土壤中红外光谱库支持下的局部建模集优化

土壤中红外（MIR）光谱能快速、无污染、低成本地估算土壤有机碳等理化属性。随着各种尺度土壤光谱库的建立,使用其进行快速土壤分析引起广泛关注,但光谱库的通用模型在局部尺度上的预测效果不理想。开发“局部化”光谱建模方法是提高土壤光谱库性能的有效途径。本文提出了一种新的方法,通过光谱相似度计算和建模子集构建,旨在从库中快速建立最优局部建模集以提高预测精度。比较了欧氏、马氏、余弦三种距离算法衡量待测样本与库样本之间的相似度并生成距离矩阵;使用连续统去除法从距离矩阵中提取库容曲线中的特征点。利用偏最小二乘回归建立土壤MIR光谱与有机碳含量间的定量关系。结果表明,三种距离算法结合连续统去除得到的第一特征点均可得到较佳的预测精度。马氏距离不仅模型精度最高（R² = 0.764,RMSE = 1.021%）而且用到的库样本数最少（14%库容）。本方法可改善MIR光谱分析的成本效率并能提高局部尺度的预测能力。     

Research on Chromium Retrieval of Black Soil with Hyperspectral Imagery in Northeast of China#br#

According to the definition of spectral integral,a new spectral characteristic parameter,with the name Reversed Spectral Absorption Integral(RSAI),is proposed and used to retrieve the chromium content based on the Partial Least Squares Regression(PLSR) model.The contrastive study with other traditional spectral characteristic parameters,including differential transformation,inverse transformation,absorption area,etc.indicates that(1) the first derivation of square root transformed model can predict the chromium content quantitatively in terms of spectral transformations.(2) the stability of the absorption area model is slightly poor,and the chromium content of samples can only be estimated roughly.(3) However,as to the inversed spectral absorption integral model,the adjustment determination coefficient(Ad-R²) of the modeling and verification is 0.73 and 0.77,while the Root Mean Squared Error(RMSE) is 2.63 mg/kg and 2.36 mg/kg respectively with Relative Percent Deviation(RPD) being 3.21,which shows that the RSAI model has excellent prediction ability.So,the inversed spectral absorption integral new model can improve the accuracy and stability used to retrieve the chromium content,which provides a new idea for monitoring the chromium contamination in soil.     

基于MATLAB的PLSR方法系统开发

提出了基于MATLAB平台的偏最小二乘回归方法的系统开发,该系统可以帮助建立多因变量与多自变量之间的数学模型,尤其对于变量具有多重相关性的模型最有针对性,通过实例的仿真验证了该系统的可行性和通用性。     

Nondestructive measurement of soluble solid content of navel orange fruit by visible-NIR spectrometric technique with PLSR and PCA-BPNN

A relationship was established between the soluble solid content (SSC) of navel orange fruit determined by destructive measurement and visible-near infrared diffuse reflectance spectra in the wavelength range of 350-1800 nm. Multiplicative scatter correction (MSC) and standard normal variate correction (SNV) were applied to the spectra, partial least squares regression (PLSR) and back propagation neural network (BPNN) based on principal component analysis (PCA) were used to develop the models for predicting the SSC of intact navel orange fruit. Thirty-eight unknown samples were used to evaluate the performance of these models. The principal component analysis-back propagation (PCA-BPNN) method with MSC spectral pretreatment obtain the best predictive results, resulting in correlation coefficient, root mean square error of prediction (RMSEP), average difference between predicted and measured values (Bias) of 0.90, 0.68 °Brix and 0.16 °Brix, respectively. Experimental results indicate that PCA-BPNN is a suitable tool to model the non-linear complex system, with additional advantages over PLSR, and the vis/NIR spectrometric technique can be used for measuring the SSC of intact navel orange fruit, nondestructively.     

Simultaneous Determination of Pentaerythritol Tetranitrate and 2,4,6‐Trinitrotoluene by High Performance Thin Layer Chromatography and Partial Least Squares Regression (PLSR) Method

High performance thin layer chromatography (HPTLC) has been used for the simultaneous determination of pentaerythritol tetranitrate (PETN) and 2,4,6‐trinitrotoluene (TNT). With this aim, the spots were developed on silica gel 60 F₂₅₄ layers with petroleum ether–acetone (2 : 1 v/v). Both PETN and TNT compounds were separated from other constituent materials, and were developed at the same speed, by this solvent system. Then ultraviolet (UV) spectra of these materials were recorded with TLC‐scanner3 of CAMAG Company, and partial least squares regression‐2 (PLSR‐2) method was applied for the calibration and quantitative determination of these materials. The figure of merit (FOM) of this method was determined, and the method was applied for the analysis of an artificial sample.     

中温发酵酸乳的游离氨基酸、脂肪酸和感官特性的研究

采用正交设计试验研究了球杆菌比例、全脂乳浓度、发酵终点以及蔗糖添加量对中温发酵酸乳的风味和游离氨基酸、脂肪酸含量的影响。以9个不同的酸乳工艺配方为研究对象,采用主成分分析法(PCA)研究了不同样品的感官特征和游离氨基酸,脂肪酸含量。采用偏最小二乘回归分析法(PLSR)研究了感官结果和仪器测定结果的相关性。结果表明,球杆菌比例和全脂乳浓度是影响奶香味酸味、游离氨基酸量,脂肪酸量的重要因素。偏最小二乘回归分析表明酸乳气味与脂肪酸相关性较强,而风味和余味与游离氨基酸相关性强。