非均匀对角减载最小方差无失真响应多目标分辨

针对最小方差无失真响应(Minimum Variance Distortionless Response,MVDR)方法在起伏非相关噪声环境下多目标分辨性能严重下降的问题,提出一种非均匀对角减载MVDR (Inhomogeneous Diagonal Unloading MVDR,IDU-MVDR)方法。该方法首先对协方差矩阵进行非均匀对角减载,然后实施MVDR方法。各阵元上的对角减载量通过求解半正定优化问题获得,优化问题中最大化减载量之和,但约束减载后协方差矩阵的最小特征值是一个较小的正值。数值仿真表明,IDUMVDR方法可通过非均匀对角减载消除大部分非相关噪声,但保留小部分噪声分量.因此IDU-MVDR方法较MVDR方法分辨力更高,空间谱中背景级更低、弱目标谱峰更加明显,并且具备一定的稳健性.海上实验结果与数值仿真相一致,验证了IDU-MVDR方法的有效性.     

A diagonal measure and a local distance matrix to display relations between objects and variables

Proper permutation of data matrix rows and columns may result in plots showing striking information on the objects and variables under investigation. To control the permutation first, a diagonal matrix measure D was defined expressing the size relations of the matrix elements. D is essentially the absolute norm of a matrix where the matrix elements are weighted by their distance to the matrix diagonal. Changing the order of rows and columns increases or decreases D. Monte Carlo technique was used to achieve maximum D in the case of the object distance matrix or even minimal D in the case of the variable correlation matrix to get similar objects or variables close together. Secondly, a local distance matrix was defined, where an element reflects the distances of neighboring objects in a limited subspace of the variables. Due to the maximization of D in the local distance matrix by row and column changes of the original data matrix, the similar objects were arranged close to each other and simultaneously the variables responsible for their similarity were collected close to the diagonal part defined by these objects. This combination of the diagonal measure and the local distance matrix seems to be an efficient tool in the exploration of hidden similarities of a data matrix. Copyright © 2009 John Wiley & Sons, Ltd.     

幻方的几个充要条件

建立等重方阵的概念,并运用它阐述幻方、2次幻方、3次幻方及中心对称互补幻方的充要条件,根据这些结论构作出一些8阶、9阶2次中心对称互补幻方.     

The Bott-Borel-Weil Theorem for direct limit groups

We show how highest weight representations of certain infinite dimensional Lie groups can be realized on cohomology spaces of holomorphic vector bundles. This extends the classical Bott-Borel-Weil Theorem for finite-dimensional compact and complex Lie groups. Our approach is geometric in nature, in the spirit of Bott's original generalization of the Borel-Weil Theorem. The groups for which we prove this theorem are strict direct limits of compact Lie groups, or their complexifications. We previously proved that such groups have an analytic structure. Our result applies to most of the familiar examples of direct limits of classical groups. We also introduce new examples involving iterated embeddings of the classical groups and see exactly how our results hold in those cases. One of the technical problems here is that, in general, the limit Lie algebras will have root systems but need not have root spaces, so we need to develop machinery to handle this somewhat delicate situation.

     

Disappearing Wakes and Dispersion in Numerically Simulated Flows Through Tube Bundles

Flow through a staggered array or bundle of parallel rigid cylinders of diameter D is computed with the help of a three-dimensional direct numerical simulation (DNS) at various values of Reynolds number between 50 and 6000. Two different spacings L of the tubes, i.e. L/D= 2 and L/D= 3, have been considered. When Re 500 the flow is laminar. In that case the converging flow between a pair of adjacent cylinders brings the oppositely signed vorticity at the two edges of the wake closer together behind the upstream cylinder so that the vorticity decreases quickly due to cancellation by diffusion. At Re 6000, when the flow is highly turbulent, the wake vorticity disappears rather by turbulent diffusion. This disappearance of the wakes in the closely packed flows (i.e. L/D 2) causes the mean flow in a cell, which consists of the region around a single cylinder, to be effectively independent of that in other cells. Another consequence is that the mean velocity field can be very well approximated by potential flow except in a thin boundary layer along the cylinder and a short wake behind it. The results have been applied to the transport of scalars in closely packed arrays. As in other complex flows, the dispersion of the scalars is dominated by the divergence and convergence of the streamlines around the cylinder rather than by the wake turbulence. Approximate expressions are derived for this topologically influenced dispersion in terms of the geometry of the array. The fact when most of the flow in the array can be approximated by a potential flow, allows us to introduce a fast approximate calculation method to compute the dispersion.     

有腹筋混凝土梁的剪切刚度分析模型

钢筋混凝土梁的挠度计算通常不计入剪切变形的贡献,然而对于斜向开裂的有腹筋混凝土梁,斜裂缝会显著降低梁体的有效剪切刚度,导致剪切变形值显著增大,因此在验算评估时应予以考虑.为评价钢筋混凝土梁斜向开裂后的有效剪切刚度,首先,基于变角桁架模型推导了钢筋混凝土梁在箍筋屈服状态下的有效剪切刚度;与弹性剪切刚度比较发现,剪切刚度退化系数的主要影响因素为材料弹模比、配箍率和斜压杆倾角.其次,基于试验剪切变形曲线表现出的刚度退化规律,提出了可用于不同开裂程度下剪切刚度计算的恒定切线刚度退化模式,并采用开裂后的剪力增量作为反映开裂程度的定量指标.最后,根据最小能量原理得到了剪切刚度退化中两个关键参数:斜压杆倾角和剪切刚度退化系数的解析公式.通过2根薄腹混凝土梁剪切变形试验以及收集的15个受剪梁段的剪切变形数据对模型有效性进行了验证,验证结果表明:有腹筋混凝土梁剪切刚度分析模型能较为准确地预测箍筋屈服状态的剪切刚度,并能反映不同开裂程度下的剪切刚度退化规律.     

Toward multipurpose NMR experiments

Standard phase cycled NMR pulse sequences were generalized such that for each individual step of the pulse phase cycle the free induction decay is stored separately without phase correction. This is in contrast to the usual practice, where pulse responses are phase shifted immediately (by applying a ‘receiver phase’) and co‐added as they are stored. The approach used here allows one to extract different types of NMR information, which are usually referred to as different ‘experiments’, from the same raw data set a posteriori by using complex linear combinations. Storing the free induction decays of individual phase cycle steps separately and using specific linear combinations of these data to obtain a particular type of information increase the overall efficiency of a given set of NMR experiments substantially, because all information can be derived from a single multiplexed data set. This ‘super‐experiment’ requires only as much time as the most complex of the derived specific experiments alone. The principle of this multipurpose approach was demonstrated by performing different multiple‐quantum filtered COSY experiments. It also becomes possible to generate linear combinations, which differ from the conventionally acquired spectra a posteriori. For example, we implemented diagonal peak reduction by using zero‐ and single‐quantum filtered COSY contributions without requiring additional experiment time. Copyright © 2009 John Wiley & Sons, Ltd.     

本质正规算子的模C_2近似酉等价

本文刻划了本质正规算子模的近似酉等价，并且改进了Ｃａｒｒｙ和Ｐｉｎｃｕｓ在［１］中的一些结果。     

Forced flexural-and-torsional vibrations of a cantilever beam of constant cross section

The paper addresses the forced flexural-and-torsional vibrations of a cantilever beam of constant cross section. The relevant boundary-value problem is solved. The system of two partial differential equations of the fourth order that describes these vibrations is analyzed in a vector-function space and is subjected to an equivalent transformation to obtain one vector equation of the fourth order with two matrices as coefficients. One is an idempotent matrix; the other is a diagonal matrix. This makes it much easier to construct a Cauchy vector function as an analytic function of these matrices __________ Translated from Prikladnaya Mekhanika, Vol. 43, No. 8, pp. 102–114, August 2007.     

Bounds for the Perron root, singularity/nonsingularity conditions, and eigenvalue inclusion sets

Using a unified approach based on the monotonicity property of the Perron root and its circuit extension, a series of exact two-sided bounds for the Perron root of a nonnegative matrix in terms of paths in the associated directed graph is obtained. A method for deriving the so-called mixed upper bounds is suggested. Based on the upper bounds for the Perron root, new diagonal dominance type conditions for matrices are introduced. The singularity/nonsingularity problem for matrices satisfying such conditions is analyzed, and the associated eigenvalue inclusion sets are presented. In particular, a bridge connecting Gerschgorin disks with Brualdi eigenvalue inclusion sets is found. Extensions to matrices partitioned into blocks are proposed.