中制导虚拟目标的定位参数研究

基于虚拟目标的优化导引中制导拦截弹道轨迹及其能耗与虚拟目标的位置紧密相关,决定该位置的参数包括中末制导交班距离和视场基准角等.首先研究了弹道特性随定位参数的变化规律并提出通过滚转实现视场基准角的负取值,在此基础上得出满足多约束条件的虚拟目标定位参数的设置方法.利用该方法所产生的虚拟目标与相应的优化导引律结合,可以很好地满足中制导的要求.     

Market entry strategy in the presence of market spillovers and efficiency differentiation

Firms form various alliances or use brand extensions to enter new markets in order to improve their operational efficiency and create a positive spillover. However, they do not always know the implications of these strategies for market entry and multimarket competition because the sale of products in one market can have negative spillover effects on product sales in other markets. We present an analytical framework to examine whether and how (i.e., by choosing alliance entry or independent entry) competing firms should enter a market in a situation where market spillovers occur when a firm enters a spillover-producing market to sell products that may increase or decrease the consumers' willingness to pay for products in the primary market. Our analysis shows that the operational efficiency (or quality differentiation ability) of firms in a spillover-producing market varies, and hence, the impact of market spillovers differs for firms. We identify the key factors, such as bargaining power, brand value difference in the primary market, and the extent of efficiencies and spillovers, that determine the firms benefitting from the different entry strategies. Specifically, we show that firms would be more willing to choose an alliance strategy to enter a spillover-producing market if the negative spillover is small and alliance efficiency is high. In contrast, if an alliance entry is not favored, the firms' relative operational efficiency is crucial for them to decide whether to enter the market independently under moderate spillover conditions. Finally, we show the implications of market entry strategies for managers.     

面向混部云失败批处理作业的预测算法

为了降低混部云失败批处理作业的风险,使用K-means聚类算法将批处理作业分为四类,在分类的基础上提出了二层嵌套分类模型(two-layer nested classification model, TLNM),实现了基于TLNM的预测算法。基于Ali Trace 2018数据集上的实验结果表明,该算法的接受者操作特性(receiver operating characteristic, ROC)曲线明显优于其他常用分类器,ROC曲线下面积(即AUC)可以达到0.978,表明该算法具有良好的分类性能。同时召回率可以达到0.951,通过混淆矩阵可以看出TLNM算法能够准确预测出执行失败的批处理作业。     

Investigation of the shock compression behaviors of Al/PTFE composites with experimental and a 3D mesoscale-model

The responses of AI/PTFE reactive materials (RMs) under shock compression were investigated by a single-stage gas gun.A 3D mesoscale-model was established based on micro-computed tomography(micro-CT) slice images and confirmed with experimental results.In the high-pressure stage,the com-posites reacted partially,whereas there were no deviations between the partially reacted Hugoniot and the inert simulation results.The simulation reveals that the Teflon matrix melting on the high shock pressure.Melts and decomposition of the PTFE accelerated the diffusion of the atoms.Thus,the reactions of the Al/PrFE composites are more like a combustion rather than a detonation.     

PAO8和45#变压器油基础油粘度调和及计算模型研究

采用高低温运动粘度测定仪测量PAO8(8#聚烯烃)合成基础油和45#变压器油基础油按不同比例调和后混合基础油的粘温性能,运用Arrhenius模型和Grunberg-Nissan模型计算混合基础油的理论粘度,研究粘度偏差率随测试温度的变化规律.实验结果表明:Arrhenius模型能准确预测混合基础油的高温运动粘度,粘度偏差率最大值不超过1%,而Grunberg-Nissan模型能精确计算混合基础油的低温运动粘度;PAO8和45#变压器油基础油调和能够获得高低温粘度的匹配.     

全球治理视角下奥巴马政府太空战略解读

作为重要的全球公域范围与全球治理对象之一,太空既面临着日益严峻的资源、环境与安全问题,又不断彰显着其极具地缘政治意义的战略地位,攸关国家利益与安全。自进入太空时代以来,争夺和维持太空领域的领先地位一直是美国不懈追求的目标。美国的经济、军事等国家实力越来越依赖太空,同时其太空资产的脆弱性也愈发凸显。面对新的国际国内形势与外空态势的复合挑战,奥巴马政府适时调整了美国的太空战略。文章以全球公域和全球治理为切入视角,在分析太空领域全球治理现状的基础上,从太空活动行为准则的制定、太空国际合作的深入、太空军备竞赛的规制三个层面解读奥巴马政府的新版太空战略,并探讨其深层次的动因与影响,进而为中国参与太空领域的全球治理提供有益的理论支撑与现实参考。     

基于能源互联网背景的储能二次电池技术发展分析

能源互联网是在现有配电网基础上,融合新能源技术和互联网技术,将分布式能源采集装置和分布式能源储存装置联合发展起来的新技术,对于优化能源结构和提高能源利用具有重要的意义。文章以能源互联网对储能电池的需求为背景,首先归纳了能源互联网对于储能电池的要求,在此基础上重点分析了多种可应用于能源互联网的储能电池的性能与特点,并对其未来发展趋势进行了展望。     

管道机器人直线行进过程中多模式模糊PID控制

针对管道机器人PID控制直线行进过程中的跑偏问题,提出了一种改进的多模式模糊PID控制方法。通过建立集成直线行进偏差维度、阈值更新时间维度和初速度维度模式切换自适应阈值模型,减小了因模式切换造成的时间滞后误差。依托能力风暴AS-UⅢ机器人搭建实验测试平台,获取了不同初始速度下驱动轮的轨迹曲线。对比采用改进控制方法前后的轨迹曲线,验证了多模式模糊PID控制方法的有效性和实用性。     

基于模型概率估计的两层交互多模型算法

在机动目标跟踪问题的研究中,针对转弯机动目标跟踪精度差的问题,设计了一种利用模型概率对模型集合进行实时计算的两层交互多模型算法。该算法由第一层目标转弯速率的粗估计和第二层目标状态向量的精估计构成,它利用第一层模型计算出的目标运动估计转弯速率,构建与当前目标运动状态匹配程度较高的第二层转弯模型集合,并将该转弯模型集合运用于状态估计中。仿真结果证明了该算法在针对转弯机动目标时仍有较好的跟踪精度。     

某型防空导弹杀伤概率评估

基于防空导弹战斗部爆炸后的杀伤特征,对战斗部爆炸后产生的有效破片、破片飞行速度、破片对目标的覆盖面积进行了分析,建立了防空导弹杀伤概率评估模型。采用仿真弹道数据,对打击目标F-16战斗机应用s im u link仿真工具绘出了杀伤概率曲线,其结果与实际分析相一致,验证了评估模型的可用性。