近红外光谱-径向基神经网络在异烟肼片无损定量分析中的应用

应用异烟肼片粉末的近红外漫反射光谱数据分别结合偏最小二乘法(PLS)和径向基神经网络(RBFNN)建立定量分析模型,并用所建模型对预测集样品进行了预测,结果表明:应用RBFNN所建立的定量分析模型优于PLS模型,相关系数(r)值由0.99593提高到0.99734,交互验证均方根误差(RMSECV)值由0.00523下降到0.00423,预测均方根误差(RMSEP)值由0.00614下降到0.00501。     

基于近红外光谱技术的亚麻纤维化学成分含量快速测定

在推进亚麻纤维的纺纱及其产业化生产过程中,快速、准确的定量分析纤维的化学成分是重要趋势。该研究利用近红外光谱技术分析亚麻纤维化学成分,以化学分析法测定值为对照,采用偏最小二乘法(PLS)建立亚麻纤维化学成分的近红外模型,从而实现了其化学成分的高效、快速定量分析。结果表明,建立的亚麻纤维纤维素、半纤维素、木质素和果胶近红外模型的校正相关系数(R_C)与验证相关系数(R_(CV))均在0.9以上,校正均方根误差(RMSEC)小于预测均方根误差(RMSEP)且均小于1。外部验证和双尾t检验表明模型预测结果较为准确,预测值与化学分析法得到的实测值无显著性差异,故该模型可用于相关化学成分含量的快速预测。     

藏药五脉绿绒蒿提取过程的在线近红外光谱质量控制研究北大核心CSCD

利用近红外光谱技术和自建的在线检测系统,实现了藏药五脉绿绒蒿提取过程中总黄酮含量的在线近红外光谱监测和提取终点的判定。以403个样品为建模集,分别获得了主成分回归(PCR)、偏最小二乘(PLS)、决策树(DT)、随机森林(RF)算法下的最佳光谱预处理方法和建模区间,以残差预测偏差(RPD)值为指标选择最佳建模方法。以62个样品为外部验证集,考察模型应用于总黄酮含量实时监测的可行性。此外,还探讨了利用模型预测值进行相对浓度变化率(RCCR)分析直接判定提取终点的可行性,并比较了标准偏差绝对距离法(ADSD)和移动窗口标准偏差法(MBSD)对提取终点判定的适用性。结果表明,在预处理方法为Constant+一阶导数+SG平滑、建模区间5300~9000 cm^(-1)条件下所建的总黄酮含量的PLS模型效果最好,其校正集和验证集的误差均方根均小于0.14、相关系数均大于0.97,RPD值为4.68。所建PLS模型对未知样品的平均预测率为79%,实际值与预测值的相关系数大于0.98,表明模型有较好的预测效果。外部验证集中RCCR法判定的预测提取终点和ADSD法判定的提取终点均与实际提取终点一致。所建模型性能较好,通过对未知样品进行准确快速的定量分析,实现了五脉绿绒蒿提取过程中总黄酮含量的实时监测,同时,以RCCR和ADSD作为提取终点的判定方法较为准确,可为藏药材提取过程在线近红外光谱分析技术的研究提供有益借鉴。     

双模型结合进一步降低预测均方根误差和均方根相对误差的方法

前期研究工作提出了以预测均方根相对误差最小为回归目标的方法(Minimization of prediction relative error,MPRE),它能使得预测结果的均方根相对误差更小.偏最小二乘法(Partial least squares,PLS)是以预测均方根误差为回归目标,能使得预测结果的均方根误差更小.基于多模型结合的思想,提出将MPRE与PLS相结合的双模型结合多元校正方法.本方法步骤为:(1)分别采用MPRE与PLS法对校正集建模;(2)计算阈值;(3)分别采用已建立好的MPRE与PLS模型进行预测;(4)将预测结果与阈值进行比较,得到预测结果.通过对酒精的近红外光谱与汽油紫外光谱进行定量分析结果表明,本方法可进一步减小预测均方根误差与相对误差.     

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。     

近红外光谱用于注射用复方甘草酸苷质量控制研究

该文通过采集70批注射用复方甘草酸苷样品的近红外光谱（NIRS）图,基于其中甘氨酸、甘草酸苷、盐酸半胱氨酸3种组分的含量,采用偏最小二乘法（PLS）建立定量分析模型,并进行了验证。所建模型中,甘氨酸、甘草酸苷、盐酸半胱氨酸的相关系数（R）分别为0.999 54、0.947 01和0.999 95,校正均方根误差（RMSEC）依次为0.289、0.098 0、0.001 07。13批验证集样品中上述3种组分测定值与预测值之间的平均相对误差分别为0.41%、0.72%、0.86%,测定值与预测值之间无显著性差异（P> 0.05）。     

应用近红外漫反射光谱法的猕猴桃品质无损检测北大核心CSCD

应用近红外漫反射光谱技术和化学计量学,研究成熟期猕猴桃内部品质与其近红外漫反射光谱之间的关系。在室温(24±2)℃下,采集猕猴桃赤道区域不同测试部位在4 000~10 000 cm^(-1)范围内的光谱数据,用基于平滑处理、归一化及基线校正的组合式处理方法对原始光谱进行预处理;另应用偏最小二乘(PLS)法、主成分回归法和多元线性回归法等方法分别建立猕猴桃硬度、可溶性固形物含量(SSC)的校正模型。结果表明:采用组合预处理方法和PLS法建立的校正模型精度最高;硬度校正集相关系数R_c、均方根误差RMSEC和预测集相关系数R_p、均方根误差RMSEP达到了0.976 5,0.548 3,0.943 2,0.612 7;SSC校正集相关系数R_c、均方根误差RMSEC和预测集相关系数R_p、均方根误差RMSEP达到了0.916 6,0.539 6,0.901 2,0.619 0;试验结果验证了本法的可行性。     

应用近红外漫反射光谱法的猕猴桃品质无损检测

应用近红外漫反射光谱技术和化学计量学,研究成熟期猕猴桃内部品质与其近红外漫反射光谱之间的关系。在室温(24±2)℃下,采集猕猴桃赤道区域不同测试部位在4 000~10 000 cm~(-1)范围内的光谱数据,用基于平滑处理、归一化及基线校正的组合式处理方法对原始光谱进行预处理;另应用偏最小二乘(PLS)法、主成分回归法和多元线性回归法等方法分别建立猕猴桃硬度、可溶性固形物含量(SSC)的校正模型。结果表明:采用组合预处理方法和PLS法建立的校正模型精度最高;硬度校正集相关系数R_c、均方根误差RMSEC和预测集相关系数R_p、均方根误差RMSEP达到了0.976 5,0.548 3,0.943 2,0.612 7;SSC校正集相关系数R_c、均方根误差RMSEC和预测集相关系数R_p、均方根误差RMSEP达到了0.916 6,0.539 6,0.901 2,0.619 0;试验结果验证了本法的可行性。     

基于QSAR模型的有机磷化合物毒性预测

随着有机磷化合物(OPs)的广泛应用,其在越来越多的环境介质中被检测出来。大多数OPs具有毒性,但人们缺乏快速且有效的预测手段来对毒性进行评估。本文将结合E-Dragon软件计算的分子描述符,采用不同的QSAR模型对36个OPs的毒性进行预测。文中采用后退法作为描述符筛选方法,以均方根误差(RMSE)作为评价标准,共找到14个对线性核函数支持向量机(SVM)模型贡献较大的描述符;在最终得到的SVM模型交叉验证结果中,计算值与实际值的相关系数为0. 913,均方根误差为0. 388;外部测试验证结果中,平均相对误差为9. 10%。此外,采用多元线性回归(MLR)、人工神经网络(ANN)以及偏最小二乘回归(PLS)模型对OPs的毒性进行预测,交叉验证结果显示,三个模型的计算值与实际值的相关系数分别为0. 878、0. 686与0. 620,没有SVM模型的预测能力好。因此采用线性核函数的SVM模型对OPs进行毒性预测是一个行之有效的方法。     

近红外反射光谱法分析玉米秸秆纤维素含量的研究

利用近红外反射光谱分析技术和偏最小二乘回归法（PLS）,通过比较不同光谱范围和光谱预处理方法,采用二阶导数光谱预处理,在7540．3-5361．1cm^-1和4882.9—4504．9cm^-1谱区内建立了近红外光谱测定玉米秸秆纤维素含量的校正模型。利用15个玉米秸秆样品对所建模型的实际预测效果进行了验证,预测值与化学值的相关系数（r）可达0．9953,最大相对误差仅为5．20。结果表明,近红外光谱技术可以快速、准确地测定玉米秸秆纤维素,该结果对玉米秸秆材料的快速鉴定和筛选利用具有重要的意义。     

Near-infrared spectroscopy quantitative determination of Pefloxacin mesylate concentration in pharmaceuticals by using partial least squares and principal component regression multivariate calibration

Pefloxacin mesylate, a broad-spectrum antibacterial fluoroquinolone, has been widely used in clinical practice. Therefore, it is very important to detect the concentration of Pefloxacin mesylate. In this research, the near-infrared spectroscopy (NIRS) has been applied to quantitatively analyze on 108 injection samples, which was divided into a calibration set containing 89 samples and a prediction set containing 19 samples randomly. In order to get a satisfying result, partial least square (PLS) regression and principal components regression (PCR) have been utilized to establish quantitative models. Also, the process of establishing the models, parameters of the models, and prediction results were discussed in detail. In the PLS regression, the values of the coefficient of determination (R²) and root mean square error of cross-validation (RMSECV) of PLS regression are 0.9263 and 0.00119, respectively. For comparison, though applying PCR method to get the values of R² and RMSECV we obtained are 0.9685 and 0.00108, respectively. And the values of the standard error of prediction set (SEP) of PLS and PCR models are 0.001480 and 0.001140. The result of the prediction set suggests that these two quantitative analysis models have excellent generalization ability and prediction precision. However, for this PFLX injection samples, the PCR quantitative analysis model achieved more accurate results than the PLS model. The experimental results showed that NIRS together with PCR method provide rapid and accurate quantitative analysis of PFLX injection samples. Moreover, this study supplied technical support for the further analysis of other injection samples in pharmaceuticals.     

Quantitative Analysis of Catalpol in Chinese Patent Medicine Lixin Pill by Near-infrared Diffuse Reflectance Spectroscopy

Lixin pill is a typical Chinese patent medicine with anti-rheumatic heart disease activity that has been widely used in clinical practice. Therefore it is very important to detect the concentration of catalpol, as the main component of the active ingredient. Near-infrared reflectance（NIR） spectroscopy was used to study the content of catalpol in the unprocessed Chinese patent medicine of Lixin pills. NIR is applied to quantitatively analyze 77 sam- ples, which were randomly divided into a calibration set containing 61 samples and a prediction set containing 16 samples. To get a satisfying result, partial least squares（PLS） regression was utilized to establish quantitative models. In PLS regression, the values of coefficient of determination（R2） and root mean square error of cross-validation （RMSECV） of PLS regression are 0.9419 and 0.0216, respectively. The process of establishing model, parameters of model, and prediction results were also discussed in detail（root mean square error of prediction is 0.0164）. The over- all results show that NIR spectroscopy can be efficiently utilized for the rapid and accurate analysis of routine chemical compositions in the Chinese patent medicine of Lixin pills. The prediction set suggests that this quantitative analysis model has excellent generalization ability and prediction precision. Accordingly, the result can provide tech- nical support for the further analysis of catalpol in unprocessed Lixin pill. Moreover, this study supplied technical support for the further analysis of other Chinese patent medicine samples.     

基于局部加权偏最小二乘的近红外光谱分析方法研究

针对近红外光谱分析技术中分析对象非线性现象突出的情况,提出了一种新的模型计算方法——局部加权偏最小二乘法(LWPLS)。以安胎丸为研究对象,采用LWPLS算法进行其近红外定量模型的建立,并比较偏最小二乘法(PLS)与LWPLS两种算法建立定量模型的精度。结果测得两种算法建立的校正模型中,阿魏酸的模型相关系数(R~2)分别为0.785 5、0.971 9,预测误差均方根(RMSEP)分别为0.126 6、0.043 8,相对预测误差(RE)分别为12.66%、9.18%;洋川芎内酯A的R~2分别为0.886 4、0.964 9,RMSEP分别为0.114 8、0.077 1,RE分别为14.01%、7.81%,显示LWPLS算法建立的模型精度更高。研究表明,采用LWPLS算法可提高安胎丸定量模型的准确性,具有可推广性和广泛的应用性。     

Components Analysis of Biochar Based on Near Infrared Spectroscopy Technology

This study aims to establish a rapid quantitative analysis method for biochar based on near infrared spectroscopy (NIRS) technology. Near infrared spectra of 163 samples in the 10000–3800 cm^–1 (1000–2632 nm) range were collected, and the contents of fixed carbon (FC), volatile matter (VM) and ash of samples were also analyzed. A partial least square (PLS) model for FC, VM and Ash was established after the model spectral ranges were optimized, the optimal factors were determined, and the raw spectra were pretreated by multiple scatter correction and second derivative (MSC + SD) method. Finally, the prediction performance of predictive model was evaluated. The results showed that the PLS model had a good prediction ability, and the predicted coefficient R²_p of actual values vs prediction values for FC, VM and ash were 0.9423, 0.9517 and 0.9265, respectively. Root mean square error of prediction (RMSEP) was 0.1074, 0.1201 and 0.1243, and ratios of prediction to deviation (RPD) were 3.51, 4.28 and 2.03, respectively. The PLS model had good accuracy and precision for both of FC and VM, and could be used as a quantitative method for FC and VM contents analysis. Nevertheless, PLS model need to improve the precision for Ash analysis according to RPD value. This method provides a fast and effective technical means for the quantitative analysis of biochar components.     

偏最小二乘线性拟合滴定甘蔗糖蜜中多组分有机酸

将滴定体系调节至pH 2.0,用碱标准溶液滴定至特定pH所消耗滴定荆为测量指标,构建了多组分有机酸滴定数据阵,分别以主成分回归法、偏最小二乘法以及人工神经元网络法进行多组分拟合.结果表明,偏最小二乘法的拟合结果最佳,对混合体系中乙酸、乳酸、草酸、琥珀酸、柠檬酸和乌头酸总量的相对预测均方根误差分别为5.80%、8.88%...     

Rapid and Sensitive Determination of Aminocaproic Acid Injection by Raman Spectroscopy Combined with Chemometric Models

A quantitative approach for the determination of aminocaproic acid in commercial injections based on Raman spectroscopy and chemometrics has been developed. The Raman spectra of aminocaproic acid injections were analyzed by chemometric models including classical least squares (CLS), partial least squares (PLS), principal component regression (PCR), and stepwise multiple linear regression (SMLR). To compare the quantitative ability of the models, two key parameters, difference value and root mean square error, were calculated. The results indicated that the SMLR method was more efficient than the other methods. The difference value of the SMLR method was 90.5% and the root mean square error was 2.08. Raman determinations agreed with results obtained with a standard titration method (p < 0.05). The recovery was (99.7 ± 0.58)% and the repeatability was (99.2 ± 0.67)% by the SMLR method. These results show that the chemometric modeling of Raman spectra is a specific, rapid, and convenient alternative to quantify aminocaproic acid in injections.     

基于小波系数的近红外光谱局部建模方法与应用研究

局部建模方法使用与预测样本相似的样本建立模型,可解决光谱响应与浓度之间的非线性问题,扩大模型的适用范围,提高预测准确度。采用小波变换进行数据压缩并利用小波系数之间的欧氏距离作为光谱相似性的判据,实现了近红外光谱定量分析的局部建模方法,避免了样本之间的依赖性。将所建立的方法用于烟草样品中氯含量的测定,100次重复计算得到的预测集均方根误差(RMSEP)平均值为0.0665,标准偏差(σ)为0.0045,优于全局建模和基于主成分的局部建模方法。     

Enhanced Characterization of Naproxen Formulation by Near Infrared Spectroscopy

Near infrared spectroscopy in combination with appropriate chemometric methods is an effective technique for quantitative analysis of parameters of interest for the pharmaceutical industry. In this study, the artificial neural network (ANN) was applied to monitor critical parameters (compression force, tablet hardness, mean particle size, and active pharmaceutical ingredient concentration of tablets) in the process of naproxen pharmaceutical preparation. The performance of ANN was compared to linear methods (partial least squares regression (PLS) and synergy interval partial squares (siPLS)). The ANN models for compression force, tablet hardness, mean particle size, and active pharmaceutical ingredient concentration of tablets yielded the low root mean square error of prediction (RMSEP) values of 0.936 KN, 0.302 kg, 4.49 mg, and 2.14 µm, respectively. The predictive ability of the PLS model was improved by siPLS with selection of spectral regions and the best performance among all calibration methods was showed by the nonlinear method (ANN). Effective models were built by using these approaches using near infrared spectroscopy.     

Determination of Ni2+ using an equilibrium ion exchange technique: Important chemical factors and applicability to environmental samples

The main purpose of this study was to investigate the relationship between some coffee roasting variables (weight loss, density and moisture) with near infrared (NIR) spectra of original green (i.e. raw) and differently roasted coffee samples, in order to test the availability of non-destructive NIR technique to predict coffee roasting degree. Separate calibration and validation models, based on partial least square (PLS) regression, correlating NIR spectral data of 168 representatives and suitable green and roasted coffee samples with each roasting variable, were developed. Using PLS regression, a prediction of the three modelled roasting responses was performed. High accuracy results were obtained, whose root mean square errors of the residuals in prediction (RMSEP) ranged from 0.02 to 1.23%. Obtained data allowed to construct robust and reliable models for the prediction of roasting variables of unknown roasted coffee samples, considering that measured vs. predicted values showed high correlation coefficients (r from 0.92 to 0.98). Results provided by calibration models proposed were comparable in terms of accuracy to the conventional analyses, revealing a promising feasibility of NIR methodology for on-line or routine applications to predict and/or control coffee roasting degree via NIR spectra.