团簇Co3FeP异构化反应的热力学与动力学研究

为了寻找团簇Co₃FeP的稳定性、结构和内部作用力等规律，利用密度泛函理论，在B3LYP/Lanl2dz的量化水平下对团簇Co₃FeP进行理论计算，得到5种优化构型，并从化学反应热力学与动力学的角度对其异构化反应中的反应物、过渡态和生成物进行了讨论分析。结果表明，所有构型均为C1对称且构型的空间结构对其熵值有影响；构型的稳定性主要由金属原子与非金属原子所决定，并且在常温常压下构型1^（3）、2^（3）和3^（3）可以稳定存在，是大部分异构化反应的最终产物。因此，在实际材料的开发中，不仅需要注意金属原子与非金属原子之间的作用力以确保构型的稳定性，还应首先考虑对稳定构型1^（3）、2^（3）和3^（3）进行相关研究。     

The Dynamics Characteristics of Flexible Spacecraft and its Closed-Loop Stability with Passive Control

Passive control is the most popular methodology for flexible spacecraft while it remains an open problem whether the closed-loop performance can be achieved only with passive control subject to the coupling modes of rigid and flexibility. Also, the closed-loop performance of passive PD control based on the dynamics of the Euler angle parameterization of spacecraft, which has been widely used in practice, is yet to be addressed. Towards these challenges, by introducing the input-output exact linearization theory and Lyapunov theory, the authors show that the closed-loop performance for flexible spacecraft with rigid and flexible modes can be achieved by adjusting the parameters of the passive controllers sufficiently large. This is done by firstly transforming the flexible spacecraft dynamics into an exact feedback linearization standard form, and then analyzing the closed-loop performance of flexible spacecraft.     

基于知识图谱的核电设备健康管理知识建模与分析

随着信息技术发展,设备健康数据与知识图谱技术结合为设备健康发展带来新的发展机遇,知识图谱利用其多种特异性优势,使企业系统中积累的海量设备维修数据得到有效整合利用.本文提出基于知识图谱的核电设备健康管理知识建模与分析方法,并根据实体关系模型构建维修知识图谱本体框架,同时基于核电企业实际的维修工单数据,从统计分析和关联分析两方面取得良好的实践效果.研究结果表明,知识图谱能够在设备维修知识的集成上解决传统的数据孤岛问题,同时能基于故障的可视化分析为维修人员和管理人员提供决策和知识支持.     

复合跟踪微分器在电容式位移传感器中的应用

针对从含噪原始信号中提取位置以及速度信息,经典跟踪微分器存在不能很好兼顾相位滞后和噪声放大问题、参数多,调试复杂等不足.在跟踪微分器等效线性分析基础上,提出复合形式跟踪微分器,用于电容式位移传感器位置信号跟踪以及速度信号估计,通过MATLAB\SIMULINK仿真以及实验平台测试,结果表明:在跟踪频率1 Hz、幅值1含噪声正弦信号中,复合跟踪微分器能光滑逼近原始位置信号,且能有效进行速度估计,相较于经典跟踪微分器,复合跟踪微分器跟踪相位滞后小0.03 rad,能更好兼顾跟踪信号相位滞后及速度信号噪声放大.     

基于差异信息量的多源数据融合方法

D-S证据理论可应用于多源数据融合领域，但在处理高度冲突的证据时，可能会出现反直觉的结果.为解决这一问题，本文提出了差异信息量的概念及融合方法.首先，通过信息熵表明证据的相对重要性，采用散度获取证据可信度.然后利用证据可信度优化证据差异度以得到差异信息量，经过计算获取数据的最终权重，并将其作为D-S证据理论中的基本概率分配进行决策.在处理冲突证据、一致证据及不同数量证据等方面的数据融合问题时与其他方法对比，所提方法收敛更快，准确度更高.故障诊断的应用实例表明，所提方法的不确定性更小，优于现存的其他方法.     

Optimal investment problem for an insurer and a reinsurer

This paper studies the optimal investment problem for an insurer and a reinsurer. The basic claim process is assumed to follow a Brownian motion with drift and the insurer can purchase proportional reinsurance from the reinsurer. The insurer and the reinsurer are allowed to invest in a risk-free asset and a risky asset. Moreover, the authors consider the correlation between the claim process and the price process of the risky asset. The authors ?rst study the optimization problem of maximizing the expected exponential utility of terminal wealth for the insurer. Then with the optimal reinsurance strategy chosen by the insurer, the authors consider two optimization problems for the reinsurer: The problem of maximizing the expected exponential utility of terminal wealth and the problem of minimizing the ruin probability. By solving the corresponding Hamilton-Jacobi-Bellman equations, the authors derive the optimal reinsurance and investment strategies, explicitly. Finally, the authors illustrate the equality of the reinsurer's optimal investment strategies under the two cases.     

基于曼哈顿距离构建非参数Q矩阵修正方法

将被试得分、理想反应距离和被试得分异常原理相结合,并加入属性计分下的被试得分特性,开发了用于多级评分情境下属性计分曼哈顿距离法(SA-MD),在不同条件下验证了SA-MD的稳定性和适宜性.通过模拟研究和实证研究表明:(i)从逻辑推导出SA-MD用于多级评分情境下Q矩阵修正更合理;(ii)在多种条件中,SA-MD 均有较优的修正效果,适用范围更广,稳定性更高;(iii)当小样本测验进行Q矩阵修正时,使用SA-MD方法可获得更优的效果.     

Defining G protein-coupled receptor peptide ligand expressomes and signalomes in human and mouse islets

Introduction

Islets synthesise and secrete numerous peptides, some of which are known to be important regulators of islet function and glucose homeostasis. In this study, we quantified mRNAs encoding all peptide ligands of islet G protein-coupled receptors (GPCRs) in isolated human and mouse islets and carried out in vitro islet hormone secretion studies to provide functional confirmation for the species-specific role of peptide YY (PYY) in mouse islets.

Materials and methods

GPCR peptide ligand mRNAs in human and mouse islets were quantified by quantitative real-time PCR relative to the reference genes ACTB, GAPDH, PPIA, TBP and TFRC. The pathways connecting GPCR peptide ligands with their receptors were identified by manual searches in the PubMed, IUPHAR and Ingenuity databases. Distribution of PYY protein in mouse and human islets was determined by immunohistochemistry. Insulin, glucagon and somatostatin secretion from islets was measured by radioimmunoassay.

Results

We have quantified GPCR peptide ligand mRNA expression in human and mouse islets and created specific signalomes mapping the pathways by which islet peptide ligands regulate human and mouse GPCR signalling. We also identified species-specific islet expression of several GPCR ligands. In particular, PYY mRNA levels were ~ 40,000-fold higher in mouse than human islets, suggesting a more important role of locally secreted Pyy in mouse islets. This was confirmed by IHC and functional experiments measuring insulin, glucagon and somatostatin secretion.

Discussion

The detailed human and mouse islet GPCR peptide ligand atlases will allow accurate translation of mouse islet functional studies for the identification of GPCR/peptide signalling pathways relevant for human physiology, which may lead to novel treatment modalities of diabetes and metabolic disease.

     

联合功率域SPMA协议的动态阈值算法

针对基于优先级概率统计的多址接入(statistic priority-based multiple access, SPMA)模式协议在高负载情况下战术信息的碰撞概率十分严重, 网络性能无法满足协同攻击要求的问题, 本文借鉴非正交多址接入技术在功率域上区分同时同频信号的思想, 改进SPMA协议信号发送机制与多优先级检测机制, 在原有的时域、频域基础上引入功率域, 提出一种联合功率域的SPMA协议性能改进策略。并针对各优先级数据包到达率的动态变化场景, 设计动态阈值算法, 以达到降低碰撞概率, 提升网络吞吐量的目的。仿真结果表明, 采用该改进策略与动态阈值算法可以使网络承载更大的业务流量, 提高网络吞吐量。     

基于新型空间注意力机制和迁移学习的垃圾图像分类算法

随着我国各级政府大力推动垃圾强制分类,分类回收各环节中实现标准化、自动化的垃圾分类识别需要适合云端部署的高准确率、低延时要求的细粒度图像分类模型.本文发挥深度迁移学习的优势建立了一套端到端的迁移学习网络架构GANet (garbage neural network);针对垃圾分类中类别易混淆、背景干扰等挑战,提出一种新型的像素级空间注意力机制PSATT (pixel-level spatial attention).为克服类别多和样本不平衡挑战,提出使用标签平滑正则化损失函数;为改善收敛速度以及模型稳定性与泛化性,提出了阶梯形OneCycle学习率控制方法,并给出了结合Rectified Adam (RAdam)优化方法和权重平滑处理技术的组合使用策略.实验使用了"华为云人工智能大赛.垃圾分类挑战杯"提供的按照深圳市垃圾分类标准标注的训练数据,验证了GANet在垃圾分类问题中的显著效果,获得了全国二等奖(第2名);同时,提出的PSATT机制优于对比方法,且在不同主干网络架构上均得到了提升,具有良好的通用性.本文提出的GANet架构、PSATT机制和训练策略不仅具有重要的工程参考价值,也具有较好的学术价值.