On estimating model complexity and prediction errors in multivariate calibration: generalized resampling by random sample weighting (RSW)

The present paper focuses on determining the number of PLS components by using resampling methods such as cross validation (CV), Monte Carlo cross validation (MCCV), bootstrapping (BS), etc. To resample the training data, random non‐negative weights are assigned to the original training samples and a sample‐weighted PLS model is developed without increasing the computational burden much. Random weighting is a generalization of the traditional resampling methods and is expected to have a lower risk of getting an insufficient training set. For prediction, only the training samples with random weights less than a threshold value are selected to ensure that the prediction samples have less influence on training. For complicated data, because the optimal number of PLS components is often not unique or readily distinguished and there might exist an optimal region of model complexity, the distribution of prediction errors can be more useful than a single value of root mean squared error of prediction (RMSEP). Therefore, the distribution of prediction errors are estimated by repeated random sample weighting and used to determine model complexity. RSW is compared with its traditional counterparts like CV, MCCV, BS and a recently proposed randomization test method to demonstrate its usefulness. Copyright © 2010 John Wiley & Sons, Ltd.     

Application of in-plasma catalysis and post-plasma catalysis for methane partial oxidation to methanol over a Fe2O3-CuO/γ-Al2O3 catalyst

Methane partial oxidation to methanol (MPOM) using dielectric barrier discharge over a Fe2O3-CuO/γ-Al2O3 catalyst was performed. The multicomponent catalyst was combined with plasma in two different configurations, i.e., in-plasma catalysis (IPC) and post-plasma catalysis (PPC). It was found that the catalytic performance of the catalysts for MPOM was strongly dependent on the hybrid configuration. A better synergistic performance of plasma and catalysis was achieved in the IPC configuration, but the catalysts packed in the discharge zone showed lower stability than those connected to the discharge zone in sequence. Active species, such as ozone, atomic oxygen and methyl radicals, were produced from the plasma-catalysis process, and made a major contribution to methanol synthesis. These active species were identified by the means of in situ optical emission spectra, ozone measurement and FT-IR spectra. It was confirmed that the amount of active species in the IPC system was greater than that in the PPC system. The results of TG, XRD, and N2 adsorption-desorption revealed that carbon deposition on the spent catalyst surface was responsible for the catalyst deactivation in the IPC configuration.     

Convex hull: A new method to determine the separation space used and to optimize operating conditions for comprehensive two-dimensional gas chromatography

It is important to develop methods of optimizing the selection of column sets and operating conditions for comprehensive two-dimensional gas chromatography. A new method for the calculation of the percentage of separation space used was developed using Delaunay's triangulation algorithms (convex hull). This approach was compared with an existing method and showed better precision and accuracy. It was successfully applied to the selection of the most convenient column set and the geometrical parameters of second column for the analysis of 49 target compounds in wastewater.     

Wave optics simulation approach for high-power laser beam combination

A numerical wave optics approach for simulating multiplex super Gaussian beam combination in high-power laser facility is presented. The approach involves the application of a sequence of random phase screens and von Karman turbulence phase to an initial beam field and the summation of the intensity results. The analytical and numerical calculation results have been given to illustrate the properties of the resulting far-field pattern after coherent, incoherent and spatially coherent combination, and some of that are compared with experimental results. The approach can be used for modeling and evaluating applications such as high-power laser shooting that might utilize different beams combining approach.     

量子棒中强耦合杂质束缚极化子的振动频率

给出了具有椭球边界量子棒经坐标变换成球形边界的哈密顿量。采用线性组合算符和幺正变换的方法研究了在非均匀抛物限制势下量子棒中强耦合杂质束缚极化子的性质。导出了量子棒中强耦合杂质束缚极化子的振动频率和声子平均数随库仑束缚势、电子-声子耦合强度、椭球的纵横比和量子棒的横向和纵向有效受限长度的变化关系。数值计算结果表明:振动频率和声子平均数随电子-声子耦合强度和库仑束缚势的增强而增加,随量子棒的横向和纵向有效受限长度和椭球的纵横比的减小而增大。表现出量子棒的奇特的量子尺寸限制效应。     

激光相干合成中单元的相位和振幅波动对合成光束远场分布的影响

运用傅里叶光学分析法推导出系统抖动造成单个光源的相位和振幅发生波动时的远场光强表达式,以1维阵列为例分析了系统抖动对远场的影响.研究表明:随着参与合束的发光单元数目的增加,尖峰变锐,能量更集中;系统抖动引起了远场峰值光强的减少.出现了本底现象,破坏了远场的对称性和光束质量;激光阵列单个光源的相位随机抖动应该控制在3 π/8范围内;相干合束发光单元数目越多,系统抖动对远场的影响越小.     

平顶高斯超短脉冲激光的时间部分相干性对相干合成的影响

 利用准单色光干涉理论,计算了2×2平顶高斯超短脉冲激光在远场的干涉图样,分析了激光的时间部分相干性及其存在束间相位差时对相干合成的影响。计算结果表明,当单路激光输出光束线宽小于100 nm时,理想波前的远场光斑图样和峰值光强基本保持不变,当光束线宽大于100 nm时,干涉减弱衍射效应逐渐增强;光束间存在相位差时,时间部分相干性将破坏远场图样随束间相位差的周期性变化,相位差越大干涉越弱;当超过相干长度时,则为非相干合成,但1/6倍相干长度范围内,仍可视为较理想的相干合成,此时焦斑的斯特列尔比大于0.9,对超短脉冲的相干合成影响可忽略。     

高平均功率固体激光功率和光束质量关系研究进展

 固体激光的一个重大发展方向和目标是实现“三高”(高平均功率、高光束质量和高效率)激光同时输出,结合相关研究工作和进展,着重论述了固体激光输出功率和光束质量的关系,给出了功率升高、光束质量非线性下降是当前“三高”固体激光研究的一个基本科学技术问题。围绕这一基本问题,评述了一些重要技术途径和手段,并探讨了实现高平均功率高光束质量固体激光输出的可能有效途径。     

激光空间相干合成效果评价方法的研究

根据目标上实际最大光强与理想合成效果下最大光强的比值定义了合成效果因子,用以评价相干合成效果.在此基础上研究了影响相干合成效果的主要因素,并对合成效果因子的传输特性进行了研究.结果表明:阵列紧密程度对合成效果影响不大,在较小的空间占空比下有可能得到较好的合成效果.影响合成效果的主要因素是随机相位差,当随机相位差增大到一定程度时,光束合成其实已经由相干合成过渡到了非相干合成,得不到预期的相干合成效果;随着传输距离的增大,空间占空比对合成效果因子的影响将进一步减小,影响合成效果的主要因素之一是随机相位差.     

多光束激光远场干涉实验研究

设计了多组多孔径光阑,将激光扩束后通过自制的多孔径光阑,获得多光束相干激光光源.基于多光束干涉理论,通过改变不同的光阑,得到多组点状阵列分布的远场干涉图样,结果表明远场中心主瓣具有随着占空比减小能量降低、旁瓣增多的分布规律.实验结果表明,相干源呈正三角形分布时能量集中度较好.对高功率激光相干合成具有一定的参考意义.