多目标的分布式协同进化MDO算法

通过引入非优超排序和排挤的多目标处理机制 ,将分布式协同进化MDO算法的能力扩展到多目标的多学科设计优化问题。多目标的分布式协同进化MDO算法在保持各学科充分自治和各学科并行设计优化协同的基础上 ,通过一次运行即可获得具有良好分布的多个Pareto最优解 ,逼近整个Pareto最优前沿。应用于导弹气动 /发动机 /控制三学科两目标设计优化问题 ,与约束法计算结果的对比表明算法能够有效逼近该问题的Pareto最优前沿 ,为设计决策提供了丰富的信息     

稳态随机过程激励下基于首次超越破坏的结构优化设计

把结构系统动力可靠性分析与最优化设计结合起来 ,以结构系统的最小质量为目标函数 ,给出了考虑在平稳随机过程激励下多自由度线性系统总的可靠性的结构优化设计方法。运用谱分析理论 ,推导了结构系统在平稳随机过程激励下响应的统计特征 ,同时结合首次超越破坏的Possion模型计算结构系统的可靠性 ,最终采用广义乘子法得到结构系统设计变量的最优值。计算结果表明该方法是可行的     

Optimality of the greedy shooting strategy in the presence of incomplete damage information

Consider a situation where a single shooter engages, sequentially, a cluster of targets that may vary in terms of vulnerability and value or worth. Following the shooting of a round of fire at a certain target, the latter may either be killed or remain alive. We assume neither partial nor cumulative damage. If the target is killed, there is a possibility that the shooter is not aware of that fact and may keep on engaging that target. If the shooter recognizes a killed target as such, then this target is considered to be evidently killed. If the objective is to maximize the weighted expected number of killed targets, where the weight reflects the value of a target, then it is shown that a certain type of a shooting strategy, called a Greedy Strategy, is optimal under the general assumption that the more a target is engaged, but still not evidently killed, the less is the probability that the next round will be effective. If all weights are equal, then the greedy shooting strategy calls to engage, at each round, the least previously engaged target that is not evidently killed. © 1997 John Wiley & Sons, Inc. Naval Research Logistics 44: 613–622, 1997     

美军战略传播中的“叙事”策略

随着互联网技术的飞速发展,信息传播对国家政治、经济、外交和国防带来巨大冲击和影响。美军意识到传播叙事是政府和军队塑造形象、谋划战略、输出文化价值观的重要手段。美国从"重讲故事"的角度出发,将"叙事"包装成对外战略传播的新型方式,从思维认知、情感操控、舆论引导等方面强化对对手的心理威慑和影响,同时以新媒体为媒介扩大叙事传播的影响受众,进而达到物理打击所难以发挥的影响。     

Balancing a one‐way corridor capacity and safety‐oriented reliability: A stochastic optimization approach for metro train timetabling

In urban rail transit systems of large cities, the headway and following distance of successive trains have been compressed as much as possible to enhance the corridor capacity to satisfy extremely high passenger demand during peak hours. To prevent train collisions and ensure the safety of trains, a safe following distance of trains must be maintained. However, this requirement is subject to a series of complex factors, such as the uncertain train braking performance, train communication delay, and driver reaction time. In this paper, we propose a unified mathematical framework to analyze the safety‐oriented reliability of metro train timetables with different corridor capacities, that is, the train traffic density, and determine the most reliable train timetable for metro lines in an uncertain environment. By employing a space‐time network representation in the formulations, the reliability‐based train timetabling problem is formulated as a nonlinear stochastic programming model, in which we use 0‐1 variables to denote the time‐dependent velocity and position of all involved trains. Several reformulation techniques are developed to obtain an equivalent mixed integer programming model with quadratic constraints (MIQCP) that can be solved to optimality by some commercial solvers. To improve the computational efficiency of the MIQCP model, we develop a dual decomposition solution framework that decomposes the primal problem into several sets of subproblems by dualizing the coupling constraints across different samples. An exact dynamic programming combined with search space reduction strategies is also developed to solve the exact optimal solutions of these subproblems. Two sets of numerical experiments, which involve a relatively small‐scale case and a real‐world instance based on the operation data of the Beijing subway Changping Line are implemented to verify the effectiveness of the proposed approaches.     

加权模糊C均值聚类算法实现BDS三频组合观测值优选

在对BDS三频载波相位组合观测值进行误差分析的基础上,确定了优选载波相位线性组合系数的筛选标准。针对传统聚类算法在高维多频混合数据集分类中存在的不足,采用一种基于加权的模糊C均值聚类算法,通过对同一维度在不同簇上赋予不同的权重值,对传统遍历搜索法所获得的部分BDS三频载波相位组合观测值进行了优化分类选取,有效解决了传统全球导航卫星系统载波相位观测值选取方法效率低的问题,同时为多系统多频数据组合观测值系数的优化选取提供了一种新的思路。对分类结果进行分析,确定了各类组合观测量的适用范围,并结合实测数据,利用无几何层叠模糊度解算方法对优选组合进行了整周模糊度的解算,结果验证了该方法的可行性。     

雾霭图像实时处理算法加速策略

为仿真前视红外导引头实时匹配导引过程,完成目标实时跟踪性判断,针对限制对比度自适应直方图均衡和归一化互相关算法开展加速策略研究,提出基于统一计算设备架构(Compute Unified Device Architecture, CUDA)的雾霭图像实时处理加速方案,并进行设计参数寻优,得出线程块、线程网格设计参数存在最优尺寸,需实验测定的结论。实验结果表明,引入CUDA加速优化后,较未进行CUDA加速前时间指标提高5~20倍,能够满足目标跟踪实时性要求,可为红外导引头目标实时跟踪系统设计提供参考。     

辐射叶加载的短波宽带鞭状天线优化设计

提出一款具有多层辐射叶结构的新型鞭状天线,在10 m鞭状天线体上设置不同的辐射叶层数、半径、长度、分支数、仰角及其分布情况,研究其对天线辐射性能的影响,综合考虑选择一套最合适的辐射叶结构,并为天线进行双加载和宽带匹配网络的算法优化。仿真结果表明,与现有的普通宽带鞭状天线相比,增益和效率得到了普遍提高,在低频段增益最高提高了3 dB,效率最高提高了5%;在高频段增益最高提高了5 dB,效率最高提高了35%,方向图上翘也得到了一定的抑制,为改善现有的宽带鞭状天线提供一种新的结构设计方法。     

基于前景理论和证据推理的混合型多属性决策方法

针对属性权重完全未知的混合型多属性决策问题,提出一种基于前景理论和证据推理的决策方法。通过直觉模糊数对精确数、区间数和语言变量3种混合型属性的决策信息进行统一,根据前景理论对决策信息进行转化;提出基于直觉模糊熵与相似度的属性可靠性评估方法,结合属性重要度确定属性权重;采用证据推理算法集结属性信息,得到方案的综合前景值,并以此进行方案排序。算例分析结果表明,所提方法具有较强的区分能力,能够有效降低决策结果的不确定性,对混合型多属性决策问题具有较好的适用性。     

电磁发射系统以太网拓扑结构建模优化

拓扑结构优化模型是网络中的一类重要模型,可以有效地优化系统整体链路性能。针对电磁发射系统以太网拓扑结构中部分节点和链路的负载过大,一旦发生堵塞,将会影响网络中关键链路性能的问题,建立网络拓扑结构优化的多目标规划模型,并提出一套基于基因环操作的遗传算法对其进行求解,通过仿真得出最优的网络拓扑结构。根据仿真结果修改实际网络节点默认配置参数,结果显示该模型和算法能有效均衡负载,降低网络冲突率,且不需要改变电磁发射系统以太网的物理链路,不会增加额外成本,对电磁发射类系统具有普适性的意义。