基于RSSI线性回归分析的无线传感器网络定位方法

在无线传感器网络（WSN）定位应用中,针对接收信号强度指示（RSSI）容易受到环境的影响不能实现准确测距的问题,提出了一种基于RSSI线性回归分析的定位方法。该方法通过信号衰减模型和线性回归理论相结合,修正实际环境下每个锚节点的测距模型,同时利用相关系数和剩余标准差对测距模型进行评估,制定出更加合理的定位策略。实验表明,采用修正的测距模型和新的定位策略,使得节点定位精度明显提高。     

预制缺陷柱壳结构的等效裂纹分析方法

针对传统有限元法在分析预制缺陷柱壳时出现的应力/应变计算精度不足、无法有效指导工程实践的问题,提出了一种具有高冗余度的等效裂纹分析方法。将预制缺陷视作典型裂纹,基于奇异裂纹元法对其进行裂纹稳定性分析,并根据稳定性分析结果来评估预制缺陷柱壳的结构完整性。通过该方法得到的分析结果具有比常规有限元分析更高的可靠性,对于导弹线式爆炸分离装置等工作环境恶劣、风险系数较高的预制缺陷柱壳结构而言,该方法有助于提升其设计的安全裕度,具有一定的工程实用性。     

军用飞机燃油消耗多元线性回归模型

为了解决军用飞机非标准动作燃油消耗建模问题,利用实际飞参数据,采用多元线性回归方法进行建模分析。建立了燃油消耗量与飞行时间、油箱平均油量、平均高度、平均速度、T1温度、T4温度、N2转速、高度变化量、飞机外挂的多元线性回归模型,并对回归模型进行统计性检验和验证。结论显示:模型拟合度较好,符合军用飞机飞行耗油规律,对军用飞机非标准动作燃油消耗规律研究具有重要意义。     

Sequential learning versus no learning in Bayesian regression models

We consider two regression models: linear and logistic. The dependent variable is observed periodically and in each period a Bayesian formulation is used to generate updated forecasts of the dependent variable as new data is observed. One would expect that including new data in the Bayesian updates results in improved forecasts over not including the new data. Our findings indicate that this is not always true. We show there exists a subset of the independent variable space that we call the “region of no learning.” If the independent variable values for a given period in the future are in this region, then the forecast does not change with any new data. Moreover, if the independent variable values are in a neighborhood of the region of no learning, then there may be little benefit to wait for the new data and update the forecast. We propose a statistical approach to characterize this neighborhood which we call the “region of little learning.” Our results provide insights into the trade‐offs that exist in situations when the decision maker has an incentive to make an early decision based on an early forecast versus waiting to make a later decision based on an updated forecast. © 2014 Wiley Periodicals, Inc. Naval Research Logistics 61: 532–548, 2014     

基于渐进性的线性方程组教学设计的研究

针对线性方程组传统教学中的一些问题,给出了引入线性方程组、解线性方程组、分析解结构的渐进性三步教学设计,同时给出了maple求解线性方程组的两种方法,将理论与实践初步结合了起来,丰富了教学手段,激发了学生的学习兴趣。     

Quadratic bottleneck problems

We study the quadratic bottleneck problem (QBP) which generalizes several well‐studied optimization problems. A weak duality theorem is introduced along with a general purpose algorithm to solve QBP. An example is given which illustrates duality gap in the weak duality theorem. It is shown that the special case of QBP where feasible solutions are subsets of a finite set having the same cardinality is NP‐hard. Likewise the quadratic bottleneck spanning tree problem (QBST) is shown to be NP‐hard on a bipartite graph even if the cost function takes 0–1 values only. Two lower bounds for QBST are derived and compared. Efficient heuristic algorithms are presented for QBST along with computational results. When the cost function is decomposable, we show that QBP is solvable in polynomial time whenever an associated linear bottleneck problem can be solved in polynomial time. As a consequence, QBP with feasible solutions form spanning trees, s‐t paths, matchings, etc., of a graph are solvable in polynomial time with a decomposable cost function. We also show that QBP can be formulated as a quadratic minsum problem and establish some asymptotic results. © 2011 Wiley Periodicals, Inc. Naval Research Logistics, 2011     

非均匀线阵天线阵因子综合方法的研究

非均匀线阵的阵因子综合与均匀线阵有很大不同,使得天线阵因子的综合无法利用经典的方法进行,副瓣电平降低困难。提出了一种针对线阵阵元空缺情况下的阵因子综合方法。从理论推导到Matlab仿真,对非均匀线阵天线阵因子综合方法进行了研究,得到一种适用于非均匀线阵天线阵因子的综合方法。     

线性规划问题新解——改进大M法

借鉴两阶段法的求解思路,在用单纯形法求解线性规划问题时,对大M法进行改进,提出一种新的算法.这种改进后的算法可以有效克服原来两种算法的不足,既能降低理解难度,又能提高算法的效率,保证算法的全局收敛性.     

军队人力资源培训问题的整数线性规划模型

文章研究了军队人力资源培训问题,并基于时间和费用两个指标,建立了一个满足培训时间约束且费用最省的0-1整数线性规划模型,给出了基于Lagrange松驰分解的模型求解算法。在算法中,采用一种简单可行的Lagrange乘子更新方法代替传统的次梯度法。另外,文章证明了算法获得最优解的两个充分条件,计算实例初步表明给出的算法是行之有效的。     

Inducing coordination in supply chains through linear reward schemes

Decentralized decision‐making in supply chain management is quite common, and often inevitable, due to the magnitude of the chain, its geographical dispersion, and the number of agents that play a role in it. But, decentralized decision‐making is known to result in inefficient Nash equilibrium outcomes, and optimal outcomes that maximize the sum of the utilities of all agents need not be Nash equilibria. In this paper we demonstrate through several examples of supply chain models how linear reward/penalty schemes can be implemented so that a given optimal solution becomes a Nash equilibrium. The examples represent both vertical and horizontal coordination issues. The techniques we employ build on a general framework for the use of linear reward/penalty schemes to induce stability in given optimal solutions and should be useful to other multi‐agent operations management settings. © 2005 Wiley Periodicals, Inc. Naval Research Logistics, 2006