人工神经网络法鉴别红外光谱

本文将反向传播人工神经网络（ＢＰ－ＡＮＮ）用于ＦＴＩＲ,鉴别未知化合物。结果表明,当训练集样本不含噪声时,纯光谱的预测结果很好。而当训练集样本有少量噪声干扰时,预测结果随预测集样本的不同,而得到不同的改善。     

A remark on similarity analysis of boundary layer equations of a class of non-Newtonian fluids

A similarity analysis of three-dimensional boundary layer equations of a class of non-Newtonian fluid in which the stress, an arbitrary function of rates of strain, is studied. It is shown that under any group of transformation, for an arbitrary stress function, not all non-Newtonian fluids possess a similarity solution for the flow past a wedge inclined at arbitrary angle except Ostwald-de-Waele power-law fluid. Further it is observed, for non-Newtonian fluids of any model only $90°$ of wedge flow leads to similarity solutions. Our results contain a correction to some flaws in Pakdemirli׳s [14] (1994) paper on similarity analysis of boundary layer equations of a class of non-Newtonian fluids.     

Analysis and use of fragment-occurrence data in similarity-based virtual screening

Current systems for similarity-based virtual screening use similarity measures in which all the fragments in a fingerprint contribute equally to the calculation of structural similarity. This paper discusses the weighting of fragments on the basis of their frequencies of occurrence in molecules. Extensive experiments with sets of active molecules from the MDL Drug Data Report and the World of Molecular Bioactivity databases, using fingerprints encoding Tripos holograms, Pipeline Pilot ECFC_4 circular substructures and Sunset Molecular keys, demonstrate clearly that frequency-based screening is generally more effective than conventional, unweighted screening. The results suggest that standardising the raw occurrence frequencies by taking the square root of the frequencies will maximise the effectiveness of virtual screening. An upper-bound analysis shows the complex interactions that can take place between representations, weighting schemes and similarity coefficients when similarity measures are computed, and provides a rationalisation of the relative performance of the various weighting schemes.     

Influence of temperature-dependent viscosity and thermal radiation on MHD forced convection over a non-isothermal wedge

An analysis has been carried out to study heat transfer characteristics of an incompressible Newtonian electrically conducting and heat generating/absorbing fluid having temperature-dependent viscosity over a non-isothermal wedge in the presence of thermal radiation. The Rosseland approximation is used to describe the radiative heat flux in the energy equation. The wedge surface is assumed to be permeable so as to allow for possible wall suction or injection. The effects of viscous dissipation, Joule heating, stress work and thermal radiation are included in the model. The governing differential equations are derived and transformed using a non-similarity transformation. The transformed equations are solved numerically by applying a fifth-order Runge-Kutta-Fehlberg scheme with shooting technique. Favorable comparisons with previously published work on various special cases of the problem are obtained. Numerical results for the velocity and temperature profiles for a prescribed magnetic field parameter as well as the development of the local skin-friction coefficient and local Nusselt number with the magnetic field and radiation parameters are presented graphically and in tabulated form to elucidate the influence of the various physical parameters.     

The Cavalcante-Tenenblat equation - Does the equation admits physical significance?

The main purpose of the given paper is to analyze a less studied third order non-linear partial differential equation, the so-called Cavalcante-Tenenblat equation (CTE) in the following form: .Since general class of solutions are of basic interest a complete characterization of the group properties is given. The traveling wave ‘ansatz’ restricts the solution manifold to special class of solutions and hence, a generalize algorithm is necessary.We determine the Lie point symmetry vector fields and calculate similarity ‘ansätze’. Further, we also derive a few non-linear transformations and some similarity solutions are obtained explicitly. Due to the complexity of some similarity solutions a numerical procedure is of advantage.Moreover, the non-classical case (potential symmetries) is studied to the first time and further, we show how the CTE leads to approximate symmetries and we apply the method to the first time. We call the disturbed equation the CTE-ε equation and we show how to derive new class of solutions.Finally, the equation does not pass the Painlevé-test and is therefore not soluble by the Inverse Scattering Transform Method (IST).Hence, suitable alternative (algebraic) approaches are necessary to derive class of solutions explicitly.     

集值信息系统在相容关系下的属性约简

借助于属性集值的相似程度在集值信息系统上定义了一种新的相客关系,给出了这种相客关系下集值信息系统的属性约简与判定,得到了集值信息系统属性约简的具体探作方法,并讨论了相似水平对集值信息系统的属性约简的影响.     

LAMOST恒星分类模板间相似性度量分析

随着获取和收集天文光谱大数据能力的与日俱增,合理利用计算科学技术正确地分析海量光谱的处理方法及结果统计。前述工作采用了欧氏距离分析判别LAMOST实测光谱与模板之间相似度的研究,研究恒星分类准确性取决于高质量的模板光谱,选取LAMOST光谱在用的分类软件中183个恒星模板光谱,分别利用欧氏距离和马氏距离方法得出A,F,G,K和M型恒星模板间的均值和最大值,完成每条谱线相互之间的相关性分析,找出相对距离较大的模板及形成原因。相似度度量可视化实验数据结果表明模板之间具有一定的区分度,通过马氏距离分析模板间相似性能更进一步辨识出相近模板之间的细微差别,具备较优良的判别效果,证实了LAMOST现有分类的各模板间距离较均匀,且分类结果较为准确。该研究可进一步优化在用光谱分类模板,提升LAMOST恒星分类模板库的精确度和可信度。     

基于核映射多光谱特征融合的高光谱遥感图像分类法

多光谱遥感图像的波段设定在理论和实践上都极具地物针对性,在更有效的数据挖掘方法帮助下可以提取足够的光谱特征以区分地物,本文将光谱匹配技术用于相似性度量,以提高分类准确度.首先选定光谱角制图、光谱相关制图、Mahalanobis距离、光谱相似度和光谱信息差异做为光谱度量;随后选择高斯核函数,在选定核函数之后,得到了核映射下的光谱特征度量来挖掘高光谱遥感数据的光谱特征.最后采用核映射多光谱特征融合法对多光谱遥感图像光谱特征的相似性进行描述,得到基于核映射多光谱特征融合的高光谱遥感图像分类算法.使用MATLAB中的LIBSVM工具箱对AVIRIS高光谱遥感数据进行分类实验,并与已有算法进行对比,结果表明本文提出的算法具有较高的分类准确度和性能.     

Analytical solution of a double moving boundary problem for nonlinear flows in one-dimensional semi-infinite long porous media with low permeability简

Based on Huang＇s accurate tri-sectional nonlin- ear kinematic equation （1997）, a dimensionless simplified mathematical model for nonlinear flow in one-dimensional semi-infinite long porous media with low permeability is presented for the case of a constant flow rate on the inner boundary. This model contains double moving boundaries, including an internal moving boundary and an external mov- ing boundary, which are different from the classical Stefan problem in heat conduction： The velocity of the external moving boundary is proportional to the second derivative of the unknown pressure function with respect to the distance parameter on this boundary. Through a similarity transfor- mation, the nonlinear partial differential equation （PDE） sys- tem is transformed into a linear PDE system. Then an ana- lytical solution is obtained for the dimensionless simplified mathematical model. This solution can be used for strictly checking the validity of numerical methods in solving such nonlinear mathematical models for flows in low-permeable porous media for petroleum engineering applications. Finally, through plotted comparison curves from the exact an- alytical solution, the sensitive effects of three characteristic parameters are discussed. It is concluded that with a decrease in the dimensionless critical pressure gradient, the sensi- tive effects of the dimensionless variable on the dimension- less pressure distribution and dimensionless pressure gradi- ent distribution become more serious; with an increase in the dimensionless pseudo threshold pressure gradient, the sensi- tive effects of the dimensionless variable become more serious; the dimensionless threshold pressure gradient （TPG） has a great effect on the external moving boundary but has little effect on the internal moving boundary.     

Locally adaptive Nakagami-based ultrasound similarity measures

The derivation of statistically optimal similarity measures for intensity-based registration is possible by modeling the underlying image noise distribution. The parameters of these distributions are, however, commonly set heuristically across all images. In this article, we show that the estimation of the parameters on the present images largely improves the registration, which is a consequence of the more accurate characterization of the image noise. More precisely, instead of having constant parameters over the entire image domain, we estimate them on patches, leading to a local adaptation of the similarity measure. While this basic idea of creating locally adaptive metrics is interesting for various fields of application, we present the derivation for ultrasound imaging. The domain of ultrasound is particularly appealing for this approach, due to the inherent contamination with speckle noise. Furthermore, there exist detailed analyses of suitable noise distributions in the literature. We present experiments for applying a bivariate Nakagami distribution that facilitates modeling of several scattering scenarios prominent in medical ultrasound. Depending on the number of scatterers per resolution cell and the presence of coherent structures, different Nakagami parameters are required to obtain a valid approximation of the intensity statistics and to account for distributional locality. Our registration results on radio-frequency ultrasound data confirm the theoretical necessity for a spatial adaptation of similarity metrics.