Volatile profile and differentiation between Albariño wines from different origins

Commercial Albariño wines from different subzones of Certified Brand of Origin ‘Rías Baixas’ (NW Spain) were analysed over four consecutive harvests. The contents of monoterpenes, volatile phenols, alcohols, sulphur compounds, acetates, fatty acids and ethyl esters were determined by gas chromatography (flame ionisation detector and flame photometric detector) and gas chromatography–mass spectrometry. These white wines presented high concentrations of free monoterpenes, volatile phenols, acetates, and ethyl esters, while the levels of sulphur compounds, known as default compounds, were very low. All the wines were elaborated from Albariño grapes, but they were produced by different enological practices, such as skin contact and by carrying out malolactic fermentation or not. No significant differences were found between wines from the same subzone and from different vintages. The results obtained, and confirmed by Principal component and linear discriminant analysis, showed a differentiation among these wines as a function of the Galician subzone of origin.     

Sparse HDLSS discrimination with constrained data piling

Regularization is a key component in high dimensional data analyses. In high dimensional discrimination with binary classes, the phenomenon of data piling occurs when the projection of data onto a discriminant vector is dichotomous, one for each class. Regularizing the degree of data piling yields a new class of discrimination rules for high dimension–low sample size data. A discrimination method that regularizes the degree of data piling while achieving sparsity is proposed and solved via a linear programming. Computational efficiency is further improved by a sign-preserving regularization that forces the signs of the estimator to be the same as the mean difference. The proposed classifier shows competitive performances for simulated and real data examples including speech recognition and gene expressions.     

Diagnosing batch processes with insufficient fault data: generation of pseudo batches

To ensure the safety of a batch process and the quality of its final product, one needs to quickly identify an assignable cause of a fault. Cho and Kim (2003 Cho, HW and Kim, KJ. 2003. A method for predicting future observations in the monitoring of a batch process. J. Qual. Tech, 35: 59–69.  ) recently proposed a diagnosis method for batch processes using Fisher's Discriminant Analysis (FDA), which showed a satisfactory performance on industrial batch processes. However, their method (or any other method based on empirical models) has a major limitation when the fault batches available for building an empirical diagnosis model are insufficient. This is a highly critical issue in practice because sufficient fault batches are likely to be unavailable. In this work, we propose a method to handle the insufficiency of the fault data in diagnosing batch processes. The basic idea is to generate so-called pseudo batches from known fault batches and utilise them as part of the diagnosis model data. The performance of the proposed method is demonstrated using a real data set from a PVC batch process. The proposed method is shown to be capable of handling the data insufficiency problem successfully, and yields a reliable diagnosis performance.     

基于距离判别分析法的山洪泥石流预报模型研究

山洪泥石流预报是山洪灾害防御中的一项重要措施。为提供一种有效的预报途径,采用距离判别分析法构建了山洪泥石流预报的距离判别分析模型,选用前期实效雨量和当日雨量为判别因子,以实测山洪泥石流资料为学习样本进行训练,建立相应的判别函数对山洪泥石流进行预报。实例结果表明,该模型预测精度高、误报率低,行之有效。     

煤层底板突水危险性的Fisher判别分析模型

为准确评价底板突水危险性,考虑到不同影响因素对于评价方法的作用与影响不同,运用逐步判别法剔除了判别效果不显著的指标,选取了含水层富水性、水压、隔水层厚度、断层导水性、构造发育程度5项主要影响因素作为判别指标。根据以往突水数据建立了底板突水危险性的Fisher判别分析模型,并绘制了判别空间中样本数据散点图。该模型通过了显著性检验,分类效果显著。对15组训练样本进行回代判别,误判率为0,通过距离判别法准确预测了测试样本的突水危险性,表明该模型在煤层底板突水危险性评价上具有较高的可信度和良好的实用性。     

Characterisation of Serbian unifloral honeys according to their physicochemical parameters

The characterisation of three unifloral Serbian honeys (acacia, sunflower and linden) was carried out based on some common physicochemical parameters (water content, electrical conductivity, free acidity, optical rotation and pH). A total of 201 honey samples, collected during the 2009 harvesting season, were analysed. Principal component analysis (PCA) and linear discriminant analysis (LDA) were used to highlight the data structure and to find the relationships between the physicochemical parameters and the botanical origin of honey. The variables that best discriminated the samples were electrical conductivity (ranging from 0.10 to 0.76 mS/cm), free acidity (ranging from 7.80 to 42.70 meq/kg) and pH (ranging from 3.17 to 5.85). LDA resulted in a classification model with a high predictive power, allowing further assessment of unknown samples of the three unifloral honeys. Determination of geographic origin of acacia honey samples based on physicochemical properties and chemometrics was attempted.     

Ageing status characterization of Chinese rice wines using chemical descriptors combined with multivariate data analysis

Wine ageing status identification is of great commercial and scientific interest, as wine quality and value are closely related to the organoleptic characteristics developed during the ageing process. In this study, Chinese rice wines from three well-known wineries (“Guyuelongshan”, “Kuaijishan” and “Pagoda”) were analyzed for 21 chemical parameters, including six conventional parameters, five sugars, lactic acid and nine macro-elements. Then the experimental data were subjected to multivariate statistical analysis to predict and classify samples of different ageing status (3, 9, 15, 21, and 33 months). Systematic differences between samples were revealed by a two-way analysis of variance (ANOVA) and principal component analysis (PCA). Discrimination model built by forward stepwise linear discriminant analysis (LDA) based on the 16 selected parameters achieved 88.5% accuracy in leave-one-out (LOO) cross-validation. The most five discriminant variables were Zn, Mn, alcohol, Cu and Al, respectively. When the discrimination was performed on the samples from each winery, the classification accuracy in LOO cross-validation was 97.7%, 91.1% and 78.0%, respectively. The results demonstrated that these chemical parameters have the potential to enable the authentication of ageing status of rice wine.     

幅度谱差分子空间雷达自动目标识别

雷达高分辨距离像是目标的重要结构特征,其维数通常很高,造成数据可分性表达差,识别过程计算复杂度高,识别率低。为降低距离像的维数,提出一种新的距离像特征提取方法,即采用直接线性判别分析(dLDA)在距离像幅度谱差分空间进行特征提取,得dLDA幅度谱差分子空间。目标识别即在所得dLDA幅度谱差分子空间中进行。采用外场实测数据,分别训练了最小距离分类器和one-against-all支撑向量机分类器,2种分类器的识别结果均表明,该方法可显著地降低数据维数并提高识别率。     

基于Fisher判别分析法的滑坡稳定性评估模型及其应用

我国西南地区滑坡灾害点多面广、规模小、危害大,预测预报很大程度依赖群测群防体系。针对这种现状,该文提出了一种基于Fisher判别分析法的滑坡稳定性定性评估模型。该模型从致灾因素方面考虑,以发育地层类型、地貌类型、坡度、前后缘相对高差、灾体面积和灾体体积为判别因子,在历史资料的基础上,建立判别模型,对未知潜在滑坡进行判别归类,是一种典型的工程类比模型。该文对建立模型的原理和方法进行了论述,并以重庆酉阳县部分滑坡灾害隐患点资料为样本,详细说明了评估过程,并验证了模型的正确率,证明该模型可靠度比较高,并且简单实用,便于在群测群防体系中大范围推广使用。     

Water content of natural cyclodextrins and their essential oil complexes: A comparative study between Karl Fischer titration and thermal methods

The paper presents a comparative study regarding the water determination in natural cyclodextrins and in their essential oil complexes (Apiaceae, Liliaceae, and Cupressaceae families) by using Karl Fischer titration (KFT) and thermal methods. For the natural cyclodextrins, the influence of the solvent hydrophobicity and the preheating temperature on the water extraction process were evaluated. The water contents, estimated by KFT in both methanol and methanol–octanol solvent systems, were 10.6% and 14.4% for α- and β-cyclodextrin, respectively; the water content, estimated by KFT in a more hydrophilic solvent system, methanol–formamide, was 0.4–0.6% higher. Thermogravimetric evaluation of water conducts to lower values. For the essential oil/cyclodextrin complexes, the KFT water content were in the range of 6.4–8.1%, higher values being obtained in the case of Juniperus essential oil/β-cyclodextrin complexes (7.5–8.1%). With some exceptions, thermal analyses of complexes are in good agreement with the KFT results.