模块化多电平变换器全桥型子模块优化均压控制方法

模块化多电平变换器(MMC)子模块(SM)的数量与直流侧电压成正比,当SM增加时,会导致MMC的开关损耗急剧增加,因此降低功率器件的开关频率一直是MMC的重要研究方向之一。采用最近电平逼近调制(NLM)方式,提出一种基于全桥型SM的改进均压排序法,旨在降低MMC中功率器件IGBT的开关频率,该方法实现相对简单,无需额外的控制器,且易于扩展。最后,通过在MATLAB/Simulink平台搭建了19个全桥SM的仿真模型,验证了该方法的有效性。验证了所提全桥型SM优化均压策略,可以有效避免IGBT不必要的反复投切,降低IGBT的开关损耗,同时对外部输出特性不会产生负面影响。     

考虑应力历史的饱和黏土一维固结渗透试验研究

为研究固结应力比和应力历史对饱和黏土一维固结渗透的影响,利用GDS三轴仪进行了一系列一维固结-渗透联合试验。研究结果表明,在不同固结应力比相同应力历史条件下,正常固结和超固结饱和黏土在一维固结渗透过程中试样的渗透系数和孔隙比均随固结应力比的增大而减小,试样的最终压缩量均随固结应力比的增大而增大。在相同固结应力比不同应力历史条件下,超固结饱和黏土的最终压缩量小于正常固结饱和黏土。将固结应力σ与土层自重应力σcz的差值σdv与试样最终压缩量sf进行拟合,得出应力差值σdv与各固结应力下的最终压缩量sf间呈良好的线性关系。通过研究发现,同时考虑固结应力比和应力历史对饱和黏土一维固结渗透的影响是有必要的,同时在得到土体自重应力后可较为准确地预测土体最终压缩量。     

新疆和田植被水分利用效率的SEM-ANN模型构建

水分利用效率（WUE）反映了消耗单位质量水所制造的干物质量,是评估植被生长适宜性的综合指标,但是WUE的复杂影响机制使得区域多种因素的驱动贡献不甚明晰。为了探究多种因素对WUE的直接或间接影响关系,以求得到更好的WUE模拟效果,本研究构建了结构方程模型（SEM）与人工神经网络（ANN）相耦合的SEM-ANN模型。由SEM确定WUE影响因素之间的结构关系及影响程度,然后转化为ANN的拓扑结构。结果表明：在新疆和田地区,不同植被类型WUE的影响因素和影响层次有所不同;温度、降水、饱和水汽压差、风速均对植被WUE有着不同程度的直接影响;在草地及耕地中,增强植被指数（EVI）可体现为为间接影响WUE的中间变量;而在灌木及常绿针叶林中,标准化降水蒸散发指数（SPEI）将作为中间变量。经过SEM优化后的ANN结构,拟合效果更好,SEM-ANN模型对于生态系统的环境控制和WUE的模拟有着很高的解释性和精确性,这为提高新疆生态系统的高效用水能力以及预测未来气候变化下WUE的响应状况提供了理论依据和模拟途径。     

基于局部拓展理论的直流多馈入系统分区方法

针对多回电网换相型高压直流输电(LCC-HVDC)馈入的交流系统容易发生同时换相失败的问题,本文采用了对多馈入系统进行电网分区的方案,以隔断多回直流之间的耦合,减少发生同时换相失败概率.首先,对多馈入系统进行N-1故障扫描,找出三永交流故障能够引起多回直流发生同时换相失败的线路集合,并将线路负载率较高的线路从线路集合中剔除.随后,基于局部拓展理论对系统中的节点进行自下而上的聚合,形成初始分区.在此过程中,首先给出了考虑PV节点和传统直流的电压/无功灵敏度矩阵的计算方法,随后采用节点聚合度函数和分区适应度函数分别计算节点聚合度和分区适应度.最后,综合考虑线路集合和分区短路比对分区结果进行修正,并给出了分区方案的流程.以IEEE 39节点系统为参考给出了算例,对整个分区过程进行验证.并结合某实际电网给出了分区方案.     

Functionalized metal–organic frameworks with strong acidity and hydrophobicity as an efficient catalyst for the production of 5-hydroxymethylfurfural

In the dehydration of fructose to 5-hydroxymethyl furfural(HMF), in situ produced water weakens the acid strength of the catalyst and causes the rehydration of HMF, causing unsatisfactory catalytic activity and selectivity. In this work, a class of benzenesulfonic acid-grafted metal–organic frameworks with strong acidity and hydrophobicity is obtained by the direct sulfonation method using 4-chlorobenzenesulfonic acid as sulfonating agent. The resultant MOFs have a specific surface area of greater than 250 m~2·g~(-1), acid density above 1.0 mmol·g~(-1), and water contact angle up to 129°. The hydrophobic MOF-Ph SO_3 H exhibits both higher catalytic activity and selectivity than MOF-SO_3 H in the HMF synthesis due to its better hydrophobicity and olephilicity. Moreover, the catalyst has a high recycled stability. At last, fructose is completely converted, and 98.0% yield of HMF is obtained under 120 °C in a DMSO solvent system. The successful preparation of the hydrophobic acidic MOF provides a novel hydrophobic catalyst for the synthesis of HMF.     

Reactivity study and kinetic evaluation of CuO-based oxygen carriers modified by three different ores in chemical looping with oxygen uncoupling (CLOU) process

In the chemical looping with oxygen uncoupling (CLOU) process,CuO is a promising material due to the high oxygen carrier capacity and exothermic reaction in fuel reactor but limited by the low melting point.The combustion rate of carbon is faster than the decoupling rate of oxygen carrier (OC).Hence,high tem-perature tolerance and rapid oxygen release rate of CuO modified by three different ores were investi-gated in this study.The kinetics analysis of oxygen decoupling with Cu-based oxygen carriers was also evaluated.Results showed that CuO modified by chrysolite had faster oxygen release rate than that of CuO.Limestone showed obvious positive effect on the oxidization process.The selected OCs could keep stable in at least 20 cycles,for about 1200 min.Shrinking core model (SCM) fitted well for the decoupling process in the temperature range of 1123-1223 K.Reduction rate kinetic information may aid in the development of chemical looping with oxygen uncoupling (CLOU) technologies during reactor design and process modeling.Ternary doped copper oxide with chrysolite and limestone could improve the reactivity of CuO in decoupling and coupling process and also improve the high temperature tolerance.     

Characterization of wet microalgal cells pretreated with steam for lipid extraction

Steam pretreatment was employed to disrupt microalgal cells for lipids extraction.Effects of steam pre-treatment on microstructure of microalgal cells were investigated through scanning electron microscopy(SEM) and transmission electron microscopy (TEM).Effect of treatment on lipid extraction was also stud-ied.Microalgal cell walls were distorted after steam pretreatment due to the hydrolysis of organic macro-molecules contained in cell wall.Maximum curvature was increased from 1.88 × 10-6 m-1 to 1.43 × 10-7 m-1 after treatment with the steam at 130 ℃.The fractal dimension of microalgal cells increased from 1.25 to 1.30 after pretreatment for 15 min,and further increased to 1.47 when the pre-treatment time was increased to 60 min.Increased steam pretreatment temperature and time enhanced the hydrolysis of organic macromolecules,and finally destroyed microalgal cell walls at pretreatment temperature of 130 ℃ and pretreatment time of 60 min.Lipid extracted from wet microalgal was signif-icantly increased (2.1-fold) after pretreatment.     

双极短路故障下海上风电MMC-HVDC并网系统的暂态运行特性

双极短路故障作为最严重的直流系统故障,研究该场景下海上风电模块化多电平换流器型高压直流输电(modular multilevel converter-high voltage direct current,MMC-HVDC)并网系统的暂态运行特性对于系统的断路器配置与控制方案设计有着重要意义.首先分析了直流断路器(direct current circuit breaker,DCCB)未动作时换流站短路电流的产生机理,并提出了风电场侧与电网侧交、直流短路电流的计算方法.其次,研究了计及DCCB控制作用下换流站交、直流侧暂态电压、电流特性,并考虑了风电场运行工况和限流电抗参数对其暂态特性的影响.最后,基于PSCAD/EMTDC仿真平台搭建了全系统的电磁暂态模型,通过仿真分析了双极短路故障下有无DCCB控制时换流站的暂态电压、电流特性.研究表明:所提短路电流计算方法具有良好的精度,且DCCB动作后风电场侧交、直流电压恢复速率与其输出有功功率呈正相关关系.该研究工作可为系统的电气设计、断路器选型提供理论依据,也可为系统故障恢复阶段控制方案的制定奠定基础.     

柔性直流配电网的N-1安全域

为完整刻画柔性直流配电网在满足N-1安全准则前提下的最大允许运行范围,提出了柔性直流配电网的N-1安全域模型与解析化求解方法.首先讨论了柔性直流配电网的典型接线模式与N-1事件前后的稳态特性;在此基础上建立了柔性直流配电网的N-1安全域模型,分析了其与交流配电网安全域模型的区别及其能适用于柔性直流配电网的原因;然后通过DistFlow支路潮流线性化并配合下垂节点的处理,实现了模型的解析化求解;最后通过算例验证了模型与求解方法的有效性.结果 表明:所提模型与求解方法可用于获得柔性直流配电网在主从控制、电压裕度控制、下垂控制下的安全域解析表达式;不同控制策略下安全域大小的规律是电压裕度控制下小于主从控制下,下垂控制下小于等于主从控制下.     

基于超分辨注意力机制改进的GIS内部细微缺陷X-DR图像检测方法

为了解决气体绝缘开关设备内部细微缺陷X-DR成像重影雾化、纹理不清晰、易造成误诊断等问题,提出一种基于超分辨率注意力机制改进的气体绝缘开关(gas insulated switchgear,GIS)内部细微缺陷X-DR图像检测算法.该方法以高效亚像素卷积神经网络为框架,引入双层计算机注意力机制SE模块,构成新颖的SE-ESPCN超分辨率网络.通过对GIS设备X-DR图像通道重要程度进行评判,赋予图像卷积层不同的权重,以增强图像故障细节的成像效果.实验结果表明:SE模块与亚像素卷积神经网络的融合,不仅保障了GIS设备故障区域成像的实时性,而且算法输出的高分辨率X-DR图像缺陷细节清晰可见,便于观察,提高了工作人员对缺陷检测的效率与准确性,为实际工程中GIS设备X-DR成像系统改良提供了一定的参考.