傅立叶变换近红外光谱法测定腐乳中总酸、蛋白质和水分

应用傅立叶变换近红外光谱技术,建立了腐乳中总酸、蛋白质和水分的分析模型。测定32份腐乳的近红外光谱数据,得到原始光谱信息,通过光谱预处理方法消除原始光谱噪声,最后采用偏最小二乘法建立回归方程。最终得到总酸、蛋白质和水分近红外光谱分析模型的决定系数(R2)依次为99.37%、99.70%、99.73%,交叉验证均方根差(RMSECV)依次为0.00871、0.11、0.0714。用该模型对11个未知腐乳样品进行外部验证,其总酸、蛋白质和水分外部验证的决定系数(R2)依次为98.74%、99.38%、99.48%,预测标准偏差(RMSEP)依次为0.00862、0.113、0.0683。内部交叉验证和外部验证均证明,近红外定量分析有较高的准确度,能满足腐乳生产中总酸、蛋白质和水分的检测精度要求。     

鱼粉近红外光谱模型传递应用研究

为探讨光栅型与傅里叶变换型近红外分析仪之间模型传递的应用效果,选取国产鱼粉为近红外光谱样本,DS2500F型近红外分析仪为源仪器,MPA型近红外分析仪为目标仪器,采用分段直接校正(PDS)方法实现近红外光谱传递。分别建立水分、粗蛋白质、粗脂肪、蛋氨酸和赖氨酸等组分的预测模型,通过交互验证决定系数(R2cv)、交互验证标准误差(RMSECV)、马氏距离(MD)、系统偏差(Bias)、预测均方根误差(RMSEP)和相对分析误差(RPD)等参数,多维度评估光谱传递后所建预测模型的效果。结果表明,DS2500F仪器的近红外光谱传递到MPA型仪器时,所建国产鱼粉的水分、粗蛋白质、粗脂肪、蛋氨酸、赖氨酸的预测模型与MPA型仪器原始预测模型各参数对比无显著差异,预测效果基本一致,说明国产鱼粉在DS2500F仪器上的近红外光谱通过传递可以替代MPA型仪器的原始光谱,间接实现了模型传递,且具有良好的适用性和共享性,可提高近红外预测模型的应用效率。     

高端近红外线水分检测仪研制成功

<正>高端近红外线水分检测仪研制成功,CSY–JH近红外水分检测仪是深圳市分析仪器制造有限公司研制的高端近红外水分检测仪,采用非接触式红外慢反射方式对样品水分进行测定,CSY–JH近红外水分检测仪采用无损检测,不破坏样品理化指标3~5 s内对样品进行快速检测水分含量。     

基于独立分量和神经网络的近红外多组分分析方法

采用小波变换对光谱数据进行压缩,用独立分量分析(ICA)方法提取近红外光谱数据矩阵的独立成分和相应的混合矩阵,再用BP神经网络对混合矩阵和实测浓度矩阵进行建模,提出了基于独立分量分析-神经网络回归(ICA-NNR)的近红外分析建模方法。进一步研究了独立分量数和网络中间隐层的神经元数对模型性能的影响,经优化后的ICA-NNR模型在相关系数与均方根误差两个指标上均优于直接用光谱矩阵作为输入所建立的模型。本方法用于玉米中水分、淀粉、蛋白质3种主要成分含量的同时测定,检验样品集的化学检测值与近红外预测值的相关系数分别达到:淀粉r=0.971,蛋白质r=0.976,水分r=0.975。     

近红外光谱法用于木薯燃料酒精品质分析

提出了用近红外光谱法测定木薯燃料酒精中乙醇和水分含量,以确定木薯燃料酒精的品质。结合修正偏最小二乘法建立了样品与其近红外光谱之间的定标模型。结果表明:乙醇定标模型交叉验证均方差和相关系数分别是0.110 2,0.960 5;水分定标模型分别是0.014 3,0.975 9。近红外光谱法预测值与化学分析法的测定值一致,该模型具有很高的预测准确性,可应用于木薯酒精品质分析的快速检测。     

在国产近红外光谱仪实验样机上用偏最小二乘法定量分析大麦成分

采用偏最小二乘法和国产近红外光谱分析仪实验样机，对大麦中的蛋白质和淀粉含量进行了定量测定．测定结果为：对蛋白质和淀粉预测的相关系数分别为０．９６１、０．９４９，相对标准偏差分别为２．７％、２．９％，相对误差分别小于５．７％、６．８％．     

傅里叶变换近红外光谱法测定大麦中蛋白质、淀粉和赖氨酸含量

采用傅里叶变换近红外光谱法测定大麦中蛋白质、淀粉、赖氨酸的含量，并用光谱影响值法(leverage)对异常值进行判断和处理。蛋白质、淀粉和赖氨酸含量近红外光谱分析模型的测定系数R。分别为0．985、0．973和0．978；检验集的化学值与模型预测值的相关系数r分别为0．9853、0．9644和0．9172，分析模型的预测相对标准偏差RSD分别为4．0％、2．4％和5．4％，该结果可替代经典分析方法，满足农产品快速分析的需要。     

蛋白质与酸性黄作用机理的研究及分析应用

在pH1.64的Clark-Lubs缓冲溶液中,牛血清白蛋白与酸性黄作用时导致吸收光谱发生变化,在532 nm附近吸光度明显降低,而在424 nm附近吸光度则呈增强趋势,其︱ΔA︱-1与cP-1呈良好线性。本文系统研究了酸性黄与蛋白质的作用机理及实验条件对酸性黄与蛋白质结合反应的影响,探讨了蛋白质与酸性黄之间的作用力。本法应用于人尿中总蛋白的测定,与经典的考马斯亮蓝G-250法结果一致。     

近红外光谱法测定成品汽油中的芳烃和烯烃含量

介绍了近红外光谱测定90#汽油及93#汽油中芳烃和烯烃含量。选择1100～1300nm的近红外光谱域,在荧光指示剂吸附法的基础上,采用偏最小二乘法建立了适合测定90#汽油及93#汽油中芳烃和烯烃含量的分析模型,通过大量试验对所建分析模型的可靠性进行了验证。近红外光谱法的测定结果与荧光指示剂吸附法的测定结果具有很好的一致性。与荧光指示剂吸附法相比,近红外光谱法可以提高分析效率,降低分析成本,具有较高的分析精密度。     

近红外波长选择结合偏最小二乘法测定混胺中微量水分

对混胺的近红外波长范围进行选择并结合偏最小二乘法建立微量水分的定量分析模型,通过配对t检验(α=0.05)验证校正模型预测的准确性.首先对混胺的性质和其近红外吸收信息进行分析,采用二阶微分(差分宽度为4nm)对光谱进行预处理.研究波长范围对微量水分测定的影响并确定了建模的最佳波长区间为940 ～ 965 nm.采用偏最...     

连续小波变换-独立成分回归算法及其在多组分分析中的应用

采用连续小波变换（CWT）对光谱数据进行处理，用独立成分分析（ICA）进行特征提取，再用回归分析方法对被测组分进行测定，建立了连续小波变换一独立成分回归（CWT-ICR）方法。方法用于肉样品中水分、脂肪和蛋白质多组分的同时测定，所得结果与化学法测得结果相符。     

Use of NIRS technology with a remote reflectance fibre-optic probe for predicting major components in cheese

In the present work the potential of near infra-red spectroscopy technology (NIRS) together with the use of a remote reflectance fibre-optic probe for the analysis of fat, moisture, protein and chlorides contents of commercial cheeses elaborated with mixtures of cow's, ewe's and goat's milk and with different curing times was examined. The probe was applied directly, with no previous sample treatment. The regression method employed was modified partial least squares (MPLS). The equations developed for the cheese samples afforded fat, moisture, protein, and chloride contents in the range 13-52%, 10-62%, 20-30%, and 0.7-2.9%, respectively. The multiple correlation coefficients (RSQ) and prediction corrected standard errors (SEP (C)) obtained were respectively 0.97 and 0.995% for fat; 0.96% and 1.640% for moisture; 0.78% and 0.760% for protein, and 0.89% and 0.112% for chlorides.     

Near and mid-infrared photoacoustic analysis of principal components of foodstuffs

Photoacoustic spectroscopy, especially in the near infrared, for principal food component analysis (protein, fat, carbohydrate and moisture contents) is discussed. The thermal diffusion length is discussed as is the problem of signal saturation. An illustrative comparison of photoacoustic spectroscopy and Fourier transform mid-infrared-attenuated total reflectance spectroscopy (FTIR-ATR) as applied to cheese analysis is given.     

Comparison of two infrared spectroscopic methods for cheese analysis

Two infrared spectroscopic methods, optothermal near infrared (NIR) spectroscopy and Fourier transform mid-infrared-attenuated total reflection (FTIR-ATR) spectroscopy, were applied to 24 cheese samples in order to obtain protein, fat and moisture contents. Reference values of the protein, fat and moisture contents in weight percent were obtained using standard wet chemistry analysis. Prediction correlation coefficients between 0.93 and 0.96 and standard errors of prediction between 2% and 5% were obtained using optothermal spectroscopy while the corresponding values for FTIR-ATR were 0.81–0.92 and 4–9%. Inhomogeneities in the cheeses, primarily due to the fat droplets, are probably the main reason for the differences in the error sizes. The superior results for optothermal spectroscopy are the more attractive because the instrument is easier to use than the FTIR-ATR instrument, it provides results more quickly with simpler statistical analysis and it is more compact and robust.     

NIR-spectrometric analysis of food. Methodical development and achievable performance values

The application of near infrared spectrometry (NIRS) to estimate the distribution of macro nutrients in food provides a rapid method for predicting food quality. Three methods have been developed for quantitative determination of the constituents of foods containing high levels of moisture (meat products, “consumable meals” and potatoes) using reflectance spectrometry in the near infrared region within 1100–2500 nm. Experience concerning these NIRS method developments is reported in order to predict e.g. the fat, crude protein and carbohydrate content. Generally, the performance values achieved so far indicate that the methods may be used as a replacement for conventional expensive and time-consuming wet chemical analysis.     

Determination of fat, moisture, and protein in meat and meat products by using the FOSS FoodScan Near-Infrared Spectrophotometer with FOSS Artificial Neural Network Calibration Model and Associated Database: collaborative study

A collaborative study was conducted to evaluate the repeatability and reproducibility of the FOSS FoodScan near-infrared spectrophotometer with artificial neural network calibration model and database for the determination of fat, moisture, and protein in meat and meat products. Representative samples were homogenized by grinding according to AOAC Official Method 983.18. Approximately 180 g ground sample was placed in a 140 mm round sample dish, and the dish was placed in the FoodScan. The operator ID was entered, the meat product profile within the software was selected, and the scanning process was initiated by pressing the "start" button. Results were displayed for percent (g/100 g) fat, moisture, and protein. Ten blind duplicate samples were sent to 15 collaborators in the United States. The within-laboratory (repeatability) relative standard deviation (RSD(r)) ranged from 0.22 to 2.67% for fat, 0.23 to 0.92% for moisture, and 0.35 to 2.13% for protein. The between-laboratories (reproducibility) relative standard deviation (RSD(R)) ranged from 0.52 to 6.89% for fat, 0.39 to 1.55% for moisture, and 0.54 to 5.23% for protein. The method is recommended for Official First Action.     

Determination of moisture, starch, protein, and fat in common beans (Phaseolus vulgaris L.) by near infrared spectroscopy

The presence of moisture, starch, protein, and fat was determined in common beans (Phaseolus vulgaris L.) by near infrared (NIR) spectroscopy without any previous sample pretreatment except grinding. A set of 96 samples was used to calibrate the instrument by modified partial least-squares regression. The following statistical results were achieved: standard error of calibration (SEC) = 0.31 and square correlation coefficient (R2) = 0.96 for moisture; SEC = 0.76 and R2 = 0.92 for starch; SEC = 0.39 and R2 = 0.98 for protein; and SEC = 0.14 and R2 = 0.80 for fat. To validate the calibration, a set of 25 bean samples was used. Standard errors of prediction were 0.39, 0.90, 0.56, and 0.13 for moisture, starch, protein, and fat, respectively, and R2 for the regression of measurements by the reference method versus NIR analysis were 0.94, 0.88, 0.94, and 0.74 for moisture, starch, protein, and fat, respectively. To compare the results obtained for all 4 components of the validation set by NIR spectroscopy with those obtained by the reference methods, linear regression and paired t tests were applied, and the methods did not give significantly different results, P = 0.05.     

Rapid determination of moisture/solids and fat in dairy products by microwave and nuclear magnetic resonance analysis

A peer-verified method is presented for the determination of percent moisture/solids and fat in dairy products by microwave drying and nuclear magnetic resonance (NMR) analysis. The method involves determining the moisture/solids content of dairy samples by microwave drying and using the dried sample to determine the fat content by NMR analysis. Both the submitting and peer laboratories analyzed various dairy products by using a CEM SMART system (moisture) and the SMART Trac (fat). The samples included milks, creams, ice cream mix, sour cream, yogurt, cream cheese, and mozzarella, Swiss, and cheddar cheeses. These samples represented a range of products that processors deal with in daily plant operations. The results were compared with moisture/solids and fat values derived from AOAC-approved methods.     

Rapid determination of moisture and fat in meats by microwave and nuclear magnetic resonance analysis--PVM 1:2003

A peer-verified method is presented for the determination of percent moisture and fat in meat products by microwave drying and nuclear magnetic resonance (NMR) analysis. The method involves determining the moisture content of meat samples by microwave drying and using the dried sample to determine the fat content by NMR analysis. Both the submitting and peer laboratories analyzed 5 meat products by using the CEM SMART system (moisture) and the SMART Trac (fat). The samples, which represented a range of products that meat processors deal with daily in plant operations, included the following: (1) fresh ground beef, high-fat; (2) deboned chicken with skin; (3) fresh pork, low-fat; (4) all-beef hot dogs; and (5) National Institute of Standards and Technology Standard Reference Material. The results were compared with moisture and fat values derived from AOAC-approved methods, 950.46 (Forced Air Oven Drying) and 960.39 (Soxhlet Ether Extraction).     

Determination of ephedrine alkaloids by capillary electrophoresis

A simple and rapid method for the simultaneous determination of six ephedrine alkaloids (ephedrine, pseudoephedrine, norephedrine, norpseudoephedrine, methylephedrine and methylpseudoephedrine) in Ephedrae herba by capillary electrophoresis was developed. A buffer solution that contained 0.005 M barium hydroxide and 0.02 M isoleucine and adjusted to pH 10.0 with ammonia solution was found to be the most suitable electrolyte for this separation. The contents of the six alkaloids in the crude drug of Ephedrae herba could be easily determined.