便携式近红外光谱仪对废旧涤/棉混纺织物中涤含量的快速检测

利用近红外光谱技术对252个涤/棉混纺织物进行研究,建立了不同光谱特征的涤/棉混纺织物的偏最小二乘(PLS)定量分析模型。将近红外光谱异常样本与光谱正常样本分别建模,显著提高了定量分析模型的预测精度、拓宽了模型的适用范围。以涤、棉主要吸收峰区间为基本建模波段,进行双向扩展,筛选出最佳建模波段,以相关系数(R)、预测集标准差(SEP)和验证集准确率优化建模条件,并与未分别建模的PLS模型相比较。用346个未参与建模的废旧涤/棉混纺织物对模型进行外部验证,外部验证准确率为92%,识别时间8s。     

近红外光谱法快速测定柴胡提取过程中的药效成分

利用近红外光谱(NIRS)技术对柴胡提取过程中的药效成分进行快速定量分析。共收集126个柴胡提取液样品,采用紫外-可见分光光度法测定总黄酮和多糖的含量,高效液相色谱法(HPLC)测定柴胡皂苷A及柴胡皂苷D的含量,以透射模式采集提取液的近红外光谱,运用偏最小二乘法(PLS)分别建立了近红外光谱与4种药效指标参考值之间的定量校正模型,并采用不同的预处理方法、光谱波段和主因子数对模型进行优化。结果表明,总黄酮、多糖、柴胡皂苷A和柴胡皂苷D 4种定量模型的近红外预测值与参考值之间的拟合性良好,模型预测精度较高,其中预测集相关系数(RP)均大于0.9;预测集误差均方根(RMSEP)分别为3.46 μg/mL、0.743 mg/mL、1.53 μg/mL、0.406 μg/mL;预测集相对偏差(RSEP)分别为1.65%、8.28%、5.74%、7.52%。该研究证实了NIRS结合PLS可成功应用于监测柴胡提取液中药效成分的含量变化,且方法具有快速、准确、无损和环保的特点。     

近红外光谱法同时测定通天口服液中天麻素和芍药苷的含量

本文应用近红外光谱结合偏最小二乘法建立了同时测定通天口服液中天麻素与芍药苷含量的方法。以高效液相色谱(HPLC)法测定通天口服液样品中天麻素和芍药苷的化学参考值,随机抽取60个样本作校正集,20个样本作预测集。用偏最小二乘法(PLS)将校正集样本的近红外光谱与相应样本的天麻素和芍药苷含量分别相关联建立模型。结果表明,天麻素和芍药苷校正模型的决定系数分别为96.28%、94.55%,模型的交叉验证均方差分别为0.0336、0.00908,预测集的决定系数分别为94.23%、92.86%,预测集均方差分别为0.0453、0.00839。同时还做了模型的精密度实验,该方法能用于大批量样品的快速分析。     

废旧聚酯/棉混纺织物的在线近红外定量分析与自动分选

在前期探究的最佳测试条件下,利用自主研制的“纤维制品主体组分高效识别与分选装置”对废旧聚酯/棉混纺织物样品进行在线原始近红外光谱采集。基于在线原始谱图,探讨出最佳光谱预处理方法为S-G平滑+最大最小归一化(MMN)+S-G导数,并利用偏最小二乘法建立了废旧聚酯/棉混纺织物的在线近红外定量分析模型,模型的交互验证均方根误差(RMSECV)为1.47,校正相关系数(RC)、验证相关系数(RV)值均不小于0.99,校正相对预测偏差(RPDC)为18.17,验证相对预测偏差(RPDV)为13.13,交互验证相对预测偏差(RPDCV)为11.76。为验证模型的可靠性,选取30个外部样本进行在线验证,验证结果的线性方程为y=(1.00±0.01)x-(0.88±0.56),预测准确率为93.3%。将模型导入分选装置的“纺织品在线主控程序”后,对设备设定不同聚酯含量织物的分选类别,即可对废旧聚酯/棉混纺织物样本进行含量预测,并通过装置的吹分分选系统将样品自动吹扫到相应的收集框中。每个样品预测并分选的时间小于2 s,机械自动分选结果无误。利用所建模型和分选装置可对废旧聚酯/棉混纺织物进行在线高效测定与自动分选。     

偏最小二乘回归用于近红外光谱分析的稳健策略

偏最小二乘法（PLS）在近红外光谱（NIR）定量分析中占有重要地位,但预测结果往往容易受到样本分组和奇异样本等因素的影响,稳健性不强。多模型PLS (EPLS）方法在模型稳健性上得到提高,然而它无法识别样本中存在的奇异样本。为了同时提高模型的预测准确性和稳健性,本文提出了一种根据取样概率重新取样的多模型PLS方法,称为稳健共识PLS（RE-PLS）方法。该方法通过迭代赋权偏最小二乘法(IRPLS)计算样本回归残差得到每个校正集样本的取样概率,然后根据样本的取样概率来选择训练子集建立多个PLS模型,最后将所有PLS模型的预测结果平均作为最终预测结果。该方法用于两种不同植物样品的近红外光谱建模,并与传统的PLS及EPLS方法进行比较。结果表明该方法可以有效的避免校正集中奇异样本对模型的影响,同时可以提高预测精确度和稳健性。对于含有较多奇异样本的,复杂近红外光谱烟草实际样本,利用简单PLS或者EPLS方法建模预测效果不是很理想,而RE-PLS凭借其独特优势则有望在这种复杂光谱定量分析中得到广泛的应用。     

采用多光程长建模方法检测血液成分含量

为了提高近红外光谱血液成分含量分析模型的预测精度,利用多个光程长(optical path length,OPL)共同参与建模的方法进行血糖等6种血液成分的定量分析。通过微米位移机构实现不同光程长血液光谱的测量,由全自动生化分析仪给出生化成分分析结果,并出具化验单。采用偏最小二乘法(PLS2)进行血液的近红外光谱建模及预测。由于血液光谱存在显著的非线性特征,不同光程长的血液样本的等效吸收系数不同,同一波长不同光程长(0.20~1.25 mm)测得的血液光谱互不相关。主动把非线性特性作为一种测量手段引入,不再利用单个的最佳光程长建模,而是用各个血液组分对应的多个最佳光程长的近红外光谱同时参与建立校正模型,进行血液成分的分析预测。研究结果表明,多光程长建模方法用于血液成分含量分析,可提高血液成分校正模型的预测精度。     

近红外光谱结合膜富集技术测定饮料中微量邻苯二甲酸二异辛酯的含量

提出了一种基于在线膜富集的近红外漫反射光谱技术,对饮料中的微量塑化剂邻苯二甲酸二异辛酯(DEHP)进行快速检测。采用聚醚砜膜对饮料中的DEHP进行富集,将富集DEHP的膜直接进行近红外漫反射检测。参考DEHP的透射近红外光谱,对波数进行选择,以4 420～4 060、4 700～4 540、6 040～5 600cm-1作为建模的波数区间。通过比较原始光谱、多元散射校正、一阶求导、二阶求导及其组合,考察了光谱预处理方法对模型的影响,用去一交互验证法建立了偏最小二乘(PLS)模型,并用所建立的校正模型对校正集样品进行了预测。结果表明,在选定的波数区间,当用一阶求导对校正集光谱进行预处理时,所建立的模型对校正集的预测效果最佳,在隐变量数为7时,对校正集所有样品的校正均方根误差(RMSEC)为0.188 7mg/L。用此模型对预测集样品进行预测时,DEHP的质量浓度在0.5～5.0 mg/L范围内,预测均方根误差(RMSEP)为0.232 4 mg/L,平均相对预测误差为6.29%。     

支持向量机回归法在近红外光谱测定植物纤维原料中甲氧基含量中的应用

以26个植物纤维原料为实验材料,由20个样品作校正样品,采用径向基核函数方法对纤维原料中甲氧基含量与纤维原料样品近红外光谱进行支持向量机(SVM)回归建模.以所建SVM回归模型对6个纤维原料样品中甲氧基含量进行预测,回归模型的预测结果与采用改良的维伯克法确定的甲氧基含量的相关系数为0.977,预测样本集的标准偏差为0.43.将SVM回归模型的预测效果与PLS回归模型的预测结果进行比较,所建近红外光谱测定植物纤维原料中甲氧基含量的SVM回归模型可用于实际植物纤维原料样品的定量分析,且具有较好的分析效果.     

近红外光谱技术定量分析连作滁菊土壤中的阿魏酸含量

应用近红外光谱(NIRS)技术定量分析连作滁菊土壤样品中阿魏酸的含量.通过标准杠杆值、学生残差和马氏距离判断异常光谱,经二阶导数和Norris平滑滤噪预处理后,在6000~4000 cm-1范围,最佳因子数为7,采用偏最小二乘法(PLS)构建数学模型.结果表明,模型校正集和验证集与高效液相色谱仪(HPLC)测定的参考值之间均呈现良好相关关系,校正相关系数Rc为0.9914,交叉验证相关系数Rcv为0.9935,校正集误差均方根(RMSEC)为0.484,预测误差均方根(RMSEP)为0.539,交叉验证误差均方根(RMSECV)为0.615.研究结果表明,NIRS分析技术能够实现连作土壤中阿魏酸的快速检测,结果准确可靠.     

近红外光谱用于注射用曲克芦丁质量控制研究

运用近红外光谱(NIRS)分析技术建立注射用曲克芦丁的定量分析预测模型,并进行相关验证。近红外积分球漫反射光谱法对73个批次样品进行采集得到NIRS图,根据高效液相色谱法(HPLC)测定样品的有效成分含量并测得样品pH值,使用TQ Analyst光谱分析软件运用偏最小二乘法(PLS)建立定量分析方法模型。模型结果显示,含量预测模型的线性相关系数(R)为0.99239,校正标准偏差(RMSEC)和预测标准偏差(RMSEP)分别为0.1790和0.5460,pH值预测模型R为0.81575,RMSEC和RMSEP分别为0.0749和0.0704。验证结果显示,含量与pH值预测模型外部验证结果的平均误差率分别为0.46%和1.57%,且t检验无显著性差异(P>0.05)。NIRS分析技术可快速稳定的对产品含量和pH值进行预测,对产品质量控制提供了新的方法依据。     

Combined wavelet transform-artificial neural network use in tablet active content determination by near-infrared spectroscopy

The pharmaceutical industry faces increasing regulatory pressure to optimize quality control. Content uniformity is a basic release test for solid dosage forms. To accelerate test throughput and comply with the Food and Drug Administration's process analytical technology initiative, attention is increasingly turning to nondestructive spectroscopic techniques, notably near-infrared (NIR) spectroscopy (NIRS). However, validation of NIRS using requisite linearity and standard error of prediction (SEP) criteria remains a challenge. This study applied wavelet transformation of the NIR spectra of a commercial tablet to build a model using conventional partial least squares (PLS) regression and an artificial neural network (ANN). Wavelet coefficients in the PLS and ANN models reduced SEP by up to 60% compared to PLS models using mathematical spectra pretreatment. ANN modeling yielded high-linearity calibration and a correlation coefficient exceeding 0.996.     

Development of a univariate calibration model for pharmaceutical analysis based on NIR spectra

Near-infrared spectroscopy (NIRS) has been widely used in the pharmaceutical field because of its ability to provide quality information about drugs in near-real time. In practice, however, the NIRS technique requires construction of multivariate models in order to correct collinearity and the typically poor selectivity of NIR spectra. In this work, a new methodology for constructing simple NIR calibration models has been developed, based on the spectrum for the target analyte (usually the active principle ingredient, API), which is compared with that of the sample in order to calculate a correlation coefficient. To this end, calibration samples are prepared spanning an adequate concentration range for the API and their spectra are recorded. The model thus obtained by relating the correlation coefficient to the sample concentration is subjected to least-squares regression. The API concentration in validation samples is predicted by interpolating their correlation coefficients in the straight calibration line previously obtained. The proposed method affords quantitation of API in pharmaceuticals undergoing physical changes during their production process (e.g. granulates, and coated and non-coated tablets). The results obtained with the proposed methodology, based on correlation coefficients, were compared with the predictions of PLS1 calibration models, with which a different model is required for each type of sample. Error values lower than 1-2% were obtained in the analysis of three types of sample using the same model; these errors are similar to those obtained by applying three PLS models for granules, and non-coated and coated samples. Based on the outcome, our methodology is a straightforward choice for constructing calibration models affording expeditious prediction of new samples with varying physical properties. This makes it an effective alternative to multivariate calibration, which requires use of a different model for each type of sample, depending on its physical presentation.     

Measurement of lactate in whole human blood with near-infrared transmission spectroscopy

We have evaluated the potential of near-infrared spectroscopy (NIRS) as a technique for rapid analysis of lactate in whole blood. To test the NIRS technique, a comparison was made with a standard clinical method using whole blood samples taken from five exercising human subjects at three different stage of exercise. To expand lactate concentration within the physiological range, standard additions method was used to generate 45 unique data points. Spectra were collected over the 2050-2400 nm spectral range with a 1 mm optical path length quartz cell. Reference lactate concentrations in the samples were determined by enzymatic measurements. Estimates and calibration of the lactate concentration with NIRS was made using partial least squares (PLS) regression analysis and leave-N-out cross validation on second derivative spectra. Separate calibrations were determined from each of the subject samples and cumulative PRESS was used to determine the number of PLS factors in the final model. The results from the PLS model presented are generated from the five individual calibration coefficient vectors and provided a correlation coefficient of 0.978 and a standard error of cross validation of 0.65 mmol l⁻¹ between the enzymatic assay and the NIRS technique. To study the parameters that impact the spectra baseline and the correlation between the calculated model and the data, referenced measurements of lactate against baseline spectrum were made for each individual. A correlation coefficient of 0.992 and a standard error of cross validation of 0.21 mmol l⁻¹ were found. The results suggest that NIRS may provide a valuable tool to assess physiological status for both research and clinical needs.     

独立分量分析预处理法提高苹果糖度模型预测精度研究

为了提高苹果近红外光谱糖度预测模型精度,利用独立分量分析方法(ICA)对苹果近红外光谱进行了预处理,并且建立了糖度的偏最小二乘(PLS)预测模型。结果表明,独立分量分析不但能分离出噪声信号,而且所分离出来的光谱信号也比原始光谱信号光滑。在预处理后的最佳PLS糖度模型校正时的相关系数rc和标准偏差SEC分别为0.9549和0.3361,用于预测时的相关系数rp和标准偏差SEP分别为0.9071和0.4355。与普通的平均处理法的PLS模型相比,其精度有所提高,且模型更加简洁。     

Rapid determination of sucrose in chocolate mass using near infrared spectroscopy

This paper reports the results of a rapid method to determine sucrose in chocolate mass using near infrared spectroscopy (NIRS). We applied a broad-based calibration approach, which consists in putting together in one single calibration samples of various types of chocolate mass. This approach increases the concentration range for one or more compositional parameters, improves the model performance and requires just one calibration model for several recipes. The data were modelled using partial least squares (PLS) and multiple linear regression (MLR). The MLR models were developed using a variable selection based on the coefficient regression of PLS and genetic algorithm (GA). High correlation coefficients (0.998, 0.997, 0.998 for PLS, MLR and GA-MLR, respectively) and low prediction errors confirms the good predictability of the models. The results show that NIR can be used as rapid method to determine sucrose in chocolate mass in chocolate factories.     

近红外光谱法测定端羟基环氧乙烷-四氢呋喃共聚醚的羟值

提出了用近红外光谱测定端羟基环氧乙烷-四氢呋喃共聚醚(PET)的羟值,结合主成分回归和偏最小二乘法建立了PET羟值与其近红外光谱之间的关联模型。结果表明,近红外光谱法与化学分析法的测定结果一致;近红外光谱法测定PET羟值的相对误差在5%以内;利用遗传算法选择部分波长建立校正可以降低模型的预测误差。     

近红外光谱法快速测定附子中6个生物碱的含量

应用近红外光谱(NIRS)技术结合偏最小二乘(PLS)和最小二乘支持向量机(LS-SVM)建立了附子中多指标成分的快速无损检测方法。选取38批样品建立了同时测定附子样品中6种成分含量的高效液相色谱(HPLC)方法;通过采集附子样品的NIRS图,分别采用PLS和LS-SVM建立了各个成分HPLC测定值与NIRS图的定量校正模型。所建立的苯甲酰新乌头原碱、苯甲酰乌头原碱、苯甲酰次乌头原碱、新乌头碱、次乌头碱、乌头碱、单酯型生物碱总量和双酯型生物碱总量LS-SVM模型的相对预测偏差(RPD)分别为3.3、3.2、4.1、7.7、8.8、7.6、4.0和8.6;验证集相关系数(rpre)分别为0.9486、0.9475、0.9668、0.9909、0.9946、0.9969、0.9669和0.9927,且LS-SVM模型优于PLS模型,说明NIRS模型验证集与HPLC测定值具有良好的非线性关系,模型预测效果良好。采用NIRS技术结合LS-SVM模型可以快速对附子中的上述6个生物碱含量以及单酯型生物碱总量和双酯型生物碱总量进行检测,方法操作简便,对控制附子中的生物碱含量具有一定的指导作用。     

高粱籽粒中多酚类物质的傅立叶变换近红外光谱分析

利用高效液相色谱(HPLC)法测定高粱籽粒中阿魏酸、原儿茶醛和花青素的含量,比色法测定总酚、总黄酮、缩合单宁的含量;运用偏最小二乘法建立NIR光谱与HPLC法和比色法分析值之间的多元校正模型,预测高粱籽粒中主要酚类物质的含量.结果表明,各成分近红外预测值与实测值之间的校正模型相关系数(R)、内部交叉验证均方差(RMSECV)、最佳主因子数分别为:总酚0.9737, 0.288, 4;总黄酮0.9660, 0.00671, 8;缩合单宁0.9558, 0.0289, 6;阿魏酸0.9818, 0.0391, 6;原儿茶醛0.9979, 0.0118, 5;花青素0.9977, 0.0523, 4;预测相对偏差(RSEP)分别为:总酚6.99%、总黄酮4.54%、 缩合单宁7.13%、阿魏酸2.68%、原儿茶醛5.46%、 花青素5.81%.结果表明,模型对样品NIR的预测值与其相应的化学值有较好的相关性,此模型可用来预测高粱籽粒中的各酚类物质的含量,在高粱优质育种和品质分析中具有广泛的应用价值.     

Fast determination of trace dimethyl fumarate in milk with near infrared spectroscopy following fluidized bed enrichment

Near infrared spectroscopy (NIRS) has been proved to be a powerful analytical tool in different fields. However, because of the low sensitivity in near infrared region, it is a significant challenge to detect trace analytes with normal NIRS technique. A novel enrichment technique called fluidized bed enrichment has been developed recently to improve sensitivity of NIRS which allows a large volume solution to pass through within a short time. In this paper, fluidized bed enrichment method was applied in the determination of trace dimethyl fumarate in milk. Macroporous styrene resin HZ-816 was used as adsorbent material, and 1?L solution of dimethyl fumarate was run to pass through the material for concentration. The milk sample was pretreated to remove interference matters such as protein, fat, and then passed through the material for enrichment; after that, diffuse reflection NIR spectra were measured for the analyte concentrated on the material directly without any elution process. The enrichment and spectral measurement procedures were easy to operate. NIR spectra in 900–1,700?nm were collected for dimethyl fumarate solutions in the concentration range of 0.506–5.060?μg/mL and then used for multivariate calibration with partial least squares (PLS) regression. Spectral pretreatment methods such as multiplicative scatter correction, first derivative, second derivative, and their combinations were carried out to select the optimal PLS model. Root mean square error of cross-validation calculated by leave-one-out cross-validation is 0.430?μg/mL with ten PLS factors. Ten samples in an independent test set were predicted by the model with the mean relative error of 5.33?%. From the results shown in this work, it can be concluded that the NIR technique coupled with on-line enrichment method can be expanded for the determination of trace analytes, and its applications in real liquid samples like milk and juice may also be feasible.