石墨水泥砂浆注浆钢纤维混凝土板融雪试验

采用渗浇工艺制作了石墨水泥砂浆注浆钢纤维混凝土(GSIFCON)板.对该板的电热性能和冬季野外融雪性能进行了试验研究.结果表明:GSIFCON板通电产生的热量可有效融化积雪.在GSIFCON板的底面和侧面分别设置3 cm厚聚乙烯泡沫保温层时,其融雪热效率可达20%以上,较无保温层的GSIFCON板其融雪效率提高约5%.应用这种技术,可有效避免除冰盐对混凝土和环境带来的负面影响.     

有效提高轻钢结构承受雪荷载的措施

为了更有效提高轻钢结构对雪荷载的承受能力,通过对轻钢结构屋顶因雪压塌实例的原因分析,认为须从设计源头上抓好雪荷载取值、选择合理的屋面结构形式并加强施工和使用管理。     

薄壁型钢檩条抗积雪荷载作用的设计方法研究

近年来我国多次发生在积雪荷载作用下门式刚架轻钢房屋发生破坏事故。本文分析了积雪荷载在房屋屋面形成的特点和屋面积雪荷载作用下薄壁型钢檩条的破坏特征。探讨了现行设计规范和技术规程中檩条抗积雪荷载设计方法的合理性,指出了计算模型存在的不足。压型钢板屋面在积雪荷载作用下,不宜作为檩条可靠的侧向支撑,檩条计算应按屋面不能阻止檩条侧移和扭转进行强度和稳定计算。通过大量算例计算和分析比较,研究了屋面坡度与扭转翘曲正应力对檩条截面正应力的影响规律和程度,指出了现行设计规范和技术规程中薄壁型钢檩条抗积雪荷载的设计方法的适用性和缺陷。对于目前大量采用的屋面坡度较小的轻钢建筑檩条进行强度和稳定计算时,不宜忽略扭转翘曲正应力的影响。探讨了在屋面积雪荷载作用下门式刚架轻钢房屋发生破坏的原因。提出了薄壁型钢檩条抗积雪荷载作用的设计方法建议。     

南方部分500 kV主网架2008年冰雪灾害中受损分析与思考

2008年初,中国南方大部分地区出现强烈的低温雨雪凝冻天气,给电网的安全稳定运行和电力供应带来极大的影响和威胁。文中介绍了中国南方黔、桂、滇、粤等4省(自治区)电网受灾主要情况及特点,描述了其500kV主网架在此次冰雪灾害中受灾的主要过程,总结了电网主网架受灾的典型特征,然后从远距离大容量输电比例、电源规划与电网规划、500kV主网架输电走廊选取等方面分析了此次冰雪灾害中暴露出来的问题,并提出了500kV主网架在电源建设和电网结构等方面提高应对灾害能力的解决方案与建议。     

Approved algorithmic security enhancement of stream cipher for advanced mobile communications

ABSTRACT

Confidentiality and integrity processes are based on SNOW and ZUC algorithms. These standardized algorithms are designed by the 3rd Generation Partnership Project (3GPP) for advanced mobile communication systems. Each algorithm has two S-boxes in its nonlinear layer structure. This paper proposes three different approaches to enhance the security level of both algorithms. The first approach aims to select the best combination of two S-boxes in the nonlinear part of each algorithm. Results showed that the best randomness properties are achieved by combining the Feistel structure (S1-box) and Rijndael (S2-box) in the SNOW algorithm and New Rijndael (S1-box) and Rijndael (S2-box) in the ZUC algorithm. The second approach aims to increase the nonlinearity and complexity of these algorithms by selecting a strong S-box to cascade the best-existing two S-boxes for both algorithms. The third approach is an extension of the former, based on a dynamic reform of such S-boxes as a function of the “Authentication Token” to increase the computational complexity of the upgraded two algorithms. These proposed upgrades are simulated by C language and tested successfully for their statistical properties using the standard NIST SP 800-22 test suite.     

Use of low-temperature scanning electron microscopy to compare and characterize three classes of snow cover

This study, which uses low-temperature scanning electron microscopy (LTSEM), systematically sampled and characterized snow crystals that were collected from three unique classes of snow cover: prairie, taiga, and alpine. These classes, which were defined in previous field studies, result from exposure to unique climatic variables relating to wind, precipitation, and air temperature. Snow samples were taken at 10 cm depth intervals from the walls of freshly excavated snow pits. The depth of the snow pits for the prairie, taiga, and alpine covers were 28, 81, and 110 cm, respectively. Visual examination revealed that the prairie snow cover consisted of two distinct layers whereas the taiga and alpine covers had four distinct layers. Visual measurements were able to establish the range of crystal sizes that occurred in each layer, the temperature within the pit, and the snow density. The LTSEM observations revealed the detailed structures of the types of crystals that occurred in the snow covers, and documented the metamorphosis that transpired in the descending layers. Briefly, the top layers from two of the snow covers consisted of freshly fallen snow crystals that could be readily distinguished as plates and columns (prairie) or graupel (taiga). Alternatively, the top layer in the alpine cover consisted of older dendritic crystal fragments that had undergone early metamorphosis, that is, they had lost their sharp edges and had begun to show signs of joining or bonding with neighboring crystals. A unique layer, known as sun crust, was found in the prairie snow cover; however, successive samplings from all three snow covers showed similar stages of metamorphism that led to the formation of depth hoar crystals. These changes included the gradual development of large, three-dimensional crystals having clearly defined flat faces, sharp edges, internal depressions, and facets. The study, which indicates that LTSEM can be used to enhance visual data by systematically characterizing snow crystals that are collected at remote locations, is important for understanding the physics of snowpacks and the metamorphosis that leads to potential avalanche situations. In addition, the metamorphosis of snow crystals must be considered when microwave radiometry is used to estimate the snow water equivalent in the winter snowpack, because large snow crystals more effectively scatter passive microwave radiation than small crystals.     

基于密集连接空洞卷积神经网络的青藏地区云雪图像分类

为了提高高纬度地区云雪卫星图像的识别准确率,提出了密集连接空洞卷积神经网络与空洞卷积相结合的方法进行云雪卫星图像识别研究。该方法首先采用常规卷积层对图像进行处理得到特征图,然后采用多个密集块和过渡层对特征图进行处理。其中,密集块中采用跨层连接的方式实现了网络中所用层的特征传递,使得大量云雪特征得到重用,同时减轻了训练过程中的梯度消失问题。密集块中的卷积核采用空洞卷积,在减少参数量的同时扩大局部感受野,对云雪的光谱信息进行特征提取。最后,该方法采用平均全局池化层与全连接层得到云雪图像的预测结果。实验结果表明,与其他机器学习方法相比,该方法能够提高卫星云雪图像的识别准确率,具有良好的泛化能力。     

Complexity and trends analysis of hydrometeorological time series for a river streamflow: A case study of Songhua River Basin,China

In China's national economic growth, an important role is being played by the Songhua River because of the river's abundant resources and natural conditions. Therefore, study of hydrometeorological time series is very important to understand the basin behaviour. This research uses the snow cover data derived from MODIS, streamflow, and meteorological records in the Songhua River Basin to evaluate similarity, complexity, and trends in the snow cover, temperature, precipitation, and streamflow. In this paper, we suggest a new method of ranking the statistics symbolic sequences to examine the degree of similarity (distance measurement) between meteorological stations and compare it with non‐parametric correlation methods and also investigate the deviations in the complexity of a hydrometeorological time series. Information‐based similarity index and multiscale entropy confirm that the hydrometeorological time series of different stations have self‐similarity and abundant complexity. Wavelet entropy is also used to investigate the basin behaviour by taking streamflow records and population. It is found that with the increase in population and urbanization, the complexity values are increased. The results also exhibit that due to increase in urbanization, it affects the hydrological process and nature of environment resulting in complex catchment behaviour. Furthermore, the streamflow trend results displayed significant decline (22.21 m³/s × year^?1) in the Songhua River. The results also indicated that the seasonal snow cover trend has no impact on changes of the streamflow. However, the decline of the streamflow may be influenced by the significant human activity upstream of the Songhua River.