Fish-scale-like nano-porous membrane based on zeolite nanosheets for long stable zinc-based flow battery

Zinc-based flow batteries receive widespread attention due to their advantages of low cost and high energy density. However, zinc dendrites are easy to appear during the charge process, pierce the membrane and thus destroy the battery, which seriously restrict its further development. In this article, MFI-type zeolite nanosheets (ns-MFIs) with high mechanical strength and hydrophobicity are in situ introduced to porous polymer membranes, which spontaneously form turnup fish-scale-like structure through the one-step phase inversion/surface segregation process. This special structure well disperses mechanical energy to provide effective protection characteristics to resist the penetration of zinc dendrites, and meanwhile promotes the uniform zinc depositions on the electrode by alleviating the water migration and accelerating zincate ion diffusion, so as to prolong the cycle life of the battery for more than 600 cycles, which is 4 times and 2.5 times longer than the commercial Nafion 212 and pristine porous polymer membrane, respectively. Moreover, the sub-nano size pores and high-aspect-ratio of ns-MFIs afford membranes extra ion sieving ability and transport area for the charging-balancing ions OH⁻ to ensure superior battery performance, delivering an average coulombic efficiency (CE) of ~98.5%, voltage efficiency (VE) of ~83.2%, and energy efficiency (EE) of ~81.9% at 80 mA/cm².     

多级纳米孔分子筛在非均相催化中的应用现状及发展前景

传统分子筛因其单一的微孔孔道,在工业应用中表现为扩散阻力差、催化易失活,尤其在涉及大分子的反应过程中催化活性较差是阻碍其工业应用的现实难题,通过优化制备路线得到的多级纳米孔分子筛催化材料可有效解决传统分子筛存在的上述应用缺陷。多级纳米孔分子筛相比传统分子筛因其特殊的孔道结构和物化性能,在非均相催化方面具有丰富的催化活性位点、较短的扩散路径、较高的传递效率和较长的催化寿命,特别在涉及非均相催化反应的现代化学工业中展现出重要的潜在应用价值。综述多级纳米孔分子筛在烃类异构化反应、加氢裂化反应、烷基化与酰基化反应、烯烃氧化反应以及甲醇制烃类等反应中的诸多优势及潜在应用。     

Sweet potato amylose and amylopectin with narrower distribution of molar mass and chain length obtained by a repeated retrogradation–hydrolysis procedure

The wide molar mass distribution of native starch creates obstacles in investigating the physicochemical characteristics of starch, such as retrogradation, because samples thought to be the same are actually compounds containing many chains with different molar masses. In this paper, the sweet potato amylose and amylopectin isolated from retrograded starch were treated with the retrogradation–hydrolysis method three times, and their physicochemical changes in this process were determined by absorbance of the starch–iodine complex, light microscopy, and molar mass and chain length distributions. The results showed that repeated retrogradation and hydrolysis caused the molar mass distribution of sweet potato amylose and amylopectin to reduce from 4.2 × 10⁷–205 and 7971–223 to 6.0 × 10⁴–730 and 4533–211 g mol⁻¹, respectively. This retrogradation–hydrolysis cut the chain length distribution of sweet potato amylose from DP 9–35 to DP 3–13, but that of amylopectin remained unchanged. The double helix in sweet potato amylopectin will not form if the percentage of chain length with DP ≥ 4 is less than 25%. Repeated retrogradation and hydrolysis was an appropriate method to obtain amylose or amylopectin with a narrower molar mass distribution. © 2016 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2016 , 133, 43849.     

带保护涂层的V—B—RE共渗膏剂工艺的研究

对４０Ｃｒ，Ｔ１０，ＧＣｒ１５，ＣｒＷＭｎ，Ｃｒ１２钢进行了带保护涂层的Ｖ－Ｂ－ＲＥ共渗膏剂工艺的研究。使用所研制的Ｖ－Ｂ－ＲＥ共渗膏剂，可实现钢在共渗后的直接淬火，渗剂在淬火时自行脱落。试验结果表明：涂层外侧是以ＶＣ为主的碳化物，内侧是以Ｆｅ２Ｂ为主的硼化物。涂层最高硬度可达ＨＶ０．１２１００～２３００，耐磨耐蚀性能良好，渗层与基体结合紧密，硬度梯度平缓。工艺设备简单，最适合单件小批或大件局部的     

汽车转向系减振器减摆效果仿真计算分析

采用仿真计算方法,对汽车转向系液压阻尼减振器消减一种国产微型汽车独立悬架转向轮摆振的效果进行了分析。建立了该车转向系的五自由度集总参数摆振分析模型,并确定了模型的各种参数;应用数值方法求解系统摆振运动微分方程组,进行了多种情况的摆振仿真计算分析,得出了关于采用阻尼减振器控制转向轮摆振的几点重要结论。     

基于人工神经网络的偏微分方程求解方法

针对各种微分方程传统的数值方法的求解中,传统的数值方法在求解微分方程时存在计算存储量、计算时间等都随着微分方程维数的增加而剧烈增长的问题,严重制约了高维问题的求解.提出一种求解偏微分方程的神经网络方法.通过数值算例说明,神经网络方法在求解微分方程问题上有高效率、很好地泛化等优点,为偏微分方程求解提供了一种有效的新途径.     

基于长短时记忆网络的航线市场份额预测

当前主流航线市场的份额预测方法是服务质量指数（QSI）模型,但此方法需要模型线性化和大量人工经验. 提出基于长短时记忆网络的航线市场份额预测模型,利用该模型对航班市场份额进行预测,并通过在简化数据集上进行试验来验证模型的有效性. 以均方根误差为评价指标,对模型的参数进行优化,分别测试运力预测,QSI模型和提出的预测模型等3种方法的预测精度. 试验结果表明提出的模型能更好地预测航线市场份额,均方根误差在0.1左右.     

基于电力物联网建设的数据中心能耗管理研究

电力物联网的建设支撑了电网业务与新兴业务的发展,并将进一步全面形成共建、共治、共享的能源互联网生态圈。依托电力物联网建设,数据中心可实现在线设备的实时监测和科学化的统一能耗管理,改善传统数据中心存在的设备分布不合理、监测设备冗杂等问题。通过对数据中心的能耗问题进行梳理,构建了基于电力物联网建设的数据中心,实现了数据中心整体设备的互联互通,降低了数据中心的整体能耗。最后对新型数据中心的发展进行了分析和展望。     

基于车载激光点云的道路交叉口检测与识别

道路交叉口是道路交通网的重要组成部分,其位置和类型是高精地图、自动驾驶等应用服务的基础数据.目前研究多关注车载激光点云的道路边界提取,较少关注道路交叉口类型识别.为此,本文提出一种基于动态图神经网络的道路交叉口分类方法.首先分析地面超体素的几何和空间分布差异进行提取道路边界点;然后计算道路边界点曲率,利用滑动窗口中曲率...