基于声矢量水听器阵列误差的自校正算法

对声矢量水听器阵列的各类误差进行了分类,推导了各类误差对阵列信号模型的影响因子,通过Monte Carlo实验分析对比了各类误差对阵列DOA估计性能的影响,然后将方向性误差和位置误差归结为幅度误差和相位误差,在传统声压阵列误差校正模型和算法的基础上,得到矢量阵列误差自校正的优化模型及自校正算法,最后,通过仿真实验和外场实验的数据处理表明,自校正算法具有良好的参数估计性能,具有一定的工程实用性.     

互相关函数在误差分离技术中的应用

在齿轮整体误差测量中，多头标准蜗杆的误差是影响齿形测量精度的一个主要误差。本文介绍应用互相关函数从齿形误差测量结果中分离出标准蜗杆误差的原理及其实现方法。     

岭回归估计的向量参数方法

本文提出岭回归估计的向量参数方法,选择均方误差函数的负梯度方向作为参数向量方向,根据均方误差与拟合误差的预期约束条件选择确定参数向量模长.文中获得了两个单调性结论,向量参数岭回归估计的均方误差是参数向量模长的单调减函数,而拟合误差是参数向量模长的单调增函数.基于两类误差的单调性结论,本文创建了关于两类误差的预期约束条件,预期条件约束下的向量参数岭回归方法有望成为兼备均方误差次优与拟合误差适度的双赢估计.文章最后是一个应用实例.     

带有变量选择过程的分类模型误差分析

偏倚一方差分析方法是在模型选择过程中权衡模型对现有样本解释程度和未知样本估计准确度的分析方法,目的是使选定的模型检验误差尽量小.在分类或回归过程中进行有效的变量筛选可以获得更准确的模型表达,但也会因此带来一定误差.提出"选择误差"的概念,用于刻画带有变量选择的分类问题中由于变量的某种选择方法所引起的误差.将分类问题的误差分解为偏倚—方差—选择误差进行研究,考察偏倚、方差和选择误差对分类问题的总误差所产生的影响.     

多工序制造过程加工误差的传播分析和控制

多工序制造过程在加工产品质量特征时出现的加工误差存在相互影响,因此为了保障产品的加工质量,定义了误差自修正能力指数来评价多工序制造过程对产品质量特征加工误差的补偿能力,并提出一种偏最小二乘回归方法应用于确定最后一道工序的加工误差与上游工序加工误差之间的关系,从而采取相应的调整措施来减少误差、保证最终的加工质量.     

伪凸集值映射的误差界

考虑了伪凸集值映射的误差界.证明了对于伪凸集值映射,局部误差界成立意味着整体误差界成立.通过相依导数,给出了伪凸集值映射存在误差界的一些等价叙述.     

线性模型中广义最小二乘参数估计误差分布的稳健性研究

研究了线性模型中广义最小二乘参数估计的误差分布稳健性问题.首先讨论了在线性统计模型里,设计矩阵为列降秩矩阵时,模型中给出了误差最大分布类,当误差向量的分布在此范围内变动时,LS估计和GLS估计在均方误差矩阵准则下是最优估计.然后进一步探讨广义最小二乘估计GLSE关于误差分布的稳健性,求出误差项所对应的最大分布族,进而证明了在该区间波动情况下,误差向量对应的始终为一致最优解.     

关于ε不敏感损失函数推广误差的界

本文研究了学习理论中推广误差的界的问题.利用ε不敏感损失函数的性质,分别获得r逼近误差和估计(样本)误差的界,并在特定的假设空间上得到了学习算法推广误差的界.     

对流扩散方程迎风有限元的自适应方法

本文对二维发展型对流扩散方程的迎风有限元格式给出了显式后验误差估计,证明了真实误差被后验误差估计器上下界定;并通过误差估计器建立了相应的自适应算法,数值例子表明了方法的有效性.     

误差绝对值的统计特征和应用

在误差服从正态分布的情况下,研究了误差绝对值的分布、数学期望和方差等统计特征,推导了高阶矩的计算公式;给出了利用残差绝对值对正态分布标准差进行估计的简便方法,残差的变异系数和分布情况可用于考察残差是否服从正态分布,利用误差限和误差平均值及正态分布标准差的关系可对模拟模型中的误差进行控制。     

Comparability Invariance Results for Tolerance Orders

We prove comparability invariance results for three classes of ordered sets: bounded tolerance orders (equivalent to parallelogram orders), unit bitolerance orders (equivalent to point-core bitolerance orders) and unit tolerance orders (equivalent to 50% tolerance orders). Each proof uses a different technique and relies on the alternate characterization.     

ContributionsTree-visibility orders

We introduce a new class of partially ordered sets, called tree-visibility orders, containing interval orders, duals of generalized interval orders and height one orders. We give a characterization of tree-visibility orders by an infinite family of minimal forbidden suborders. Furthermore, we present an efficient recognition algorithm for tree-visibility orders.     

Proper and Unit Trapezoid Orders and Graphs

We show that the class of trapezoid orders in which no trapezoid strictly contains any other trapezoid strictly contains the class of trapezoid orders in which every trapezoid can be drawn with unit area. This is different from the case of interval orders, where the class of proper interval orders is exactly the same as the class of unit interval orders.     

Unit and Proper Tube Orders

In 2005, we defined the n-tube orders, which are the n-dimensional analogue of interval orders in 1 dimension, and trapezoid orders in 2 dimensions. In this paper we consider two variations of n-tube orders: unit n-tube orders and proper n-tube orders. It has been proven that the classes of unit and proper interval orders are equal, and the classes of unit and proper trapezoid orders are not. We prove that the classes of unit and proper n-tube orders are not equal for all n ≥ 3, so the general case follows the situation in 2 dimensions.     

算子偏序的刻画及性质

本文定义了Hilbert空间上两个算子间的四种关系:星序、左星序,右星序及减序,使用了算子分块矩阵的方法,给出了两个算子具有上述四种关系之一时它们几何结构的刻画,证明了这四种关系是真正的偏序关系,进一步研究了它们之间的关系和性质.     

Line directionality of orders

Some orders can be represented by translating convex figures in the plane. It is proved thatN-free and interval orders admit such representations with an unbounded number of directions. Weak orders, tree-like orders and two-dimensional orders of height one are shown to be two- directional. In all cases line segments can be used as convex sets.     

Some positive dependence stochastic orders

In this paper we study some stochastic orders of positive dependence that arise when the underlying random vectors are ordered with respect to some multivariate hazard rate stochastic orders, and have the same univariate marginal distributions. We show how the orders can be studied by restricting them to copulae, we give a number of examples, and we study some positive dependence concepts that arise from the new positive dependence orders. We also discuss the relationship of the new orders to other positive dependence orders that have appeared in the literature.     

Decomposition of partial orders

A decomposition theory for partial orders which arises from the split decomposition of submodular functions is introduced. As a consequence of this theory, any partial order has a unique decomposition consisting of indecomposable partial orders and certain highly decomposable partial orders. The highly decomposable partial orders are completely characterized. As a special case of partial orders, we consider lattices and distributive lattices. It occurs, that the highly decomposable distributive lattices are precisely the Boolean lattices.     

Global Dependence Stochastic Orders

Two basic ideas, that give rise to global dependence stochastic orders, are introduced and studied. The similarities and differences between the resulting global dependence orders, and the known common positive dependence orders, are discussed. Some desirable properties that global dependence orders may expected to satisfy are listed and checked. Three particular global dependence orders, that come up from the two general ideas, are studied in detail. It is shown, among other things, how these orders can be verified. Finally, some applications in auction theory, in reliability theory, and in economics, are described.     

Order acceptance and due-date quotation in low machine rates

In real manufacturing environments, some customer orders include multiple jobs. However, a single due-date should be assigned to each order. Further, machine processing rate is not constant at all times. In effect, in many manufacturing operations, the machine processing rate decreases to a subnormal level during time and needs a special type of maintenance to bring the normal state back. Due to this reduction in capacity, production schedulers may decide to reject some orders. In this paper, the novel extensive problem of selecting a subset of orders, assigning due-dates to selected orders, scheduling the selected orders and jobs within each one, and scheduling the rate-modifying maintenance is studied. The objective function is minimizing total cost of lost-sales of rejected orders together with due-date length and maximum of earliness and tardiness of selected orders. The problem is proved polynomial and an optimal approach is developed.