具有策略偏好的模糊双矩阵对策模型及其应用

针对局中人对策略有偏好、支付值模糊的双矩阵对策给出了一种具体的求解方法,首先根据加权算子集结得到局中人的策略偏好向量、利用模糊数排序方式将模型转化为清晰的双矩阵对策,再利用粒子群优化算法求解.最后以雷达电子对抗为例,建立了具有策略偏好的模糊双矩阵对策的作战效能评估模型,研究结果对于双方资源分配,提高作战效能具有一定的军...     

基于二人有限零和对策的防空兵火力分配方法

为了优化防空作战过程中火力分配问题,使射击达到最佳效果,运用对策论、线性规划等理论方法,采用对策矩阵建立了防空火力分配的线性规划模型,并通过计算示例和计算机仿真初步预测了敌方的空袭兵器使用情况和我方相应的兵力分配对策。该模型建立的防空火力分配方法较好地满足了要地防空装备的战术应用问题,对提高作战效能具有一定的参考价值。     

An inspection game: Taking account of fulfillment probabilities of players' aims

This paper deals with an inspection game of customs and a smuggler. The customs can take two options of assigning a patrol or not. The smuggler has two strategies of shipping its cargo of contraband or not. Two players have several opportunities to take actions during a limited number of days. When both players do, there are some possibilities that the customs captures the smuggler and, simultaneously, the smuggler possibly makes a success of the smuggling. If the smuggler is captured or there remain no days for playing the game, the game ends. In this paper, we formulate the problem into a multi‐stage two‐person zero‐sum stochastic game and investigate some characteristics of the equilibrium solution, some of which are given in a closed form in a special case. There have been some studies so far on the inspection game. However, some consider the case that the smuggler has only one opportunity of smuggling or the perfect‐capture case that the customs can certainly arrest the smuggler on patrol, and others think of a recursive game without the probabilities of fulfilling the players' purposes. In this paper, we consider the inspection game taking account of the fulfillment probabilities of the players' aims. © 2006 Wiley Periodicals, Inc. Naval Research Logistics, 2006     

Capacity allocation with traditional and Internet channels

In this paper we study a capacity allocation problem for two firms, each of which has a local store and an online store. Customers may shift among the stores upon encountering a stockout. One question facing each firm is how to allocate its finite capacity (i.e., inventory) between its local and online stores. One firm's allocation affects the decision of the rival, thereby creating a strategic interaction. We consider two scenarios of a single‐product single‐period model and derive corresponding existence and stability conditions for a Nash equilibrium. We then conduct sensitivity analysis of the equilibrium solution with respect to price and cost parameters. We also prove the existence of a Nash equilibrium for a generalized model in which each firm has multiple local stores and a single online store. Finally, we extend the results to a multi‐period model in which each firm decides its total capacity and allocates this capacity between its local and online stores. A myopic solution is derived and shown to be a Nash equilibrium solution of a corresponding “sequential game.” © 2006 Wiley Periodicals, Inc. Naval Research Logistics, 2006     

The Banzhaf value and communication situations

Myerson 3 and 4 proposed and characterized a modification of the Shapley value in the context of communication situations. In this paper we propose several characterizations of the extension of the Banzhaf value proposed by Owen 7 to communication situations. In particular we characterize this value with the properties of fairness, isolation, and pairwise merging. © 2006 Wiley Periodicals, Inc. Naval Research Logistics, 2006     

落点散布鉴定中Bayes序贯截尾方法的运用

本文研究再入飞行器落点散布的鉴定方法。构造了 Ｂａｙｅｓ 序贯截尾方案；分析了可能冒的风险，给出了计算方法； 对于 Ｂａｙｅｓ 方法运用中的关键问题——验前信息的运用及验前概率的计算， 作了较详细的论述。文中提出的方法， 适用于小子样场合下的试验鉴定问题。     

序列近似优化方法

随着工程优化中仿真模型精度和计算时间的不断提高,常规的智能优化方法难以在可接受的计算代价中得到最优解。序列近似优化方法通过将近似模型技术引入优化过程,并采用不断更新采样点的方法来指导寻优,在基于计算耗时模型的优化中得到了越来越广泛的应用。通过论述序列近似优化方法的若干关键技术及其发展现状,可有效指导其在工程优化中的应用,并给出了序列近似优化方法可能的改进方法及发展趋势。     

二维非线性正交各向异性材料的瞬态热传导反问题数值方法

针对正交各向异性材料的二维非线性热传导反问题,采用顺序函数法进行表面热流辨识问题的研究。在求解反问题时,采用有限体积法、牛顿-拉夫逊法并引入未来时间步的概念。在每个时间步内,将待辨识热流视为非线性方程组的未知量,通过一个迭代过程进行求解。算例的研究表明,热流辨识结果与真实热流相近,从而证明了本方法在辨识二维非线性热传导反问题时是准确、稳定、有效的。     

基于合作博弈的智能集群自主聚集研究

智能集群是指由大量智能体构成、采用自组织策略协作完成任务的群体。以无人车集群系统协同监视再入体着靶过程为任务背景,开展智能集群自组织策略的关键技术研究。设计了无人车集群执行再入体着靶协同监视的集群行为模式,提出了基于合作博弈的智能集群自组织策略,各智能体以实现群体聚集为“合作目标”,以降低自身能量消耗为“竞争目标”,开展博弈,基于微粒群算法规划局部路径,最终使群体系统涌现出聚集行为。仿真实验验证了设计的自主聚集策略的有效性。     

基于SPRT的组合导航故障检测与隔离互辅算法

针对SPRT(序贯概率比检验)算法无法准确判断缓变故障结束时间的问题,提出了一种故障检测与隔离互辅算法。采用SPRT算法对滤波残差进行故障检测,确认故障后隔离故障观测量,重新建立卡尔曼滤波观测方程并完成量测更新,将隔离后的滤波结果作为下一时刻的滤波初值辅助故障检测。仿真结果表明,该方法能准确定位故障观测量并判断出故障的结束时间,提高了系统可靠性。