Lower Bounds for the Robustness of Multilevel Coherence

International Journal of Theoretical Physics - Quantum coherence, coming from quantum superposition, occupies a significant position in the field of physics. We put forward a lower bound of the...     

n维有限射影空间上d~e——析取矩阵的新构作

利用n维有限射影空间上的一些性质,构作了组合群验的数学模型de-析取矩阵,并研究了它的参数和Hamming距离.     

盐酸伊达比星杂质A的合成工艺优化及其高效液相分析

通过对盐酸伊达比星脱羟基杂质A的合成研究，为盐酸伊达比星的质控标准提供参考。以盐酸伊达比星为起始原料，经过4步反应得到脱羟基杂质A。本文提供了盐酸伊达比星杂质A的制备方法以及高效液相色谱分析方法，为杂质含量控制提供了依据。该工艺可以稳定、快速制备高纯度脱羟基杂质A，制备总收率为57.20%，纯度为57.20%，其结构经¹H NMR， ¹³C NMR和MS(ESI)表征。     

Metabolic profiles of dioscin in rats revealed by ultra‐performance liquid chromatography quadrupole time‐of‐flight mass spectrometry

Dioscin (DIS), one of the most abundant bioactive steroidal saponins in Dioscorea sp., is used as a complementary medicine to treat coronary disease and angina pectoris in China. Although the pharmacological activities and pharmacokinetics of DIS have been well demonstrated, information regarding the final metabolic fates is very limited. This study investigated the in vivo metabolic profiles of DIS after oral administration by ultra‐performance liquid chromatography quadrupole time‐of‐flight mass spectrometry method. The structures of the metabolites were identified and tentatively characterized by means of comparing the molecular mass, retention time and fragmentation pattern of the analytes with those of the parent compound. A total of eight metabolites, including seven phase I and one phase II metabolites, were detected and tentatively identified for the first time. Oxidation, deglycosylation and glucuronidation were found to be the major metabolic processes of the compound in rats. In addition, a possible metabolic pathway on the biotransformation of DIS in vivo was proposed. This study provides valuable and new information on the metabolism of DIS, which will be helpful for further understanding its mechanism of action. Copyright © 2015 John Wiley & Sons, Ltd.     

基于深度学习的ELM实时识别研究

基于深度学习的方法，在HL-2A装置上开发出了一套边缘局域模(ELM)实时识别算法。算法使用5200次放电数据(约24.19万数据切片)进行学习，得到一个深度为22层的卷积神经网络。为衡量算法的识别能力，识别了HL-2A装置自2009年实现稳定ELMy H模放电以来所有历史数据(约26000次放电数据)，共识别出1665次H模放电，其中误识别35次，误报率为2.10%。在实际的1634次H模放电中，漏识别4次，漏识别率为0.24%。该误报率和漏报率可以满足ELM实时识别的精度要求。识别算法在实时控制环境下，对单个时间点的平均计算时间为0.46ms，可以满足实时控制的计算速度要求。     

边剖分、点扩张与图的最大亏格的可约性

设γM（G）是连通图G=（V，E）的最大亏格，记EM^-（G）={e∈E(G)|G\e连通，且γM（G\e）=γM（G）}。若EM^-(G)≠0，则称G是γ（G）-可约的；否则称G是γM（G）-不可约的。本文证明了边的剖分不改变图的最大亏格可约性，点的扩张不改变上可嵌入图的最大亏格可约性；并给出了两类满足EM^-（G）=E（G）的非4-边连通图。