模糊多因素多层次评判在目标威胁度中的应用

目标威胁度评估是目标分配的重要依据,是区分目标的轻、重、缓、急为发射决策提供依据的保证。结合模糊优选理论和AHP法,以某新型地空导弹武器系统为依托,建立相应的威胁度评估模型,为火力单元级自动化作战指挥系统的目标威胁度评估提供了一种简捷有效的方法。     

A comprehensive review on material selection for polymer matrix composites subjected to impact load

Polymer matrix composites (PMC) are extensively been used in many engineering applications. Various natural fibers have emerged as potential replacements to synthetic fibers as reinforcing materials composites owing to their fairly better mechanical properties, low cost, environment friendliness and biodegradability. Selection of appropriate constituents of composites for a particular application is a tedious task for a designer/engineer. Impact loading has emerged as the serious threat for the composites used in structural or secondary structural application and demands the usage of appropriate fiber and matrix combination to enhance the energy absorption and mitigate the failure. The objective of the present review is to explore the composite with various fiber and matrix combination used for impact applications, identify the gap in the literature and suggest the potential naturally available fiber and matrix combination of composites for future work in the field of impact loading. The novelty of the present study lies in exploring the combination of naturally available fiber and matrix combination which can help in better energy absorption and mitigate the failure when subjected to impact loading. In addition, the application of multi attributes decision making (MADM) tools is demonstrated for selection of fiber and matrix materials which can serve as a benchmark study for the researchers in future.     

Mitigation effects on the reflected overpressure of blast shock with water surrounding an explosive in a confined space

The mitigation of blast shock with water has broad application prospects. Understanding the mitigation effects on the reflected overpressure of the explosion shock with water surrounding an explosive in a confined space is of great significance for military explosives safety applications. To estimate the effects of the parameters on the reflected overpressure of blasted shock wave, a series of experiments were carried out in confined containers with spherical explosives immersed in a certain thickness of water, and numerical simulations were conducted to explore the corresponding mechanisms. The results reveal that the reflected overpressure is abnormally aggravated at a small scaled distance. This aggravation is due to the high impulse of the bulk accelerated water shell converted from the explosion. With increasing scaled distance, the energy will be gradually dissipated. The mitigation effects will appear with the dispersed water phase front impacting at a larger scaled distance, except in the case of a dense water phase state. A critical scaled distance range of 0.7—0.8 m/kg1/3 for effective mitigation was found. It is suggested that the scaled distance of space walls should be larger than the critical value for a certain water-to-explosive weight ratio range (5—20).     

系统可靠性模糊分配的熵权方法

采用模糊数来表示不同模糊因素的相对强度,即通过指标得分的比较,用三角模糊数来表示m个模糊因素的权重系数,建立模糊判断矩阵,根据α截集和决策者对模糊判断矩阵的满意乐观程度指标λ进行模糊区间运算,得出各分系统可靠性指标所占的熵权,再根据熵权对系统可靠性指标进行分配。这种方法可以正确把握处理模糊问题的尺度,减少人为干扰。     

飞机空-空作战效能加权模糊综合评估

飞机空-空作战效能是飞机空中作战能力的综合反映,也是评估敌我双方空战实力的主要指标。针对飞机结构复杂和空战涉及因素较多等特点,建立了两级递阶结构模型,利用层次分析法计算了飞机空-空作战中各品质因素的权值,并给出了进行一致性检验的方法。根据飞机空战效能指标多样性的特点,针对飞机不同品质因素运用模糊数学理论建立了的模糊隶属度函数,并对几种飞机的空-空作战效能进行了评估,充分验证了评估方法的有效性。     

LiDAR点云纹理特征提取方法

针对图像纹理应用于LiDAR点云分类过程中存在的多义性问题,提出点云纹理特征的概念。该特征属性反映了点与其邻域点的属性值分布情况,提取过程基于KD树数据检索结构和灰度共生矩阵算法。分析搜索邻域、移动步长和灰度等级等参数对点云纹理特征的影响,并利用支持向量机分类方法验证点云纹理特征,可以有效地辅助高程和强度信息以改善LiDAR点云的地物分类结果。实验还证明了相比于栅格格式的图像纹理特征,点云纹理特征约束的地物分类具有更高的分类精度,并且点云纹理特征在微小地物的甄别和水陆的区分方面具有突出的能力。该特征的这些优秀特性可以为海岸带机载LiDAR数据的精细化分类、海岸带高精度DEM构建和海岸线提取等工作发挥重要作用。     

基于反正切函数的测试性分配方法

针对现有主流测试性分配方法存在的分配指标过低和过高等不合理问题,分析认为其根源在于现有方法采用线性分配函数,而该函数不符合测试性指标提升规律,为此,构造了一种基于反正切函数的测试性分配函数,设计故障检测率与故障隔离率指标的分配算法,开展案例应用并与经典故障率分配法和综合加权分配法进行对比分析,结果表明随着分配权重(故障率)的变大,本方法分配指标的增幅逐渐减小,不会出现极低指标和超1指标的不合理情形。     

基于单类集成分类器的JPEG通用隐写分析

针对JPEG图像通用隐写检测问题,提出了一种基于单类集成分类器的新方法。算法提取图像DCT块内、块间和小波层内、层间的共生特征以及小波子带系数的直方图特征对图像进行描述,并计算检测图像及其原始估计图像所提共生矩阵和直方图分布的对称交互熵作为隐写分析特征;然后,随机构造若干个特征子空间,利用bootstrap方法构造载体图像训练子集,分别进行单类训练得到数个基分类器;最后,将基分类器的分类结果按多数投票法进行融合作为单类集成分类器的分类结果。实验结果表明,单类集成分类器能显著提高算法的检测效果,而且,本方法相比基于多超球面OC-SVM分类器的单类隐写分析方法,具有更高检测率。     

基于shearlet与PCNN的多聚焦图像融合方法

针对传统脉冲耦合神经网络(PCNN)模型在多聚焦图像融合领域应用中面临的参数繁杂等问题,提出一种基于剪切波(shearlet)变换与改进型PCNN的多聚焦图像融合方法。相比以往的变换域方法,shearlet具有理想的图像信息捕捉性能以及较低的计算复杂度,因此,可利用shearlet将待融合图像进行多尺度多方向分解。其次,对经典PCNN模型加以改进,综合运用清晰度水平以及协调矩阵完成低频子带图像以及一系列高频子带图像的融合过程。最后,运行shearlet反变换得到最终融合图像。仿真实验选取了若干组待融合图像进行仿真,验证了该方法在主、客观评价两方面的优越性。     

权重信息不完全的机载雷达作战效能评估模型

为解决指标权重信息不完全时的机载雷达作战效能评估问题,提出了一种评估模型。对专家关于指标权重给出的残缺判断矩阵,给出了其一致性评判标准。结合残缺判断矩阵的最佳一致性比例,构造了一个单目标优化模型,用于计算指标权重。在此基础上提出了权重信息不完全的机载雷达作战效能评估模型。最后给出一个实例验证了评估模型的有效性。