迭代目标转换因子分析与人工神经网络法用于邻,间,对硝基甲苯的分光光度

本文将迭代目标转换因子分析与人工神经网络法用于分光光度法同时测定邻、间、对硝基甲苯，并与目标转换因子分析的结果进行了比较。结果表明，迭代目标转换因子分析法与线性网络法的效果都很好。     

光度法同时测定地质样品中的钼钨锡锑

以水杨基荧光酮为显色剂,确定了光度分析法同时测定钼、钨锡、锑四组份的最佳条件。并采用迭代目标转换因子分析光度法,对地质样品中上述四组份进行了测定,测定结果的相对误差一般小于１０％。浓度矩阵的实验设计、主因子数的确定、干扰及消除等问题进行了讨论。     

优化迭代目标转换因子分析法在多组分混合物红外光谱解析中的应用

优化迭代目标转换因子分析法在多组分混合物红外光谱解析中的应用何锡文,陈鼎,王永泰（南开大学化学系、中心实验室,天津,３００００７１）关键词因子分析,相关性,红外光谱因子分析是一种多元统计方法,用其解析多组分混合物的红外光谱已有报道［１,２］。迭代目标...     

迭代目标转换因子分析光度法同时测定化探样品中钼、钨

本文以水杨基荧光酮为显色剂、采用迭代目标转换因子分析光度法,不经分离对化探样品中钼、钨进行同时测定。结果表明,各组分的相对误差一般小于10％,Mo、W的相对标准偏差分别为1.6％,1.5％。取得满意结果。对标准数据阵的构成、因子数的确定以及矿样分解,干扰消除等诸问题进行了探讨。     

迭代目标转换因子分析法在光度分析中的应用研究——用三溴偶氮氯膦同时测定15个稀土元素

以三溴偶氮氯膦为显色剂,应用迭代目标转换因子分析光度法对15个稀土元素混合物合成样品进行同时测定,结果的平均相对误差为2.9％～5.6％。对标准数据阵的构成。主因子数的确定以及组份间浓度比对计算结果的影响等问题作了讨论。     

改进的迭代目标转换因子分析研究

本文用非线性迭代偏最小二乘法分解原始数据阵,提出了改进的迭代目标转换因子分析法,大大简化了运算,方法简单、直观、准确.用此法对复方扑热息痛片中三组分(乙酰基水杨酸、扑热息痛及咖啡因)的含量进行紫外光谱测定,结果满意.     

迭代目标转换因子分析法及其在光度分析中的应用——轻稀土元素多组份同时测定的研究

本文提出一种新的因子分析计算方法,即迭代目标转换因子分析法,避免了对转换矩阵求逆的运算过程,降低了实验误差在计算过程中放大的可能性,从而提高了浓度计算的准确度。用本文方法对La、Ce、Pr、Nd、Sm、Eu、Gd等稀土元素7组份混合物合成样品进行了分析,获得了满意的结果。     

优化迭代目标转换因子分析法

本文采有一系列步骤，如优化波长集合与优化标准浓度阵的选择，其目的是为了尽量发掘组份间的内在差异且在计算过程中予以保持，降低组份间相关性来改善迭代目标转换因子分析法的分辨混合体系的性能。应用该优化迭代目标转换因子分析法，对二元稀土和三元氨基酸体系进行光度法多组份同时测定，结果满意。     

迭代目标转换因子分析光度法用于铜锌钴镍锰同时测定的研究

本运用迭代目标转换因子分析光度法，在ｐＨ９的介质条件下，以５－Ｂｒ－ＰＡＤＡＰ－０Ｐ体系，对Ｃｕ，Ｚｎ，Ｎｉ，Ｍｎ五组分的合成试样及地质标样进行了分析测定。结果表明，各组份的相对误差一般都小于１０５，最低检测限为３．８×１０＾－＾９－５．２×１０＾－＾９，获得满意结果。该算法避免了一般因子分析法中对转换矩阵求逆的运算工程，降低了实验误差在计算过程中被放大的可能性，从而提高了对浓度计算的准确度。     

一种新的目标转换因子分析算法用于混合物红外光谱中纯组分光谱的解析

本文针对运用目标转换因子分析从混合物红外光谱中解析出纯组分光谱的有关限制,进行了有效的改进.把抽象正交特征光谱与以单一向量为初始向量的目标转换因子分析相结合,提出了一种新的算法,结果令人满意.     

神经网络方法在血管紧张素转换酶抑制剂定量构效关系建模中的应用

对20个ACEI化合物用量子化学方法进行结构优化并计算出10个参数,用9种不同隐含层节点数的BP神经网络研究了ACEI的定量构效关系,建立了节点为10/6/1的三层BP神经网络模型。结果表明:以量化理论计算所得参数可以构建合理的ACEI定量构效关系模型,神经网络模型M6的r2=0.995,S=0.050,6个验证集化合物的残差平方和为0.002,预测能力明显强于多元线形回归模型,亦优于同类文献报道,可作为ACEI研发领域中预测先导化合物活性的理论工具。     

Comparison of several curve resolution methods for drug impurity profiling using high-performance liquid chromatography with diode array detection

The performance of five curve resolution methods was compared systematically for the identification and quantification of impurities in drug impurity profiling. These methods are alternating least-squares (ALS) with either random or iterative key-set factor analysis (IKSFA) initialisation, iterative target transformation factor analysis (ITTFA), evolving factor analysis (EFA), and heuristic evolving latent projections (HELP). Real and simulated high-performance liquid chromatography diode array detection (HPLC-DAD) data were obtained for drug mixtures containing one main compound and two impurities. The elution order of the main compound and the impurities was varied. Furthermore, resolutions were varied from 0.56 to 3.36 and impurity levels from 30% down to 0.1%. For simulated data, ALS with IKSFA initialisation and HELP perform better than ITTFA and EFA, which perform better than ALS with random initialisation. ITTFA works better than EFA for almost completely separated data, while the opposite is true for moderately or strongly overlapping data. Only ALS with IKSFA initialisation and HELP were found to resolve the required 0.1% level for moderately overlapping data. For real data, comparison of the methods provides similar results. ITTFA performs clearly better than EFA. However, none of the curve resolution methods can identify or quantify impurities at the required 0.1% level. The results for real data are worse than for simulated data because of heteroscedasticity, nonlinearity, and the acquisition resolution of the A/D-converter.     

Artificial Neural Networks Applied to the Quantitative Structure-Activity Relationship Study of Para-substituted Phenols

The artificial neural network (ANN) model with back-propagation of error is used to study the quantitative structure-activity relationship of para-substituted phenol derivatives between the biological activity and the physicochemical property parameters. Network parameters are optimized, and an empirical rule for dynamically adjusting the network's learning rate is proposed to improve the network's performance. The results showthat the three-layer ANN model gives satisfactory performance, with f(x)=1/(1+exp(-x)) as the network node's input-output transformation function and the number of hidden nodes 10. The network gives the mean square error (rose) of 0.036 when predicting the biological activity of 26 para-substituted phenol derivatives. This result compares favourably with that obtained by the conventional methods.     

Simultaneous polarographic analysis of pyrazine and its methyl derivatives by iterative target transformation factor analysis

The polarographic waves of pyrazine and its methyl derivatives are seriously overlapping, so they cannot be determined individually by polarographic methods without a prior separation. In this paper, a chemometric approach, iterative target transformation factor analysis (ITTFA), is developed and applied to the determination of mixtures of pyrazines at trace level (2.0–9.0 × 10⁻⁶moll⁻¹) by using differential pulse polarography (DPP) and a static mercury drop electrode (SMDE). Different from the general ITTFA method, only one-dimensional measurement data of n − 1 standards and an unknown were used in this work. It produced acceptable results with average recoveries in the 96–108% range and relative standard errors in the 3.4–9.5% range.     

A comparative study of diesel analysis by FTIR, FTNIR and FT-Raman spectroscopy using PLS and artificial neural network analysis

Diesel properties determined by ASTM reference methods as cetane index, density, viscosity, distillation temperatures at 50% (T50) and 85% (T85) recovery, and the total sulfur content (%, w/w) were modeled by FTIR-ATR, FTNIR, and FT-Raman spectroscopy using partial last square regression (PLS) and artificial neural network (ANN) spectral analysis. In the PLS models, 45 diesel samples were used in the training group and the other 45 samples were used in the validation. In the ANN analysis a modular feedforward network was used. Sixty diesel samples were used in the neural network training and other 30 samples were used in the validation. Two different ATR configurations were compared in the FTIR, a conventional (ATR1) and an immersion (ATR2) cell. The ATR1 cell presented the best results, with smaller prediction errors (root mean square error of prediction, RMSEP). The comparison of the three PLS models (FTIR-ATR1, FTNIR, and FT-Raman) shows that reasonable values of R² and RMSEP were obtained by the FTIR-ATR1 and FTNIR models in the evaluation of density, viscosity, and T50. The PLS/FT-Raman models presented reasonable results only for the T50 property. None of the techniques was able to generate suitable PLS calibration models for the determination of sulfur content. The ANN/FT-Raman models presented the best performances, with all models presenting R²-values above 85% some of them with RMSEP values significantly smaller than those obtained with FTIR-ATR and FTNIR. The ANN/FT-Raman and ANN/FTIR-ATR1 models were able to estimate the total sulfur content of diesel with 0.01% (w/w) accuracy.     

A comparative study of laser induced breakdown spectroscopy analysis for element concentrations in aluminum alloy using artificial neural networks and calibration methods

A comparative study of analysis methods (traditional calibration method and artificial neural networks (ANN) prediction method) for laser induced breakdown spectroscopy (LIBS) data of different Al alloy samples was performed. In the calibration method, the intensity of the analyte lines obtained from different samples are plotted against their concentration to form calibration curves for different elements from which the concentrations of unknown elements were deduced by comparing its LIBS signal with the calibration curves. Using ANN, an artificial neural network model is trained with a set of input data of known composition samples. The trained neural network is then used to predict the elemental concentration from the test spectra. The present results reveal that artificial neural networks are capable of predicting values better than traditional method in most cases.     

迭代目标转换因子分析法在吸附伏安法中的应用──混合色素的同时测定

将迭代目标转换因子分析法应用于混合色素溶液吸附伏安波谱的解析,完成了苋菜红、日落黄、柠檬黄和胭脂红混合色素在磷酸氢二钠-柠檬酸缓冲溶液(pH=5.7)中的吸附伏安法同时测定,取得了较满意的结果。     

Resolving co-eluting chromatographic patterns by means of dissimilarity analysis in iterative target transformation factor analysis

The initialization of concentration vector for iterative target transformation factor analysis (ITTFA) and identification of pure or key variables are the important issue in MCR. In this study, dissimilarity analysis and evolving factor analysis (EFA) are combined to find the selective or key variables and subsequently obtain initial estimates of the concentration vectors for resolution of gas chromatography/mass spectrometry (GC/MS) data by ITTFA. For systems containing components with highly similar mass spectra, a new constraint setting the elements out of elution window to 0 is used to improve convergence rate and accuracy of results. Tested by standard mixture of two wax esters and real GC/MS data of gasoline 97#, dissimilarity based ITTFA could obtain accurate results (average Dot product of concentration vectors, average deviation of peak area ratio and average similarity of mass spectra are 0.9929, 0.0228 and 981.0, respectively).