一种适用于混合型分类数据的聚类算法

传统的K-modes算法采用简单的属性匹配方式计算同一属性下不同属性值的距离，并且计算样本距离时令所有属性权重相等。在此基础上，综合考虑有序型分类数据中属性值的顺序关系、无序型分类数据中不同属性值之间的相似性以及各属性之间的关系等，提出一种更加适用于混合型分类数据的改进聚类算法，该算法对无序型分类数据和有序型分类数据采用不同的距离度量，并且用平均熵赋予相应的权重。实验结果表明，改进算法在人工数据集和真实数据集上均有比K-modes算法及其改进算法更好的聚类效果。     

Constant hue bands in boundary colors discovered using a new appearance model

Boundary colors are observed when light from a scene is dispersed by a prism or diffraction grating. We discovered that patterns with repeating black and white stripes can produce repeating bands of boundary colors with two hues. These hues are virtually constant as measured by chromaticity or CIELAB. We found seven cases of this kind using a new appearance model for boundary colors. The model correctly predicts that green and magenta bands recur as stripe widths and dispersion strength vary. The first green/magenta case in the sequence traces out an accurate ellipse in XYZ color space. Green and magenta bands are prominent in supernumerary rainbows and interference rings, and we explain why that might be the case. The explanation is based on an interesting property of the visible spectrum. In addition to the green/magenta cases, the other cases are orange/cyan, yellowish‐green/purple, and yellow/violet. The success of the boundary color appearance model implies that bands are perceived as if the wavelength responses of the cones were essentially independent, which contradicts the actual behavior of cones. © 2014 Wiley Periodicals, Inc. Col Res Appl, 40, 135–146, 2015     

Methodology for the application of velocimetry by molecular tagging of hypersonic flows

A laser-induced NO fluorescence technique was applied to measure velocity in a hypersonic shock tunnel nozzle exit. For the application of the technique, a detailed study of the density and fluorescence lifetime of the tracer radical, flow velocity and effective test time is proposed, resulting in a methodology for the application of the technique in hypersonic pulsed facilities. The study has demonstrated that it is necessary to jointly evaluate the flow velocity, the fluorescence lifetime of the radical and the width at half height of the laser beam, resulting in a kind of indicator for the feasibility of the technique. The variation of the laser incidence time with respect to the Pitot signal showed that it is not enough to select a stable Pitot pressure signal region to define the laser incidence time, preliminary trial and error analysis are necessary for each device used. Furthermore, the analysis of the velocity values calculated from the linear fit method shows that the adoption of such a method eliminates the effect of the systematic error of the measurements.     

一种改进的K-Modes聚类算法

传统的K-Modes算法采用0-1简单匹配方法计算对象与类中心(Modes)之间的距离,并将每个对象分配到离它最近的类中去。采用基于频率方法重新计算各类的类中心(Modes)、定义目标函数,然而,对象的归类方法和目标函数的定义没有充分考虑分类数据的特点。对此,提出一种改进的K-Modes算法,采用期望熵最小的衡量方法进行归类,并且采用期望熵作为新的目标函数。通过实验将该算法与传统的K-Modes算法进行比较,表明该算法是更有效的。     

新高阶多变量马尔可夫模型

本文提出了一种新高阶多变量马尔可夫模型,并对其收敛性进行了分析.给出了模型的参数估计方法.数值实验表明在预测精度方面新高阶多变量马尔可夫模型比高阶多变量马尔可夫模型更加有效.     

Acquiring real kinetics of reactions in the inhibitory atmosphere containing product gases using micro fluidized bed

This study proposed an isotope-tagging method to investigate reactions under the atmosphere of product gas. To illustrate this method, the calcination kinetics of calcium carbonate Ca¹³CO₃ in CO₂ atmospheres were investigated by monitoring ¹³CO₂ produced using a micro fluidized bed reaction analyzer (MFBRA). The results demonstrated that the presence of CO₂ in reaction atmosphere increases the apparent activation energy. The increase in the apparent activation energy is, however, significantly overestimated by the thermogravimetric analyzer (TGA) because of the excessive suppression by stagnated product gas inside the sample crucible. Comparatively, the apparent activation energy increases with CO₂ from the MFBRA due primarily to the thermal equilibrium limitation, because the gas diffusion in the MFBRA is essentially eliminated. It is thus concluded that the MFBRA is quite capable of acquiring the real kinetics of reactions in such inhibitory atmospheres.     

Factorial 2 p–q Plans Robust Against Linear and Quadratic Trends

When the substrate of a 2 ^n–q factorial or fractional factorial experiment may be expected to show trends representable by linear and quadratic terms in time, then certain orderings spaced at equal time-intervals permit better estimation of the effects than do others. Some of these ordered plans are given for p – q = 2, 3, 4, 5. Simple methods are given for computing effects, trends, and efficiencies.     

Variance plus bias optimal response surface designs with qualitative factors applied to stem choice modeling

This paper explores the issue of model misspecification, or bias, in the context of response surface design problems involving quantitative and qualitative factors. New designs are proposed specifically to address bias and compared with five types of alternatives ranging from types of composite to D‐optimal designs using four criteria including D‐efficiency and measured accuracy on test problems. Findings include that certain designs from the literature are expected to cause prediction errors that practitioners would likely find unacceptable. A case study relating to the selection of science, technology, engineering, or mathematics majors by college students confirms that the expected substantial improvements in prediction accuracy using the proposed designs can be realized in relevant situations. Copyright © 2011 John Wiley & Sons, Ltd.     

AutoCAD技术在机械设计中的应用与实践

圆柱齿轮传动是机械传动中应用最广泛的一种传动装置。齿轮齿形为渐开线结构,一直以来AutoCAD中渐开线的绘制都使设计人员感到头疼。文章介绍了一种基于Excel的齿轮渐开线的绘制过程,同时详细论述了齿轮的标注,包括尺寸标注、尺寸公差、粗糙度、形位公差和技术要求标注等。     

An improved run‐to‐run process control scheme for categorical observations with misclassification errors

When product quality characteristics are evaluated and assigned to exclusive categories, measurement errors (misclassification of products) always exist unless a perfect measurement system is used to identify the categories. In run‐to‐run (R2R) process control, a categorical controller has been developed for process adjustments with categorical variables. However, if process outputs are misclassified, an adjustment bias will be introduced by the controller. In this study, an improved categorical R2R controller that utilizes the misclassification probabilities to decrease process deviation is proposed. Simulation results show that the proposed controller exhibits better performance when misclassification exists. Copyright © 2008 John Wiley & Sons, Ltd.