Design of accelerated life testing plans under multiple stresses

Accelerated life testing (ALT) using multiple stresses is commonly used in practice to resemble the operating stresses at normal operating conditions and obtain failure observations in a much shorter time. However, to date, there is little research into the theory of planning ALT for reliability estimation with multiple stresses. ALT with multiple stresses can result in a large number of stress‐level combinations which presents a challenge for implementation. In this article, we propose an approach for the design of ALT plans with multiple stresses and formulate multistress test plans based on different objectives and practical constraints. We develop a simulated annealing algorithm to efficiently determine the testing plan parameters. We demonstrate the proposed method with examples based on an actual test conducted using three stress types. The obtained optimal test plans are compared with those based on fractional factorial design. © 2013 Wiley Periodicals, Inc. Naval Research Logistics 60: 468–478, 2013     

多无人机梯次攻击航路规划

为了解决多无人机协同攻击航路规划问题,基于一种雷达威胁等效方法,以及无人机在不同姿态下雷达散射截面RCS(Radar Cross-Section)值随之改变的特性,结合多机作战战术思想,提出了一种多无人机梯次协同攻击同一目标的方法,并利用Dijkstra算法进行多无人机航路规划.仿真结果表明,该方法具有更高的应用价值.     

多传感器目标识别和火力分配群决策算法

针对多传感器目标识别的群决策问题,考虑到多目标威胁程度评价指标属性的模糊性,首先将多传感器目标评价属性的定量、定性描述指标统一转化为三角模糊数,充分利用三角模糊数性质构建群决策函数,实现定量指标与定性指标的规范化处理,在此基础上提出一种基于加权矩阵排序的三角模糊数多传感器目标识别的群决策算法,实现对多传感器目标的威胁程度的综合评估.最后,通过多传感器目标评估实例分析并证明了算法的可行性和有效性.同时实验发现该算法对于多属性、多评价者的群体决策具有优势,且不易受外界参数变化的影响.     

基于多假设跟踪弹道导弹主动段跟踪算法

针对弹道导弹主动段防御中多枚弹道导弹同时跟踪问题,提出了基于多假设思想的主动段跟踪算法.重点阐述了该算法中假设生成、假设概率计算、假设约简以及假设剪枝等环节.从工程实用的角度出发,采用求解一个线性分配问题(LAP)方法得到M个最优假设,大大减少了假设数量,并运用N-scan回溯剪枝方法对假设进行剪枝,确定要输出的航迹,提高了算法的效率和实用性.仿真实验表明,该算法能够对主动段多枚弹道导弹目标准确关联跟踪.     

面向多最优解组合优化问题的决策求解算法

针对具有固定物品总和、多最优解特征的组合优化问题,以固定总和实数子集问题和购买鸡翅问题为例,给出了这类多最优解组合优化问题的形式化表示。在分析枚举等经典算法基础上,提出了基于整数状态表示和实数状态表示的0-1决策递归搜索多最优解动态规划算法。针对该算法在最优解数量较大时,时间复杂度趋向O(mⁿ)的问题,提出了基于相同决策路径合并和基于0-x决策的两种改进算法。实验中两种改进算法的计算时间基本符合与O(nb+nm)的正比关系,表明对于这类多最优解组合优化问题具有良好的求解性能。     

灰色关联投影法的威胁排序算法

威胁评估与排序是作战系统和C4I系统必须具备的重要功能.结合多指标决策理论和灰色系统理论,提出了基于多指标决策灰色关联投影法的威胁排序算法.首先通过建立灰色关联判断矩阵将威胁排序转化为多指标决策问题;其次为缩小极端值的影响提出了基于信息熵的威胁指标权重算法;最后从矢量投影角度应用灰色关联投影值进行威胁排序;并分别研究了分辨系数和指标权重对威胁排序的影响.通过现代防空作战实例验证了算法的有效性和工程实用性.     

一种时空数据流中范围聚集查询共享策略

根据查询谓词的相似性来实现计算共享是数据流查询优化的主要策略之一,主要针对时空数据流中基于滑动窗口的范围聚集问题进行查询优化,采用时间片段和空间片段分别描述滑动窗口之间和空间范围谓词之间的重叠。提出基于时空分片技术将时空数据流元组分成互不重叠的时空片段。范围聚集查询首先基于时空片段生成中间结果,然后根据查询条件与时空片段的映射关系生成查询结果。实验表明,提出的时空分片策略能大大提高时空数据流中范围聚集查询的性能。     

An MX/G/1 queueing system with disasters and repairs under a multiple adapted vacation policy

We consider a queueing system with batch Poisson arrivals subject to disasters which occur independently according to a Poisson process but affect the system only when the server is busy, in which case the system is cleared of all customers. Following a disaster that affects the system, the server initiates a repair period during which arriving customers accumulate without receiving service. The server operates under a Multiple Adapted Vacation policy. The stationary regime of this process is analyzed using the supplementary variables method. We obtain the probability generating function of the number of customers in the system, the fraction of customers who complete service, and the Laplace transform of the system time of a typical customer in stationarity. The stability condition for the system and the Laplace transform of the time between two consecutive disasters affecting the system is obtained by analyzing an embedded Markov renewal process. The statistical characteristics of the batches that complete service without being affected by disasters and those of the partially served batches are also derived. © 2015 Wiley Periodicals, Inc. Naval Research Logistics 62: 171–189, 2015     

多准则下反导作战单、多遭遇点预测模型

整个反导射击过程可以简单看作一个基于预测的遭遇点起始的,不断重复、修正的过程。分析遭遇点预测的时间和空间前提,并给出遭遇点预测的3个基本准则;在多准则下建立了单遭遇点预测模型,给出了模型公式中未知量的转化方法;分射击-观察-射击和射击-射击两种情况建立了多准则下多遭遇点预测和优化模型,并给出了拦截弹最晚发射时间的计算方法;就观察时机对遭遇点预测的影响进行了分析与建模。部分研究属于探索性的研究,相关结论对指控模型开发和实施连续反导,从方法和作战理念方面提供了一些参考。     

面向通信信号调制识别的半监督生成对抗网络框架

针对只有少量标签数据的弱监督条件下现有调制信号识别模型准确率较低的问题,提出基于生成对抗网络的半监督学习框架。该方法通过对通信信号进行冗余空域变换,使其在适应生成对抗网络模型的同时保留丰富的信号相邻特征;通过梯度惩罚Wasserstein生成对抗网络的引入,构建适宜电磁信号处理的半监督学习框架,实现对无标签信号样本的有效利用。为了验证所提算法的有效性,在RADIOML 2016.04C数据集上进行测试。实验结果表明,该方法在半监督条件下能训练出高效的分类器,获得优异的调制识别结果。