近红外光谱定量分析的新方法:半监督最小二乘支持向量回归机

在近红外光谱定量分析中,样品化学值测定的准确度是运用数学模型进行定量分析精确度的理论极限。但能够准确获取化学值的样品数量比较少,许多模型在建模时只考虑这部分样品数据,而不考虑大量的无化学值的样品数据。针对该问题,本文在LS-SVR的基础上,提出了可以同时利用有化学值(标签)和无化学值样品数据的半监督LS-SVR(S2LS-SVR)模型。类似于LS-SVR,该模型也只需求解一个线性方程组。最后,以烤烟样品数据集为实验材料,建立了四种样品成分(总糖、还原糖、总氮和烟碱)的定量分析模型。四种样品成分的预测值与实际值的平均误差分别为6.62%,7.56%,6.11%和8.20%,相关系数分别为0.974 1,0.973 3,0.923 0和0.948 6。经分析比较发现S2LS-SVR模型优于PLS和LS-SVR,从而验证了S2LS-SVR模型的可行性和有效性。     

近红外漫反射线性加和光谱在烟叶复烤配方中的应用

应用近红外光谱分析技术进行烟叶复烤配方的辅助分析以及设计过程中,需要大量的不同混合比例的样品光谱,而混合样品光谱在实际生产过程中是很难获得的;且不同分级的样品很难实现均匀混合,不可避免的引入误差.文章提出了应用近红外漫反射光谱的线性加和原理产生"理论混合均匀光谱"来代替实际的混合样品光谱,不仅能够消除混合不均匀造成的误...     

Ion implantation effects on the microhardness and microstructure of GaN

We study the effect of N⁺ and O⁺ implantation on the microhardness and the microstructure of epitaxially grown GaN. The microhardness is measured using a Knoop diamond indenter while information on the effect of implantation on the surface morphology, microstructure and electronic structure is provided by atomic force microscopy, cross-section transmission electron microscopy and near edge X-ray absorption fine structure (NEXAFS) spectroscopy. It is demonstrated that implantation increases the surface microhardness. A possible mechanism for the surface hardening effect is based on the formation of N interstitials that pin the dislocations and prohibit the plastic deformation. In addition to the hardening effect, the implantation induced N interstitials introduce a characteristic resonance in the NEXAFS spectra, at 1.4 eV below the absorption edge.     

Bounded rational expectations and the stability of interest rate policy

The New Keynesian model has recently been subject to two serious criticisms: the model cannot produce plausible inflation and output dynamics following a monetary shock, and the stability of its dynamics suffers from indeterminacy. The procedures that have been proposed to eliminate these two shortcomings fall into two categories: the introduction of some sort of backward price indexation into the standard model and/or other forms of stickiness (like sticky information); and the adoption of some form of policy rule that completely offsets the effects of forward looking dynamics in the optimization process. In this paper we do not eradicate forward looking behavior from the dynamics of the New Keynesian model, neither do we impose some form of backward price indexation. We assume that private economic agents have forward looking behavior and that they do try to optimize with all available information; the only novelty is that they are allowed to make small mistakes near the rational expectations equilibrium, in a fully deterministic setup. These “near rational” or “bounded rational” expectations show that the dynamics of the model with active interest rate rules is much richer than the simple problem of local indeterminacy as is usually found in the literature.     

Near optimality of stochastic control in systems with unknown parameter processes

Stochastic control for systems with an unknown parameter is considered in this paper. The underlying problem is to minimize a functional subject to a system described by a singularly perturbed differential equation with an unknown parameter process driven by fast fluctuating random disturbances. This problem arises in the context of stochastic adaptive control, adaptive signal processing, and failure-prone manufacturing systems. Due to the nature of the wide-bandwidth noise processes, identifying the parameter process for eacht is very hard since the driving noise changes very rapidly. An alternative approach is used, and an auxiliary control problem is introduced to overcome the difficulties. By means of weak convergence methods and comparison control techniques, nearly optimal controls are obtained.This research was supported in part by the National Science Foundation under Grant DMS-9022139 and DMS-9224372.     

小波系数压缩用于柴油十六烷值近红外光谱分析建模

近红外光谱分析建模中存在多变量高维数据处理问题,导致计算量大,不利于过程控制中应用.为此提出利用小波变换压缩近红外光谱数据的算法与准则,并结合柴油十六烷值定量分析研究压缩数据的建模效果.研究表明,经小波方法处理后,变量维数压缩30倍左右,光谱主要信息基本保留,而模型的预测精度和常规预处理方法分析相比有明显提高.光谱数据压缩的同时包含了噪声滤除和基线校正,简化数据处理步骤,有利于NIRS实际应用时提高建模效率.     

A consensus successive projections algorithm – multiple linear regression method for analyzing near infrared spectra

The successive projections algorithm (SPA) is widely used to select variables for multiple linear regression (MLR) modeling. However, SPA used only once may not obtain all the useful information of the full spectra, because the number of selected variables cannot exceed the number of calibration samples in the SPA algorithm. Therefore, the SPA-MLR method risks the loss of useful information. To make a full use of the useful information in the spectra, a new method named “consensus SPA-MLR” (C-SPA-MLR) is proposed herein. This method is the combination of consensus strategy and SPA-MLR method. In the C-SPA-MLR method, SPA-MLR is used to construct member models with different subsets of variables, which are selected from the remaining variables iteratively. A consensus prediction is obtained by combining the predictions of the member models. The proposed method is evaluated by analyzing the near infrared (NIR) spectra of corn and diesel. The results of C-SPA-MLR method showed a better prediction performance compared with the SPA-MLR and full-spectra PLS methods. Moreover, these results could serve as a reference for combination the consensus strategy and other variable selection methods when analyzing NIR spectra and other spectroscopic techniques.     

Combining linear polarization spectroscopy and the Representative Layer Theory to measure the Beer–Lambert law absorbance of highly scattering materials

Visible and Near Infrared (Vis–NIR) Spectroscopy is a powerful non destructive analytical method used to analyze major compounds in bulk materials and products and requiring no sample preparation. It is widely used in routine analysis and also in-line in industries, in-vivo with biomedical applications or in-field for agricultural and environmental applications. However, highly scattering samples subvert Beer–Lambert law's linear relationship between spectral absorbance and the concentrations. Instead of spectral pre-processing, which is commonly used by Vis–NIR spectroscopists to mitigate the scattering effect, we put forward an optical method, based on Polarized Light Spectroscopy to improve the absorbance signal measurement on highly scattering samples. This method selects part of the signal which is less impacted by scattering. The resulted signal is combined in the Absorption/Remission function defined in Dahm's Representative Layer Theory to compute an absorbance signal fulfilling Beer–Lambert's law, i.e. being linearly related to concentration of the chemicals composing the sample. The underpinning theories have been experimentally evaluated on scattering samples in liquid form and in powdered form. The method produced more accurate spectra and the Pearson's coefficient assessing the linearity between the absorbance spectra and the concentration of the added dye improved from 0.94 to 0.99 for liquid samples and 0.84–0.97 for powdered samples.     

Water revealed as molecular mirror when measuring low concentrations of sugar with near infrared light

Near infrared spectroscopy is an overtone spectroscopy regarded as a quick and non-destructive method that provides analytical solutions for components that represent approximately 1% or more of the total mass of the investigated composite samples. Aquaphotomics offers the possibility for disentanglement of information remaining hidden in the spectra when conventional data evaluation methods are used, since this concept utilizes changes of the water structure induced by the measured solute as specific molecular vibrations at water bands. Here, near infrared technique and aquaphotomics are applied for non-destructive identification and quantification of mono- and di-saccharide solutes at 100–0.02 mM concentration that is accepted as unachievable with near infrared spectroscopy. The results presented in this study support the aquaphotomics' water molecular mirror concept that explores spectral changes related to water molecular rearrangements caused by minute changes of the solutes in the aqueous systems. The method provides quick and accurate alternative for classical analytical measurements of saccharides even at millimolar concentration levels.     

基于共轭聚合物的近红外光传感器

近红外光(NIR)传感器在军事警戒、空间勘探、科研检测、医疗诊断等领域有着极为重要的应用价值。传统的NIR传感材料主要是基于半金属的无机材料,借助窄带隙来吸收低能量的NIR而改变材料自身的电导率,继而实现检测。无机传感材料由于加工繁琐、不具备柔性、成本高昂以及难降解等因素致其进一步发展受到限制。新兴的共轭聚合物材料通过光热转换或者能级可调的光电效应实现对NIR的高灵敏传感,同时能够实现快速响应。该类聚合物材料具有π电子离域的共轭结构,同样具有较窄的带隙,在NIR照射下能够通过自身电导率的改变或光热转换作用将热量传递至其他超热敏材料来实现对光的检测。借助柔性、环境友好、制备简单、便于掺杂、灵敏度高等优势,共轭聚合物材料为新一代NIR传感器的应用与普及开辟了新的前景。