件系统，为轨道交     

Determination of the uranium enrichment with the NaIGEM code

The traditional method used to non-destructively determine the uranium enrichment with an NaI detector is based on the “enrichment meter principle” (Progress report LA-4605-MS, Los Alamos National Laboratory, NNM, 1970, p. 19), which involves measuring the intensity of the 186 keV line of ²³⁵U by selecting two regions of interest for the peak and the background. This type of method suffers from several limitations, the most limiting of which are the impossibility to make wall thickness correction or to take the inference of foreign radioisotopes into account. The NaIGEM software (A guide for using NaIGEM code, version 1.5 for DOS and Windows, 2001; Nucl. Instr. and Meth. A 458 (2001) 196) was developed to overcome these limitations by calculating the 186 keV line intensity with a fitting procedure. The code was tested in different measurement conditions on the wide variety of certified samples, in particular, on reprocessed uranium and on depleted material with thick steel filters interposed between the source and the detector. The results are presented to illustrate the performance and limitations of the tested version (A guide for using NaIGEM code, version 1.5 for DOS and Windows, 2001). The general performance is good except in the case of low-enriched uranium in thick containers.     

铁路桥梁横向振动模拟及适应快速行车结构型式的研究

针对铁路提速后出现的上承式简支钢板梁桥跨中横向振幅过大的问题,以羊楼司桥(L=32m单跨上承式简支钢板梁桥)为例进行实桥检测和数值模拟,建立车辆、线路、桥梁三者的联合模型,考虑轮轨接触力,应用MSC/DYTRAN大型结构分析程序进行车线桥动力相互作用的仿真分析。研究揭示我国用量最多的C62型货车以70∽80km/h的速度空载通过时,该跨度的钢板梁桥发生横向谐振的机理。此外,用该方法对32m跨长的组合梁结构进行桥梁动力特性及列车走行性分析,表明该结构满足提速货车和高速电动旅客列车走行的要求。研究结果对我国高速铁路桥梁建设和既有桥梁的改造具有积极意义。     

快堆伪裂变产物数据的计算

从微观评价核数据库出发 ,计算给出了快堆谱的伪裂变产物数据 ,包括一群的截面数据和多群的全套中子数据 ,裂变核有2 3 5U和2 3 9Pu。截面数据有全截面、弹性散射、非弹散射、(n ,γ)、(n ,2n)截面 ,多群全套中子数据还包括次级中子角分布和能谱。数据以ENDF B -6格式给出 ,同时还分析讨论了数据误差和可靠性 ,数据可用于实际的快堆计算     

Application of Code Approach for Emergency Evacuation in a Rail Station

This paper presents an emergency evacuation analysis during a mid-platform train fire in an underground rail station. A time-based evacuation scenario is developed taking into account the stairway inaccessibility due to smoke blockage. The calculation approach in NFPA 130 is then applied to a parametric study of this time-based scenario and its reliability is examined using a micro-simulation model. Both methods give consistent predictions of the exiting times. The results demonstrate that with the division of the station into two zones, the exiting time could not meet the code requirements, whereas it does meet the required exiting time without the division. Such division is due to smoke blockage and creates an uneven ratio of the occupant load to the available exiting capacity. This shows that appropriate consideration of the fire and smoke effects is important for station egress evaluation. Other issues such as the fire growth rate and the pre-evacuation time are also discussed.