基于熵的空中目标识别模型及应用

空中目标识别是现代防空作战的重要研究内容。研究根据不同类型的目标产生不同的辐射和不同的雷达回波信号对目标进行识别。提出利用熵理论来处理目标识别问题,采用层次分析法与熵权的组合来确定指标的权重。仿真实例表明,基于熵的识别是一种运算简单,结果可靠的目标识别算法。     

Directional change‐point detection for process control with multivariate categorical data

Most modern processes involve multiple quality characteristics that are all measured on attribute levels, and their overall quality is determined by these characteristics simultaneously. The characteristic factors usually correlate with each other, making multivariate categorical control techniques a must. We study Phase I analysis of multivariate categorical processes (MCPs) to identify the presence of change‐points in the reference dataset. A directional change‐point detection method based on log‐linear models is proposed. The method exploits directional shift information and integrates MCPs into the unified framework of multivariate binomial and multivariate multinomial distributions. A diagnostic scheme for identifying the change‐point location and the shift direction is also suggested. Numerical simulations are conducted to demonstrate the detection effectiveness and the diagnostic accuracy.© 2013 Wiley Periodicals, Inc. Naval Research Logistics, 2013     

Analysis of the finite‐source multiclass priority queue with an unreliable server and setup time

In this article, we study a queueing system serving multiple classes of customers. Each class has a finite‐calling population. The customers are served according to the preemptive‐resume priority policy. We assume general distributions for the service times. For each priority class, we derive the steady‐state system size distributions at departure/arrival and arbitrary time epochs. We introduce the residual augmented process completion times conditioned on the number of customers in the system to obtain the system time distribution. We then extend the model by assuming that the server is subject to operation‐independent failures upon which a repair process with random duration starts immediately. We also demonstrate how setup times, which may be required before resuming interrupted service or picking up a new customer, can be incorporated in the model. © 2013 Wiley Periodicals, Inc. Naval Research Logistics, 2013     

非周期不完全检测下的维修优化

针对gamma退化型单部件系统,基于更新和半更新过程理论建立了以期望费用率最小为目标的非周期不完全检测下的维修优化模型。利用检测规划函数使系统的检测间隔期可以依据系统的状态进行调整。为更加贴近实际,模型考虑了测量误差对维修优化的影响。通过计算可以得到最优预防性维修阈值和检测方案。给出了模型的应用算例,通过对退化速率和费用参数的灵敏度分析说明了模型的有效性。与不考虑检测误差的优化结果进行对比,证明了考虑检测误差的重要性。     

基于TCP Veno算法的异构网络边界代理改进机制研究

针对在异构网络环境下无线网络的特点使TCP性能下降的问题,借鉴TCP Veno算法原理,提出基于网络状况评估的拥塞控制的性能增强代理改进工作机制,动态评估当前网络的状况,调整预应答发送窗口的阈值,提高性能增强代理在较大差异网络下的传输控制效能．基于NS-2的网络仿真测试对算法进行了验证,结果显示改进后的工作机制在丢包率较大的异构网络中性能有明显提升．     

基于灰色系统理论的军队油料节约人员360度绩效评估

基于360度绩效评估理论,从德、能、勤、绩4个方面构建了包含4个1级指标和16个2级指标的军队油料节约人员360度绩效评估指标体系。利用层次分析法确定了各评估指标的权重,运用灰色系统理论的原理和方法,构建了多层次灰色评估模型,并对某油料节约小组5位油料节约人员进行了360度绩效评估。案例应用表明,此方法具有系统综合各层评估者提供的"贫信息"的优点,不仅能得出被评估者的灰类及优劣顺序,而且得到了各指标所属灰类和具体的灰色评价权值,使360度绩效较好地实现了绩效反馈的目的。该研究为军队油料节约人员绩效管理工作提供了新的思路和方法。     

一种鲁棒的场地球类视频中的球门探测方法

在场地球类视频中,球门常常会伴随着精彩片段而出现。因此,球门探测是体育视频的基础,同时也是视频高层语义概念探测的研究热点。目前,利用机器学习方法进行视频中的对象探测是非常有前途的研究领域。基于此,提出一种基于模糊决策树的球门探测算法,用来探测场地球类视频中球门帧的出现,为了提高分类准确性,在模糊决策树训练的过程中加入了平衡处理。实验结果表明,与基于阈值和决策树的算法相比,该算法可以得到更好的分类结果(F-measure>95%),并且我们可以从所建立的树中推导出模糊规则来解释分类模型。     

基于删除平均的VI-CFAR检测器

针对MLVI-CFAR检测器在多个随机分布干扰目标背景下检测性能不好,以及OSVI-CFAR检测器在均匀环境和仅单滑窗存在干扰目标的环境下检测性能欠佳问题,提出了一种基于删除平均(CML)的VI-CFAR检测器(CMLVI-CFAR).仿真验证了CMLVI-CFAR检测器在均匀背景及多目标干扰环境下,都具有比MLVI和OSVI-CFAR检测器好的检测性能,并且其在杂波边缘背景下的虚警控制能力与前两者也基本一致,说明其在实际杂波环境中具有更强的鲁棒性和良好的应用潜力.     

利用深度学习的硬件计数器复用估计算法

利用深度学习方法,为硬件计数器复用(multiplexing,MPX)提供结果精度更高的估计模型。通过对MPX估计得到的结果与实际采集的真实数据进行相似性分析,证明相同程序多次运行之间得到的硬件计数值是线性相关的。采用神经网络多层感知器(multilayer perceptron,MLP)和双向门控神经网络(bidirectional gated recurrent unit, Bi-GRU)这2种深度学习模型,对MPX数据进行拟合。基于动态时间规整(dynamic time warping, DTW),提出一个全新的评估MPX数据精度的指标DTW-cost。实验结果表明,同时收集15个硬件事件数据时,MLP方法拟合得到的13个高性能计算应用平均准确率比现有使用最广的固定插值法高出10.53%,最多可提升19.8%;而在MLP表现较差的事件上,Bi-GRU方法得到的平均准确率提升了28.8%。     

Tuning the reactivity of Al–Ni by fine coating of halogen-containing energetic composites

In this paper, various core-shell structured Al–Ni@ECs composites have been prepared by a spray-drying technique. The involved ECs refer to the energetic composites (ECs) of ammonium perchlorate/nitrocellulose (AP/NC, NA) and polyvinylidene fluoride/hexanitrohexaazaisowurtzitane (PVDF/CL-20, PC). Two Al–Ni mixtures were prepared at atomic ratios of 1:1 and 1:3 and named as Al/Ni and Al/3Ni, respectively. The thermal reactivity and combustion behaviors of Al–Ni@ECs composites have been comprehensively investigated. Results showed that the reactivity and combustion performance of Al–Ni could be enhanced by introducing both NA and PC energetic composites. Among which the Al/Ni@NA composite exhibited higher reactivity and improved combustion performance. The measured flame propagation rate (v = 20.6 mm/s), average combustion wave temperature (T_max = 1567.0 °C) and maximum temperature rise rate (γ_t = 1633.6 °C/s) of Al/Ni@NA are higher than that of the Al/Ni (v = 15.8 mm/s, T_max = 858.0 °C, and γ_t = 143.5 °C/s). The enhancement in combustion properties could be due to presence of the acidic gaseous products from ECs, which could etch the Al₂O₃ shell on the surface of Al particles, and make the inner active Al to be easier transported, so that an intimate and faster intermetallic reaction between Al and Ni would be realized. Furthermore, the morphologies and chemical compositions of the condensed combustion products (CCPs) of Al–Ni@ECs composites were found to be different depending on the types of ECs. The compositions of CCPs are dominated with the Al–Ni intermetallics, combining with a trace amount of Al₅O₆N and Al₂O₃.