基于激光诱导击穿光谱(LIBS)的煤灰熔点快速检测

炉内结渣是影响火电机组和气化工艺可靠运行的关键因素之一,准确预测灰熔点可以提前调整炉膛出口温度以避免结渣。本论文采用激光诱导击穿光谱（LIBS）采集煤灰样中金属元素的光谱,分别建立煤灰中的金属元素的谱线强度与煤灰熔点的随机森林模型、支持向量机回归模型和线性回归模型,直接预测煤灰熔点温度。采用基于马氏距离(MD)的异常数据剔除算法和基于稀疏矩阵的基线估计与降噪算法（BEADS）,对粉煤灰样的全光谱数据进行了预处理。随机森林模型对粉煤灰熔点的预测平均相对误差（MRE）为54.74%,支持向量机回归模型的预测平均相对误差为60.08%,而线性回归模型的预测平均相对误差达到了9.78%。研究结果表明,线性回归模型对煤灰熔点的预测结果更准确。     

交叉分段差分进化支持向量回归的气体超声 流量计测量方法

为了进一步提高全量程气体超声流量计的测量精度,基于多通道声波到时和实时温度,提出了一种交叉分段差分进化(Differential Evolution)支持向量回归(Support Vector Regression)DE-SVR模型。考虑到气体在不同流量条件下的流体状态不同,提出了交叉分段处理的方法,采用DE算法优化选取SVR参数。实验结果表明,对于16~1600m3/h全量程,交叉分段DE-SVR和传统积分方法计算气体流量的平均相对误差分别为0.00447和0.02781,前者较后者降低了83.93%;对于16~160m3/h小流量,交叉分段DE-SVR和无分段DE-SVR算法计算结果平均相对误差分别为0.00436和0.03214,前者较后者降低了86.43%。该方法有效避免了声道长度、探头角度以及管道直径等参数不确定性对流量计算的影响,为全量程气体流量的高精度测量提供了保障。     

金催化乙炔氢氯化的密度泛函理论研究进展

乙炔氢氯化(AH)是生产氯乙烯的主要途径之一,传统上使用高毒性的汞催化剂,因此开发无汞催化剂迫在眉睫。金(Au)催化剂是最有潜力的替代催化剂之一,然而其活性Au物种、反应物的活化过程或反应过渡态结构等催化机理仍不够清晰。密度泛函理论（DFT）在研究由Au催化AH的反应机理中发挥了极其重要的作用。本文综述了DFT对金催化剂活性位点、反应物在催化剂上的吸附性质及反应机理的研究进展。重点讨论了DFT对阳离子金和金簇催化AH反应过程的模拟计算,包括Au电子状态、其它原子掺杂及金簇尺寸和形状对催化AH反应影响的模拟。结果表明DFT模拟计算在微观分子尺度上研究反应物的吸附、反应中间体及过渡态等方面发挥了关键作用,对理解Au催化AH反应机理做出了重要贡献。     

On Vector Quasi-Equilibrium Problems with Set-Valued Maps

In this paper, we introduce a new class of vector quasi-equilibrium problems with set-valued maps. Almost all the vector equilibrium models of the Blum-Oettli type in the literature are special cases of our new class of equilibrium problems under consideration. Moreover, a number of C-diagonal quasiconvexity properties are proposed for set-valued maps, which are natural generalizations of the -diagonal quasiconvexity for real functions. Together with an application of continuous selection and fixed-point theorems, these conditions enable us to prove unified existence results of solutions for such vector equilibrium problems.     

海洋气溶胶多角度偏振辐射特性研究

海洋气溶胶是对流层气溶胶的重要组成部分,对全球的辐射收支平衡及气侯变化均有重要的影响。评价气溶胶的直接与间接辐射效应需要对气溶胶的性质进行深入的研究。多角度偏振为气溶胶光学物理性质研究提供了新的方法。在对可见光550 nm和近红外860 nm波段处海洋气溶胶的光学性质的研究基础上,采用矢量辐射传输模型模拟了TOA（top of atmosphere）反射率和偏振反射率与下垫面性质、观测方位角、气溶胶光学厚度之间的敏感性。模拟结果表明,海洋型气溶胶的多角度偏振信息可以有效地体现气溶胶的光学性质,可以利用多角度偏振遥感信息进行海洋气溶胶的反演,为利用多角度偏振遥感数据进行海洋气溶胶提供了理论基础。     

激光偏振性的量子理论与实验研究

本文建立了激光偏振性的全量子理论。分析了各向异性腔中多模运转下激光束的偏振特性。给出He-Ne激光器在6328振荡谱线上各模相互间角度随各向异性变化的基本方程、计算曲线及偏振组态。本文还给出了由应力导致相位各向异性对模式偏振性影响的实验结果。理论分析与实验结果相符。     

基于支持向量机的肤色滤波器

为了探测图像中的肤色像素,提出了一种新的方法-支持向量机（SVM：Support Vector Machine）方法.它是一种基于肤色的非特定人的面部定位方法,是非接触人机交互技术和机器视觉中的一个重要内容.实验结果表明,采用支持向量机方法较传统人工神经网络方法不仅有更高的探测准确性,而且具有更好的推广性能.由于SVM采用结构风险最小化（SRM：Structural Risk Minimization）准则,在最小化训练误差（经验风险）的同时,尽量缩小模型预测误差的上界,从而使模型有更好的泛化能力.     

偏心和倾斜光学系统的像差特性

运用一种新的分析偏心和倾斜光学系统初级像差特性的方法—矢量像差理论,分析了具有较大交叉倾斜角的平行平板光学系统和具有较大倾斜角的光楔光学系统的像差特性.通过ZEMAX光学设计软件像差分析功能中的垂轴像差曲线、场曲曲线和点列图加以比较,两者非常吻合,证明该理论可以对具有较大倾斜元件的非共轴光学系统设计起指导作用.     

A recursive color image edge detection method using Green's function approach

In this paper, an extended version of image edge detector using Green's function approach is proposed for detection of edges in the color vector space field. In the proposed method, the relationship between the Red, Green and Blue components is considered to design a differential operator for detection of edges in color images. By using the proposed operator, partial derivatives of all components of color image can simultaneously affect on the edge detection process. Therefore the proposed method can preserve the vector nature of color images during the edge processing stages. Also, the proposed method is compared both quantitatively and qualitatively with other color edge detectors. Experimental results show that the proposed method can efficiently preserve the edges even when the color images corrupted with different levels of noise.     

Curvelet based image compression via core vector machine

In this paper, a novel curvelet based digital image compression scheme is proposed. Aiming at achieving high compression ratio, the proposed scheme embeds a representative machine learning method, core vector machine (CVM), in the encoding process of the image compression technique. The core vector machine (CVM) has been introduced as an extremely fast classifier which is demonstrably superior to standard support vector machine (SVM) on very large datasets. In this scheme, we appropriately utilize the characteristic of CVM to reduce huge numbers of curvelet coefficients. Compared with image compression algorithms do not use CVM and methods based on wavelet transform, experimental results show that the compression performance of our method gains much improvement in peak-signal-to-noise-ratio (PSNR) and CPU time. Moreover, the algorithm works fairly well for declining block effect at higher compression ratios.