Applications of Diglycolamide Based Solvent Extraction Processes in Spent Nuclear Fuel Reprocessing,Part 1: TODGA

Over the last decade there has been much interest in the applications of diglycolamide (DGA) ligands for the extraction of the trivalent lanthanide and actinide ions from PUREX high active raffinates or dissolved spent nuclear fuel. Of the DGAs, the N,N,N’,N’-tetraoctyldiglycolamide (TODGA) is the best known and most widely studied. A number of new actinide separation processes have been proposed based on extraction with TODGA. This review covers TODGA-based processes and extraction data, specifically focusing on how phase modifiers have been used to increase metal loading and thus enhance the operating process envelopes. Effects of third phase formation and the organic phase speciation are reviewed in this context. Relevant aspects of the extraction chemistry of important solvents (TODGA-modifier-diluent combinations) are described and their performances demonstrated by a consideration of the published flowsheet tests. It is seen that modifiers are successfully enabling the use of TODGA in actinide separation processes but to date the identification and testing of suitable modifiers has been rather empirical. There is a growing understanding of the fundamental chemistry occurring in the organic phase and how that affects extractant speciation and metal loading capacity but studies are still needed if TODGA-based flowsheets are to become an industrially deployable option for minor actinide (MA) recovery processes.     

牵引变压器模糊故障诊断技术

牵引变压器是牵引供电系统的重要组成部分之一，对其DGA成分和含量进行在线监测并进行故障判断，能有效地提高电气化铁路运行可靠性。目前对变压器的故障性质判断大多是根据IEC三比值法进行的。而实践中，一些DGA结果与已存在的IEC编码不符，或造成缺码现象。论文采用模糊三比值诊断法进行故障性质判断能有效地解决这一问题。     

用溶解气体分析法诊断变压器故障

用溶解气体分析法对一台大型变压器内部乙炔超标现象进行了分析，并结合该变压器的结构，判断出故障所在位置。     

基于MCD稳健统计分析的变压器油色谱异常值检测及分布特性

由于人为因素以及现场干扰的影响,变压器油色谱在线监测数据在测量和传输中会不可避免地出现偏误,导致异常数据值,使得经典统计方法的结果出现偏差甚至错误。为了解决上述问题,基于稳健统计理论,根据油色谱监测数据异常值的特点,提出了油色谱H2、CO和总烃3类特征气体异常值的最小协方差行列式MCD稳健多元检测方法。利用迭代和Mahalanobis距离的思想构造一个稳健的协方差估计量,并进行异常值检测,然后再将异常数据和正常数据分类处理。针对H2、CO和总烃这3类特征气体的实例统计分析表明,异常值剔除后可有效减少这3种气体的测量值对经典统计方法的干扰,使得油色谱数据的统计规律更加明显。通过对异常值区间的跟踪评估,还可更加明显地反映变压器运行状态的变化。     

一起500 kV自耦变压器油色谱数据异常缺陷诊断及处理

本文对一起500 kV自耦变压器油色谱数据异常缺陷进行了诊断与处理.通过对该变压器油色谱与电气试验数据及历史运维情况的综合分析,判断设备内部存在导体接触不良引起的过热缺陷.经设备内部检查发现其无载分接开关触头烧蚀,验证了缺陷诊断结果,开展电镜扫描与元素分析,推断了缺陷原因及发展过程,对同类型变压器的运维给出了建议.     

不平衡样本下基于变分自编码器预处理深度学习和DGA的变压器故障诊断方法

为提高变压器故障诊断效果,并改善训练样本数量不平衡对故障诊断的不利影响,提出了一种基于变分自编码预处理深度学习和油中溶解气体分析(dissolved gas-in-oil analysis,DGA)的变压器故障诊断方法.该方法以各样本DGA特征量为诊断模型输入,以各故障状态概率分布为诊断模型输出.首先通过变分自编码器对...     

神经网络结构与算法对变压器故障诊断的影响

BP神经网络算法本质上是基于梯度下降的一种迭代学习算法,存在学习收敛速度慢、收敛精度低、易陷入局部极小、学习率难以选取、隐层数及隐层神经元个数难以确定等缺陷。为了选择出更适宜变压器DGA故障诊断的神经网络结构及算法。本文采用了常用的几种智能算法对变压器故障样本进行了诊断性能对比实验。结果得出Levenberg-Marquardt神经网络算法是收敛速度较快的算法,有动量和自适应的梯度下降法是收敛稳定性较佳的算法;网络最优结构设计过程。为用于变压器DGA故障诊断的神经网络的结构和算法提供了系统化的试验方法。     

油溶解气体色谱分析中的小型真空在线脱气技术

基于当前气相色谱分析技术的发展,针对变压器绝缘油中溶解气体色谱在线分析的状况,提出了色谱系统对气相中气体组分体积分数最小检测值极限算法的数学模型。应用该模型,可对脱气室和集气室体积进行小型化计算和试验典型配置。在此基础上,文中设计了一种新型的小型真空在线脱气系统(10cm×10cm×20cm)和新型的全自动进样器(5cm×5cm×2cm)。系统采用全自动化控制。用标准气体试验确定了定量器体积(0.769mL),用样品气体试验对整个系统进行了检测。试验结果表明,文中所设计的小型真空脱气系统能较好地用于变压器绝缘油中溶解气体在线色谱分析。     

基于APCNN和BiGRU-Att的单词DGA域名检测方法

为了提高对基于单词的域名生成算法(domain generation algorithm, DGA)生成的恶意域名的检测准确率,提出了一种结合改进的并行卷积神经网络(APCNN)和融合简化注意力机制的双向门控循环单元(BiGRU-Att)的网络模型,该模型能充分学习单词特征、单词之间的组合关系和关键字符信息。实验结果表明,相比Bilbo和CL模型,APCNN-BiGRU-Att模型的分类准确率和F₁值更高,表明该模型具有更好的检测效果、多分类效果和稳定性。     

Comparison of MEA and DGA performance for CO2 capture under different operational conditions

Carbon capture and storage from flue gases is the most common method to reduce greenhouse gas emissions. Using a primary amine as the solvent of CO₂ capture unit is popular because of its high activity and ability to be used for streams with low concentration and low partial pressure of CO₂. Monoethanolamine(MEA) and Diglycolamine(DGA) are the most common kinds of primary amines which have been traditionally used in many natural gas sweetening plants. In this research, the capture plant has been designed for these two solvents at various CO₂ concentrations in the feed flue gas. This paper proposes different possible alters to overcome the high energy requirements of capture plant. It also presents the results of technical evaluation of different parameters, in order to design an actual plant with minimum energy requirement. The results of different parameters show that for DGA solvent, there will be an improvement in overall energy usage in the capture plant rather than MEA for some special cases. To gain the practical results, actual stages have been used for absorber and stripper instead of equilibrium stages. Copyright © 2011 John Wiley & Sons, Ltd.