拉曼光谱结合后向间隔偏最小二乘法用于调和汽油辛烷值定量分析

采用后向间隔偏最小二乘(Backward interval partial least squares,BiPLS)提取汽油拉曼光谱特征谱段,并用于研究法辛烷值(Research octane number,RON)的定量分析。实验中首先使用SPXY(Sample set partitioning based on joint x-y distances)方法划分训练集、交叉验证集和测试集,并采用稳健回归方法剔除异常的样本数据,再结合BiPLS方法筛选特征谱段,利用特征谱段建立偏最小二乘模型。与全谱段偏最小二乘模型的预测性能对比结果表明,后向间隔偏最小二乘方法可使输入模型的特征数据维数降低50.00%,交叉验证均方根误差(Root mean square error of cross validation,RMSECV)降低18.92%,预测均方根误差(Root mean square error of prediction,RMSEP)降低13.86%。后向间隔偏最小二乘方法可有效提取汽油拉曼光谱的特征谱段,降低模型复杂度,同时提高模型预测精度,在调和汽油研究法辛烷值定量分析方面有较好的应用前景。     

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PLS-红外光谱的汽油辛烷值测定方法研究

辛烷值是反映汽油抗爆性的重要指标,现有的辛烷值测试方法具有分析周期长、测试成本高等缺点。本文以红外光谱法结合偏最小二乘法(PLS)建立了汽油辛烷值快速测定方法。实验采集了113个汽油样品的光谱数据,以研究法辛烷值(RON)测得的实际辛烷值为参数,建立了预测汽油辛烷值的PLS模型。结果表明:20个预测集的相关系数Rp~2为1.0184,预测均方根误差RMSEP为0.4639。说明此方法对汽油辛烷值具有较好的预测效果,且操作简单、分析速度快,具有一定的可行性。     

基于粒子群算法的波长选择方法用于苹果酸度的近红外光谱分析

采用便携式近红外光谱分析仪,对苹果样品进行扫描获得光谱数据,运用偏最小二乘法结合基于粒子群算法的波长选择方法对苹果试验数据进行多元统计分析,建立数学模型,利用该模型对苹果酸度进行了预测。对于基于粒子群算法和全谱偏最小二乘方法,校正集样品的酸度预测值和实测值之间的相关系数分别为0.9880和0.9553,校正均方根误差分别为0.0197和0.0388;预测集样品的酸度预测值和实测值之间的相关系数分别为0.9833和0.9596,预测均方根误差分别为0.0193和0.0304。与全谱偏最小二乘法相比,基于粒子群算法的偏最小二乘法,不仅较大地减少波长变量而降低计算量,而且也较大地提高了模型性能而增强了模型预测的准确性。该方法可建立较好的定量分析模型,能广泛应用于现场或野外苹果酸度的快速分析。     

樱桃含糖量的无损检测实验研究

利用便携式可见-近红外光谱仪,研究了600~1100 nm波段内无损检测樱桃含糖量的可行性。以烟台大樱桃为研究对象,采集了每个樱桃的含糖量。利用小波去噪法对光谱数据进行预处理,并用主成分回归分析法(PCR)建立了樱桃含糖量定量分析模型。实验结果为:多尺度小波去噪法滤除了原始光谱中的噪声,同时保留了原始光谱的主要信息;所建立的主成分回归定量分析模型的校正样本集的相关系数(R)为0.9394,校正均方根误差(RMSEC)为0.1384;预测样本集的相关系数(R)为0.9071,预测均方根误差(RMSEP)为0.1495。同时与偏最小二乘回归法(PLSR)所建模型得出的预测结果相差很小。研究表明:应用便携式光谱技术在600~1100 nm范围内无损检测樱桃含糖量具有可行性,为樱桃内部品质的野外在线动态检测提供了理论依据。     

THz光谱技术定量分析异丙醇、甲醇、汽油混合物各组分

应用太赫兹时域光谱技术结合偏最小二乘法,定量分析了异丙醇、甲醇、汽油混合物中各组分含量。分析得到的甲醇建模集和预测集均方根误差分别为1.43%和2.07%,相关系数分别为0.9991和0.9980;92号汽油建模集和预测集均方根误差分别为2.38%和1.36%,相关系数分别为0.9975和0.9992;异丙醇建模集和预测集均方根误差分别为1.53%和2.84%,相关系数为0.6467和0.8954。结果表明,方法能够较好地定量分析异丙醇、甲醇、汽油混合物中的甲醇和汽油含量,而异丙醇的分析结果相关系数低。     

基于小波变换-遗传算法-偏最小二乘的草莓糖度检测研究

采用近红外漫反射光谱分析技术,对草莓糖度进行了无损检测研究。利用便携式近红外光谱仪采集草莓样品在600～1 100 nm波段内的漫反射光谱数据。首先利用小波变换(WT)多分辨率方法对光谱数据进行去噪预处理,然后利用遗传算法(GA)优选特征波长,最后运用偏最小二乘法(PLS)建立草莓糖度的WT-GA-PLS校正模型。该模型校正集的相关系数R_C为0.9395,校正集的均方根误差RMSEC为0.1615,预测集的相关系数R_P为0.9652,预测集的均方根误差EMSEP为0.5042。与全光谱模型(FS-PLS)和小波变换模型(WT-PLS)相比,该模型预测能力更强,稳健性更优。     

拉曼光谱结合改进稀疏编码器特征优选的成品油混合浓度预测方法

成品油混合浓度的预测对成品油顺序输送过程中的安全监控、混油段分割具有重要的意义。本研究配制92#汽油-3#航煤以及3#航煤-0#车柴两组包含不同浓度的混合样品,并对其进行拉曼光谱采集;依次采用归一化、多元散射校正、BaselineWavelet基线校正3种光谱预处理方法进行优化;之后采用改进的栈式稀疏自编码器(Stacked Sparse Autoencoder, SSAE)模型对预处理之后的拉曼光谱进行稀疏特征提取,并结合全连接层进行回归预测;最后根据均方根误差(Root Mean Square Error, RMSE)和决定系数(R²)两项评价指标,与偏最小二乘回归(Partial Least Square Regression, PLSR)、最小二乘支持向量回归(Least Square Support Vector Machine, LSSVR)以及SSAE 3种模型进行对比。结果表明：改进的SSAE-FC模型表现出更优的预测精度和稳定性,92#汽油-3#航煤混油测试集的R²和RMSEC指标分别为0.9952和0.8932,3#航煤-...     

拉曼光谱法测定芳烃物料的馏程

应用拉曼光谱法测定了芳烃物料的馏程.采用芳烃样品60个,其中50个为校正样品集,10个为预测样品集,在拉曼光谱位移为400 ～1 800 cm~(-1)范围内进行光谱预处理,并应用偏最小二乘回归法(PLS)建立了各馏程的校正模型,其相关系数(r~2)分别为0.87、 0.89、0.98、0.97、0.94、0.89、0.88(相应蒸馏回收百分数分别为5%、10%、30%、50%、70%、90%、95%).在置信水平99.5%,α为0.005时,各馏程t值均小于3.69(临界值),表明拉曼光谱法预测结果与常压蒸馏法的测定结果无显著性差别.采用拉曼光谱技术可以快速测定芳烃物料的馏程.     

Quality assessment of gasoline using comprehensive two‐dimensional gas chromatography combined with unfolded partial least squares: A reliable approach for the detection of gasoline adulteration

Comprehensive two‐dimensional gas chromatography and flame ionization detection combined with unfolded‐partial least squares is proposed as a simple, fast and reliable method to assess the quality of gasoline and to detect its potential adulterants. The data for the calibration set are first baseline corrected using a two‐dimensional asymmetric least squares algorithm. The number of significant partial least squares components to build the model is determined using the minimum value of root‐mean square error of leave‐one out cross validation, which was 4. In this regard, blends of gasoline with kerosene, white spirit and paint thinner as frequently used adulterants are used to make calibration samples. Appropriate statistical parameters of regression coefficient of 0.996–0.998, root‐mean square error of prediction of 0.005–0.010 and relative error of prediction of 1.54–3.82% for the calibration set show the reliability of the developed method. In addition, the developed method is externally validated with three samples in validation set (with a relative error of prediction below 10.0%). Finally, to test the applicability of the proposed strategy for the analysis of real samples, five real gasoline samples collected from gas stations are used for this purpose and the gasoline proportions were in range of 70–85%. Also, the relative standard deviations were below 8.5% for different samples in the prediction set.     

Characterization of Gasoline by 1H Nuclear Magnetic Resonance and Chemometrics

Automotive fuel adulteration is an old and significant problem. One common type of fuel adulteration is the addition of diesel to gasoline. Unsupervised models were developed through hierarchical cluster and principal component analysis models. Supervised models through partial least square discriminant analysis using ¹H nuclear magnetic resonance spectra as the input were used to classify samples as adulterated or unadulterated. Quantitative models were developed using partial least squares to determine the gasoline and diesel concentrations in the samples. This set contained samples composed of pure gasoline and anhydrous ethanol reproducing commercial gasoline and other samples treated with diesel. Hierarchical cluster and principal component analysis did not distinguish between adulterated and unadulterated samples except for the most adulterated materials. However, partial least square discriminant analysis classified 100% of the samples correctly. The partial least square algorithm provided excellent regression models for the gasoline and diesel content. The determination coefficient was 0.9920 for both models, whereas the root mean square error of cross-validation and root mean square error of prediction for the diesel model were 2.32 and 1.42%, respectively, and 2.40 and 1.38% for the gasoline model.     

Determination of Basic Nitrogen in Residues of Crude Oil Distillation Using ATR-FTIR and Chemometric Methods

A method for basic nitrogen determination in residues of crude oil distillation using infrared spectroscopy and chemometrics algorithms was developed. Interval partial least squares, synergy interval partial least squares, and backward interval partial least squares were evaluated for calibration model construction. The samples were divided into a calibration and prediction set containing 40 and 15 samples, respectively. The first derivative with a Savitzky-Golay filter and the mean centered data showed the best results and were used in all calibration models. The backward interval partial least squares algorithm with spectra divided in 60 intervals and combinations of 4 intervals (1407 to 1372; 1117 to 1082; 971 to 936; 914 to 879 cm^?1) showed the best root mean square error of prediction of 0.016 wt%. This calibration model displayed a suitable correlation coefficient between reference and predicted values.     

Simultaneous Quantification of Three Polymorphic Forms of Carbamazepine using Raman Spectroscopy and Multivariate Calibration

Carbamazepine is a pharmaceutical product used to treat epilepsy and bipolar disorder. Some active pharmaceutical ingredients, such as carbamazepine, present polymorphism that may alter the bioavailability. Consequently, the determination of different polymorphic forms has become important for the pharmaceutical industry. In this work, polymorphic forms were synthesized and characterized by differential scanning calorimetry and X-ray diffraction. Raman spectroscopy was used to quantify mixtures of the three common polymorphic forms of carbamazepine. A ternary mixture design was used to create the calibration set of ten samples and six levels of concentration for each polymorph. Partial least squares was performed to build the prediction models. Ten spectra were obtained to obtain representative Raman spectra of the mixtures. The calibration models were built using the average spectra, and an external set of samples was used to evaluate the models. The partial least squares model gave a root mean square error of prediction of 6.2% for carbamazepine I, 6.8% for carbamazepine III, and 11.6% for carbamazepine dihydrate. The results showed that good results were obtained for the solid state characterization of the mixtures of polymorphs using a fast strategy for simultaneous analysis.     

Cellulose I crystallinity determination using FT–Raman spectroscopy: univariate and multivariate methods

Two new methods based on FT–Raman spectroscopy, one simple, based on band intensity ratio, and the other using a partial least squares (PLS) regression model, are proposed to determine cellulose I crystallinity. In the simple method, crystallinity in cellulose I samples was determined based on univariate regression that was first developed using the Raman band intensity ratio of the 380 and 1,096 cm^?1 bands. For calibration purposes, 80.5% crystalline and 120-min milled (0% crystalline) Whatman CC31 and six cellulose mixtures produced with crystallinities in the range 10.9–64% were used. When intensity ratios were plotted against crystallinities of the calibration set samples, the plot showed a linear correlation (coefficient of determination R ² = 0.992). Average standard error calculated from replicate Raman acquisitions indicated that the cellulose Raman crystallinity model was reliable. Crystallinities of the cellulose mixtures samples were also calculated from X-ray diffractograms using the amorphous contribution subtraction (Segal) method and it was found that the Raman model was better. Additionally, using both Raman and X-ray techniques, sample crystallinities were determined from partially crystalline cellulose samples that were generated by grinding Whatman CC31 in a vibratory mill. The two techniques showed significant differences. In the second approach, successful Raman PLS regression models for crystallinity, covering the 0–80.5% range, were generated from the ten calibration set Raman spectra. Both univariate-Raman and WAXS determined crystallinities were used as references. The calibration models had strong relationships between determined and predicted crystallinity values (R ² = 0.998 and 0.984, for univariate-Raman and WAXS referenced models, respectively). Compared to WAXS, univariate-Raman referenced model was found to be better (root mean square error of calibration (RMSEC) and root mean square error of prediction (RMSEP) values of 6.1 and 7.9% vs. 1.8 and 3.3%, respectively). It was concluded that either of the two Raman methods could be used for cellulose I crystallinity determination in cellulose samples.     

基于组合偏最小二乘的特征波段优选方法在氨、碱化处理玉米秸秆粗蛋白检测中的研究

该文构建了玉米秸秆粗蛋白定量分析模型,并对光谱特征波段选取方法进行探讨及验证。首先对107个样本进行预处理,剔除两个异常样本后采用DB2小波缺省阈值4层分解方式进行光谱重构,预处理后粗蛋白模型交互验证决定系数R2CV从0.788 9提高至0.920 8,采用间隔偏最小二乘(IPLS)及其改进型方法后向区间间隔偏最小二乘(BIPLS)、组合间隔偏最小二乘(SIPLS)进行特征波段选取,并对比主成分分析、竞争性自适应重加权采样法、相关系数法、遗传算法、移动窗口最小二乘等结果,发现基于IPLS及其改进型BIPLS、SIPLS均可有效、准确定位特征波段区间,其中采用SIPLS 30 波段间隔在10 128~10 398 cm-1与11 196~11 462 cm-1时具有最优模型,验证集相关系数(rp)为0.978 4,验正集决定系数(R2P)为0.957 2,验正集均方误差根(RMSEP)为0.221 1,相比于其他波段选取方法表现出较好的实时准确性,该方法可为玉米秸秆氨碱化最优条件判定提供重要的数据支撑。     

Characterization of Gasoline by Raman Spectroscopy with Chemometric Analysis

A rapid Raman spectroscopy protocol is reported to classify gasoline according to its distributor and to identify and quantify common adulterants. Gasoline from three distributors was collected from 19 stations in São Paulo, Brazil. Principal component analysis (PCA) showed specific clusters for each distributor, and partial least squares discriminant analysis (PLS-DA) correctly identified the origin of the samples. To evaluate the technique for the identification and quantification of the adulterants, authentic samples from each distributor were fortified at levels from 2.5 up to 25.0% (v/v) using ethanol, methanol, toluene, and turpentine to obtain 120 altered samples. PCA showed clear separation among the samples with the adulterants and PLS-DA precisely identified the adulterants (478 in 480 predictions by cross-validation), irrespective of the distributor and the concentration. One classification model was used to characterize all distributors. To quantify the adulterants, 36 multivariate calibration models were constructed using partial least squares (PLS), interval PLS, and PLS genetic algorithm for each distributor and for each adulterant. Cross-validation errors of less than 5.0% were obtained for all adulterants regardless of the distributor. Raman spectroscopy and multivariate analysis were shown to be powerful for rapid and inexpensive for the characterization of gasoline origin and the identification and quantification of common adulterants.     

Influence of different laser irradiation methods on determination of composition content in polypropylene/low-density polyethylene blends using Raman spectroscopy

Sample movement makes a difference to raw Raman spectra and determination of composition content using Raman spectroscopy. Therefore, it is necessary to have further studies in this aspect. In this paper, different laser irradiation methods were investigated for determination of composition content in polypropylene (PP)/low-density polyethylene (LDPE) blends using Raman spectroscopy. Raw Raman spectra of PP sample were firstly collected using different laser irradiation methods. It was shown that the relative standard deviations (RSD) of PP sample under circle irradiation were ten times bigger than that under point irradiation at the little sacrifice of signal-to-noise ratio (SNR). In other words, rotating (or moving) PP sample during Raman spectra collection could signally improve sample representation. Owing to this, in combined with partial least squares (PLS), Raman quantitative analysis of PP concentration in PP/LDPE blends were performed by different laser irradiation methods. The results validated that blend samples with rotation during Raman measurement led to lower prediction errors in prediction of PP concentration. The best root-mean-square error of prediction (RMSEP) and coefficient of determination (R²) that were obtained for PP were respectively 2.10% and 0.9884.     

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.     

基于局部最小二乘支持向量机的光谱定量分析

提出了一种基于局部最小二乘支持向量机(LSSVM)的回归方法,以克服待测参数和光谱数据间的非线性。本方法首先通过欧式距离选取局部训练样本子集,然后利用该子集建立LSSVM校正模型。由于每个测试样本建模时要选取不同的训练样本,因此提出相对距离的概念用来改进高斯核函数,使LSSVM的参数对于不同的训练样本具有自调整功能。针对一批汽油样本的实验结果表明,本方法的预测精度优于常见的局部线性建模方法和全局建模方法。