模式识别用于压电晶体传感器阵列识别可燃物质

用７个压电晶体组成传感器阵列,每个晶体上分别涂有不同种类的气相色谱固定液,通过测定各种可燃物质燃烧时放出的混合气体来识别所燃物质,在识别中分别应用了人工神经网络法（ＡＮＮ）和逐步判别分析法（ＳＤＡ）。讨论了解决神经网络开始训练时不收敛或产生麻痹现象的方法,提出了训练数据选取的新方法－训练集逐步扩展法,实验证明：人工神经网络对被测物质的识别准确率达１００％,高于逐步判别分析法（８３％）。     

Simultaneous Quantification of Heavy Metals Using a Solid State Potentiometric Sensor Array

A potentiometric sensor array of four nonspecific electrodes with solid‐state membranes is developed and tested for simultaneous analysis of copper(II), mercury(II), and silver(I) ions. The cross‐sensitivity responses of the sensors for these ions are evaluated. The array potentiometric signals are processed by partial least‐squares regression (PLS) and back propagation artificial neural networks (ANN) to determinate analyte concentrations. The ANN configuration is optimized and two different training algorithms of the ANN are also evaluated. Best results are obtained when the potentiometric sensors are activated and the data are processed using ANN and the gradient descent adaptive algorithm. The system is used to quantify these heavy metals in synthetic samples and in dental amalgams with successful results.     

人工神经网络在纸浆卡伯值光学定量分析中的应用

卡伯值 (硬度 )是纸浆的重要质量指标 ,是制浆过程控制的关键参数 .目前的测量方法包括化学分析法和光学分析法两大类型 ,国内大多数的制浆造纸厂采用离线的传统化学分析法来测定纸浆的卡伯值 ,需要比较长的时间 .而光学分析法因具有实时性好、精度和可靠性高等优点 ,已逐步用于卡伯值的在线测量和控制 .研究 [1] 发现 ,在 460～ 580 nm的可见光谱范围内 ,蒸煮液吸光度的变化可以表征纸浆中木素含量的变化 .本文将可见分光光谱技术应用于蒸煮液中木素含量的在线测量 ,根据蒸煮液在所选波段的吸光度来预测纸浆的卡伯值 ,建立纸浆卡伯值与蒸煮…     

Comparative study of artificial neural network and multivariate methods to classify Spanish DO rose wines

Classical multivariate analysis techniques such as factor analysis and stepwise linear discriminant analysis and artificial neural networks method (ANN) have been applied to the classification of Spanish denomination of origin (DO) rose wines according to their geographical origin. Seventy commercial rose wines from four different Spanish DO (Ribera del Duero, Rioja, Valdepeñas and La Mancha) and two successive vintages were studied. Nineteen different variables were measured in these wines. The stepwise linear discriminant analyses (SLDA) model selected 10 variables obtaining a global percentage of correct classification of 98.8% and of global prediction of 97.3%. The ANN model selected seven variables, five of which were also selected by the SLDA model, and it gave a 100% of correct classification for training and prediction. So, both models can be considered satisfactory and acceptable, being the selected variables useful to classify and differentiate these wines by their origin. Furthermore, the casual index analysis gave information that can be easily explained from an enological point of view.     

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.     

烃类混合气体的神经网络模型检测

八十年代末科学家模仿生物鼻研制一种传感器阵列与计算机模式识别的气体检测系统．传感器阵列相当于生物鼻的嗅觉细胞，计算机模式识别系统相当于嗅泡和大脑「‘］．传感器阵列对气体的响应是一个多维空间的响应模式，这种响应模式经过一定的数学处理后可以实现气体的种类识别或浓度检测［’－‘j．传感器的响应和混合气体浓度之间呈非线性关系，这一特性给定量检测多组分气体混合物造成很大的限制．应用人工神经元网络技术（ANN）可以克服这一缺陷，并使检测气体的选择性大大提高．本工作运用ANN中的反向传播（BP）算法识别由16个不同…     

Artificial neural networks in analysis of indinavir and its degradation products retention

Artificial neural networks (ANN) are biologically inspired computer programs designed to simulate the way in which the human brain processes the information. In the past few years, coupling of experimental design (ED) and ANN became useful tool in the method optimization. This paper presents the application of ED-ANN in analysis of chromatographic behavior of indinavir and its degradation products. According to preliminary study, full factorial design 2⁴ was chosen to set input variables for network training. Experimental data (inputs) and results for retention factors from experiments (outputs) were used to train the ANN with aim to define correlation among variables. For networks training multi-layer perceptron (MLP) with back propagation (BP) algorithm was used. Network with the lowest root mean square (RMS) had 4-8-3 topology. Predicted data were in good agreement with experimental data (correlation was higher than 0.9713 for training set). Regression statistics confirmed good ability of trained network to predict compounds retention.     

模式识别用于压电晶体传感器阵列定量检测有机蒸汽混合物

本文应用9个压电晶体组成传感器阵列, 每片晶体上分别涂有不同种类的冠醚衍生物, 用它来定量检测二元及三元有机蒸汽混合物, 在数据处理中比较了两种模式识别方法---偏最小二乘法(PLS)和人工神经网络法(ANN), 实验证明, ANN法在预测准确度上明显优于PLS法, 本文还讨论了解决神经网络训练过拟合现象的方法。     

The prediction of total anthocyanin concentration in red-grape homogenates using visible-near-infrared spectroscopy and artificial neural networks

This study compares the performance of partial least squares (PLS) regression analysis and artificial neural networks (ANN) for the prediction of total anthocyanin concentration in red-grape homogenates from their visible-near-infrared (Vis-NIR) spectra. The PLS prediction of anthocyanin concentrations for new-season samples from Vis-NIR spectra was characterised by regression non-linearity and prediction bias. In practice, this usually requires the inclusion of some samples from the new vintage to improve the prediction. The use of WinISI LOCAL partly alleviated these problems but still resulted in increased error at high and low extremes of the anthocyanin concentration range. Artificial neural networks regression was investigated as an alternative method to PLS, due to the inherent advantages of ANN for modelling non-linear systems. The method proposed here combines the advantages of the data reduction capabilities of PLS regression with the non-linear modelling capabilities of ANN. With the use of PLS scores as inputs for ANN regression, the model was shown to be quicker and easier to train than using raw full-spectrum data. The ANN calibration for prediction of new vintage grape data, using PLS scores as inputs, was more linear and accurate than global and LOCAL PLS models and appears to reduce the need for refreshing the calibration with new-season samples. ANN with PLS scores required fewer inputs and was less prone to overfitting than using PCA scores. A variation of the ANN method, using carefully selected spectral frequencies as inputs, resulted in prediction accuracy comparable to those using PLS scores but, as for PCA inputs, was also prone to overfitting with redundant wavelengths.     

Artificial neural network combined with principal component analysis for resolution of complex pharmaceutical formulations

A chemometric approach based on the combined use of the principal component analysis (PCA) and artificial neural network (ANN) was developed for the multicomponent determination of caffeine (CAF), mepyramine (MEP), phenylpropanolamine (PPA) and pheniramine (PNA) in their pharmaceutical preparations without any chemical separation. The predictive ability of the ANN method was compared with the classical linear regression method Partial Least Squares 2 (PLS2). The UV spectral data between 220 and 300 nm of a training set of sixteen quaternary mixtures were processed by PCA to reduce the dimensions of input data and eliminate the noise coming from instrumentation. Several spectral ranges and different numbers of principal components (PCs) were tested to find the PCA-ANN and PLS2 models reaching the best determination results. A two layer ANN, using the first four PCs, was used with log-sigmoid transfer function in first hidden layer and linear transfer function in output layer. Standard error of prediction (SEP) was adopted to assess the predictive accuracy of the models when subjected to external validation. PCA-ANN showed better prediction ability in the determination of PPA and PNA in synthetic samples with added excipients and pharmaceutical formulations. Since both components are characterized by low absorptivity, the better performance of PCA-ANN was ascribed to the ability in considering all non-linear information from noise or interfering excipients.     

Discrimination of Beers by Cyclic Voltammetry Using a Single Carbon Screen-printed Electrode

A fast, simple and costless methodology without sample pre-treatment is proposed for the discrimination of beers. It is based on cyclic voltammetry (CV) using commercial carbon screen-printed electrodes (SPCE) and includes a correction of the signals measured with different SPCE units. Data are submitted to partial least squares discriminant analysis (PLS−DA) and support vector machine discriminant analysis (SVM−DA), which allow a reasonable classification of the beers. Also, CV data from beers can be used to predict their alcoholic degree by partial least squares (PLS) and artificial neural networks (ANN). In general, non-linear methods provide better results than linear ones.     

Usage of Artificial Neural Network (Back Propagation) in Optimising Salicylic Acid Determination with Ferric(III) Nitrate

Abstract

The use of artificial neural networks (ANN) in optimizing salicylic acid (SA) determination is presented in this paper. A simple and rapid spectrophotometric method for salicylic acid (SA) determination was carried out based on the complexation of salicylic acid–ferric(III) nitrate, SAFe(III). The SA forms a stable purple complex with ferric(III) nitrate at pH 2.45. The useful dynamic linear range is 0.01–0.35 g/L. It has a maximum absorption at 524 nm and the stability is more than 50 hours. The results were used for artificial neural networks (ANNs) training to optimize data. For training and validation purposes, a back‐propagation (BP) artificial neural network (ANN) was used. The results showed that ANN technique was very effective and useful in broadening the limited dynamic linear response range mentioned to an extensive calibration response (0.01–0.70 g/L). It was found that a network with 22 hidden neurons was highly accurate in predicting the determination of SA. This network scores a summation of squared error (SSE) skill and low average predicted error of 0.0078 and 0.00427 g/L, respectively.     

VOCs分子的半导体型传感器识别检测研究进展

具有体积小、功耗低、灵敏度高、硅工艺兼容性好等优点的金属氧化物半导体(MOS)气体传感器现已广泛地应用于军事、科研和国民经济的各个领域。然而MOS传感器的低选择性阻碍了其在物联网(IoT)时代的应用前景。为此,本文综述了解决MOS传感器选择性的研究进展,主要介绍了敏感材料性能提升、电子鼻和热调制三种改善MOS传感器选择性的技术方法,阐述了三种方法目前所存在的问题及其未来的发展趋势。同时,本文还对比介绍了机器嗅觉领域主流的主成分分析(PCA)、线性判别分析(LDA)和神经网络(NN)模式识别/机器学习算法。最后,本综述展望了具有数据降维、特征提取和鲁棒性识别分类性能的卷积神经网络(CNN)深度学习算法在气体识别领域的应用前景。基于敏感材料性能的提升、多种调制手段与阵列技术的结合以及人工智能(AI)领域深度学习算法的最新进展,将会极大地增强非选择性MOS传感器的挥发性有机化合物(VOCs)分子识别能力。     

Selection of quasi-optimal inputs in chemometrics modeling by artificial neural network analysis

Instrumentation spectra used for chemometrics analysis are often too unwieldy to model, as many of the inputs do not contain important information. Several mathematical methods are used for reducing the number of inputs to the significant ones only. Artificial neural networks (ANN) modeling suffers from difficulties in training models with a large number of inputs. However, using a non-random initial connection weight algorithm and local minima avoidance and escape techniques can overcome these difficulties. Once the ANN model is trained, the analysis of its connection weights can easily identify the more relevant inputs. Repeating the process of training the ANN model with the reduced input set and the selection of the more relevant inputs can proceed until a quasi-optimal, small, set of inputs is identified. Two examples are presented—finding the minimal set of wavelengths in benchmark diesel fuel NIR spectra, and in spectra generated in a recent work, modeling of “artificial nose” sensor array. In the last example, 1260 inputs were reduced to optimal sets of <10 inputs. Causal index calculation can analyze the influence of each of selected wavelengths on the predicted property. Some of the resulting minimal sets are not unique, depending on the ANN architecture used in the training. The accuracy of the resulting ANN models is usually better, and more robust, than the original large ANN model.     

Artificial neural network data analysis for classification of soils based on their radionuclide content

The artificial neural network (ANN) data analysis method was used to recognize and classify soils of an unknown geographic origin. A total of 103 soil samples were differentiated into classes according to the regions in Serbia and Montenegro from which they were collected. Their radionuclide (²²⁶Ra, ²³⁸U, ²³⁵U, ⁴⁰K, ¹³⁴Cs, ¹³⁷Cs, ²³²Th, and ⁷Be) activities detected by gamma-ray spectrometry were then used as inputs to ANN. Five different training algorithms with different numbers of samples in training sets were tested and compared in order to find the one with the minimum root mean square error (RMSE). The best predictive power for the classification of soils from the fifteen regions was achieved using a network with seven hidden layer nodes and 2500 training epochs using the online back-propagation randomized training algorithm. With the optimized ANN, most soil samples not included in the ANN training data set were correctly classified at an average rate of 92%. The text was submitted by the authors in English.     

神经网络方法用于分辨3种化学物质

神经网络方法用于分辨３种化学物质冯伟,胡上序（浙江大学化工系,杭州,３１００２７）关键词模式分类,神经网络,模拟退火,遗传算法,传感器阵列传感器阵列技术是利用传感器阵列所提供的交叉敏和模式识别及微机处理技术,来提高传感器的选择性和传感器的测量精度［１...