苯胺加氢制环己胺与二环己胺催化剂进展

综述了国内外苯胺加氢制环己胺与二环己胺催化剂的进展，按活性组分分别对钴系，钌系，镍系，钯系苯胺加氢催化剂研究现状进行叙述，并提出了今后的研究方向。     

Aromatization of n-hexane on Mo2C catalysts

The deposition-precipitation method was used to prepare gold catalysts based on different supports. Their catalytic activities for combustion of carbon monoxide (CO) and formaldehyde (HCHO) were investigated. All these catalysts showed good activity for the two reactions and the Au/CeO₂-a catalyst exhibited the highest activity for the two reactions. Furthermore, catalysts derived from the as-precipitate hydroxides exhibited higher activity than that from corresponding oxide supports. The BET, XRD, TEM and XPS were carried out. The results indicated that the gold dispersed more homogeneous on the as-precipitate hydroxide supports than that on the corresponding oxide supports.     

Direct conversion of methane to higher hydrocarbons via an oxygen free,low-temperature route

The direct conversion of methane to higher hydrocarbons over a silica-supported Ru catalyst has been investigated via an oxygen free, two-step route. The reaction consists of decomposition of methane over a 3% silica-supp orted Ru catalyst at temperatures between 400 and 800 K to produce surface carbonaceous species followed by rehydrogenation of these species to higher hydrocarbons at of 368 K. It was found that the Ru/SiO₂ catalyst exhibits a trend similar to that for single-crystal Ru catalysts. However, the temperature at which a maximum in ethane selectivity occurs shifts toward a higher temperature. It was also found that the ethane yield can be optimized by changing the surface carbon coverage. Under optimum conditions a net ethane yield of about 13–15% has been realized. For this two-step reaction sequence, only a few reaction cycles could be operated without intermediate high temperature rehydrogenation and without significant loss in ethane yield. This is attributed to large amounts of inactive carbon that could not be hydrogenated at 368 K. Higher methane partial pressures were found to be desirable for this reaction. The activity of the catalyst could also be maintained at total pressures up to 10 atm.     

混凝土现浇箱形拱圈与拱架联合作用研究

混凝土箱型拱桥在无支墩拱式拱架上现浇施工时,采用分环分段的施工方法,可以利用拱圈与拱架的联合作用,减少钢拱架的用钢量,达到节省施工成本的目的.通过大型通用有限元软件MSC.Marc2005的模拟及实桥测试数据的分析,发现先期浇注的混凝土拱环与拱架的联合作用明显,拱圈可承担其自重的43%,即拱架只承担拱圈重量的57%,利用联合作用可节约用钢量达30%.同时提出了一个估算拱架承担拱圈重量的公式,该公式简便实用,并通过两种施工工艺的计算结果验证了该公式的准确性,可用于指导初步设计.     

Nd(OR)_3催化异戊二烯聚合动力学及聚合物结构

研究了Nd(OR)_3—A1(i—Bu)_3一A1Et_2Cl体系催化异戊二烯聚合的动力学和所得聚异戊二烯的分子结构。结果表明,本体系的有效利用率为6％左右。在Al(i—Bu)_3用量较高时,主要发生活性中心向A1的链转移;在A1/Nd≤20,聚合温度低于室温条件下可获得“活的”聚合物。聚合物分子量高、分布宽,主要受A1(i—Bu)_3用量和聚合温度影响。分子量分布宽与其GPC图呈双峰有关。催化剂的相分离实验表明,体系中有两种不同的活性种存在:高活性生成高分子量聚合物的固相活性种和低活性生成低分子的可溶性活性种。聚异戊二烯的顺1,4—链节含量在95％左右,其余为3,4—链节。     

An EXAFS study of the Mo/TiO2 hydrodesulphurization catalysis

In-situ EXAFS studies of sulphided Mo/TiO₂ catalysts have shown that at low Mo loadings (2–4 wt%), an active species with a short Mo-S distance of 2.25 Å is formed, while on Mo/TiO₂ with high Mo loadings as well as on Mo/-Al₂O₃, bulk MoS₂ (Mo-S, 2.42 Å) is formed. The species with the short Mo-S distance has Mo in an oxidation state close to 6 + and is likely to result from the sulphidation of the tetrahedral molybdate species present in the oxidic precursor at low Mo loadings. The calcination temperature of the oxidic precursor appears crucial, a high calcination temperature of 973 K favouring the formation of MoS₃ on sulphidation, and a low calcination temperature of 623 K favouring MoS₂.Contribution No. 790 from the Solid State and Structural Chemistry Unit.     

智能变电站过程层网络异常流量检测

为了避免智能变电站过程层网络通信出现异常变动的现象,需要准确检测智能变电站过程层网络异常流量,为此提出新型的检测方法.设计了基于网络演算的变电站通信网络流量计算模型,将根节点全部设成信源,通过流量路由实现周围输入与输出端口的联系,获取智能变电站过程层网络中全部设备端口输入与输出流量.还应用优化支持向量模型进行异常流量检测,将网络异常流量与正常流量分类,实现智能变电站过程层网络异常流量检测.实验结表明:在检测,网络流量特征提取、异常流量检测效果均符合应用需求.     

中继协同无人机辅助的认知无线电网络安全通信

针对中继协同无人机（UAV）辅助的无线通信网络,提出一种基于认知无线电网络（CRN）的物理层安全通信方案。利用二次发射机协同解码转发中继向目的接收机发送机密消息,将UAV用作移动干扰器发送干扰噪声,以降低窃听者的解码能力。在不影响主用户通信的前提下,通过联合优化UAV的飞行轨迹和发射功率提高系统的平均保密率,采用基于连续凸逼近的算法求解近似凸规划的保密率最大化问题。仿真结果表明,相较于优化功率和优化轨迹这两种传统方案,该方案能够进一步提高CRN系统通信的安全性。     

一类SP结构的不可能差分区分器证明

针对分组密码SP结构的不可能差分区分器轮数的下界进行证明,提出的方法使用线性代数的理论,对系数矩阵P及P-1进行分析,提出了系数矩阵部分子空间存在两个行向量线性相关时,可证明至少存在四轮不可能差分区分器。uBlock算法是SPN结构,提出的方法对uBlock算法进行了分析验证,说明了结论的正确性,进一步,使用该算法搜索到比uBlock算法设计文档更多的不可能差分区分器。针对SPN结构线性扩散层P,使用了本原指数的概念,使用线性扩散层P的本原指数对SPN结构不可能差分的轮数进行论证。分析结果表明,分组密码SP结构至少存在四轮不可能差分区分器。     

基于加速残差网络的图像超分辨率重建方法

针对深层网络架构的图像超分辨率重建任务中存在网络参数多、计算复杂度高等问题,提出了一种基于加速残差网络的图像超分辨率重建方法。首先,构建一个残差网络对低分辨率图像与高分辨率图像之间的高频残差信息进行重建,以减少冗余信息的深层网络传输过程,提高重建效率;然后,通过特征收缩层对提取的低分辨率特征图进行降维,从而以较少的网络参数实现快速映射;之后,对高分辨率特征图通过特征扩展层进行升维,从而以较丰富的信息重建高频残差信息;最后,将残差与低分辨率图像求和得到重建的高分辨率图像。实验结果表明,该方法取得的峰值信噪比（PSNR）及结构相似性（SSIM）均值结果较基于卷积神经网络的图像超分辨率（SRCNN）取得的结果分别提升了0.57 dB和0.013 3,较基于中间层监督卷积神经网络的图像超分辨率重建（ISCNN）取得的结果分别提升了0.45 dB和0.006 7;在重建速度方面,以数据集Urban100为例,较现有方法提高了1.5~42倍。此外,将该方法应用于运动模糊图像的超分辨率重建时,获得了优于超深卷积神经网络的图像超分辨率（VDSR）的性能。所提方法以较少的网络参数快速获得较好的重建质量,并且为图像超分辨率重建提供了新的思路。