基于未确知测度的多目标决策方法及其应用

定义策略、目标和评判三个空间以及满足非负性、归一性和可加性的未确知测度函数,给出了适合于有序分割类评判空间寻优的最优决策准则,从而提出了一种未确知信息数学处理理论基础上的未确知测度决策方法,对含多因素特别是难以量化因素的决策问题寻优建立了一个合理、精细、分辨率高、计算简单高效的数学模型,用来求解多目标决策问题中的最优策略,并结合火力打击选择的一个实例分析,获得了较为成功的应用.     

二代光点式坦克火控系统的校准操作

通过分析坦克射击准备误差因素的地位变化情况,得出了火控系统的校准误差是影响首发命中关键因素的结论。在此基础上,重点阐述了二代光点式坦克火控系统的校准操作程序和方法,以满足部队科技练兵的需要。     

三维信息提取系统研究

建立了从两幅图像提取三维信息的数学模型,应用广义逆法,求解三维坐标的超定线性方程组.推导了通过摄像机校准来确定摄像机参数的数学过程.最后通过图像测量实验验证了以上算法的正确性.     

基于虚拟仪器的导弹综合校准系统研制

简要介绍了虚拟仪器技术的发展,并以某导弹校准系统为例,阐述了基于虚拟仪器的装备计量校准系统的软硬件研制。     

蚁群算法在光电载荷标校中的应用

针对空中平台光电载荷安装误差的梗准问题,提出了一种利用蚁群算法实现动态标校的方法.通过光电裁荷测量得到的目标方位角、俯仰角和平台导航信息,结合GPs转化得到的方位角、俯仰角真值,建立了基于最优估计理论的目标函数,并利用蚁群算法求得了系统的安装误差.通过蒙特卡洛模拟仿真,验证了该方法能够较为准确地标定出安装误差;通过校准,有效避免了由安装误差引起的结果有偏,使得目标跟踪结果收敛至真值附近.     

舰船最优概念设计方案的构造方法研究

在舰船概念设计阶段,设计者往往需要从大量设计方案中筛选优秀方案,此问题属于多属性多目标优化的问题,而进化算法的兴起为此类问题的解决提供了较好的途径.针对进化算法中Pareto最优解集的特点,结合舰船设计实际,提出一种新的构造最优解集的方法--提取法.案例验证表明,提取法是合理和有效的.     

基于辐照度标准传递的辐亮度校准系统

介绍了一种新型辐亮度校准系统,通过溯源至国家辐照度标准,构建了军用光辐射校准平台。对系统的测量不确定度分析表明,该系统可以满足装备计量保障对测试不确定度比的通用要求,适用于装备专用检测设备的辐亮度校准。     

系统标定的一种新方法

以2008年高教社杯全国大学生数学建模竞赛的A题为研究背景,设计了利用数码相机进行系统标定的一种新方法,建立了相应的数学模型,并设计遗传算法实现了模型的快速求解,算例表明了模型及算法的有效性。     

Coordinated dynamic control of marketing and production

We study the dynamic profit maximization problem for a firm exercising control on both marketing and production. The firs marketing effort impacts the current‐period demand, which in turn affects future demand in a dissipating fashion. Under linear‐cost and zero‐leadtime assumptions, we show that the firm should follow base‐point rules for both marketing and production, whereas trends of the base points reflect a certain complementarity between marketing and production. We obtain comparable results when marketing costs are convex. Our computational study identifies conditions under which simple fixed‐marketing‐effort and fixed‐marketing‐target heuristics would perform well. © 2009 Wiley Periodicals, Inc. Naval Research Logistics 2009     

Optimal design of accelerated life testing plans for periodical replacement with penalty

Accelerated life testing (ALT), enhanced by optimal test plans, has been widely accepted in practice as a quick approach for estimating the reliability of a product. From the estimation result, preventive maintenance schedules can be determined to ensure the performance of the product under its normal operating conditions. By default, maintenance decision‐making is regarded as the last and least rewarding step. However, sometimes the maintenance schedules, such as preventive maintenance intervals, are predetermined due to customer concerns and/or by various mandatory regulations and rules. Under such circumstances, how to accurately estimate the expenditure (e.g., on maintenance or spare parts management) associated with these maintenance requirements becomes an important issue. A viable solution is to incorporate the maintenance requirements into ALT plans. This paper provides an approach for the optimal design of ALT plans oriented by a mandatory periodical replacement schedule subject to a discounted penalty. The objective is to improve the estimation accuracy of the economic impact of this maintenance requirement. A numerical experiment is provided to demonstrate the approach in practical use. © 2008 Wiley Periodicals, Inc. Naval Research Logistics 2009