Resolution of overlapping signals in spectrometry using a wavelet packet transform and an Elman recurrent neural network

A novel method named a wavelet packet transform based Elman recurrent neural network (WPTERNN) was proposed for the simultaneous UV–visible spectrometric determination of Cu(II), Cd(II) and Zn(II). This method combined wavelet packet denoising with an Elman recurrent neural network. A wavelet packet transform was applied to perform data compression, to extract relevant information, and to eliminate noise and collinearity. An Elman recurrent network was applied for nonlinear multivariate calibration. In this case, using trials, the kind of wavelet function, the decomposition level, and the number of hidden nodes for the WPTERNN method were selected as Daubechies 14, 3, and 8, respectively. A program (PWPTERNN) was designed that could perform the simultaneous determination of Cu(II), Cd(II) and Zn(II). The relative standard errors of prediction (RSEP) obtained for all components using WPTERNN, a Elman recurrent neural network (ERNN), partial least squares (PLS), principal component regression (PCR), Fourier transform based PCR (FTPCR), and multivariate linear regression (MLR) were compared. Experimental results demonstrated that the WPTERRN method was successful even where there was severe overlap of spectra. The results obtained from an additional test case also demonstrated that the WPTERNN method performed very well. Figure The part of WP coefficients obtained by wavelet packet transforms     

基于小波和小波包变换的苯丙酮尿症傅里叶变换衰减全反射红外光谱筛查模型的研究

应用小波和小波包变换对傅里叶变换衰减全反射红外光谱(FTIR/ATR)进行去噪处理,以提高苯丙酮尿症(PKU)筛查模型的性能。首先优化小波和小波包变换的参数,然后分别对原始光谱(OS)、9点平滑光谱(9S)和一阶微分9点平滑光谱(1D9S)进行去噪处理,以均方根误差(RMSE)、平均相对误差(MRE)、预测准确率(Acc)等为指标,考察小波和小波包变换对模型性能的影响。结果与变换前相比,模型性能均有所提高,其中小波变换以1D9S+sym12处理结果为最优,而小波包变换以1D9S+sym1为最优;Acc全部提高为100%。     

小波神经网络用于钼和钨的同时测定

介绍了小波神经网络的结构和算法，并首次用于混合试样中的钼和钨含量测定，在小波神经网络中，采用Ｍｏｒｌｅｔ母小波和一维搜索变步长共轭梯度优化方法，结果表明，小波神经网络优于ＢＰ网络，其预测含量的回收率在９６．０％～１０４．６％之间。     

小波包压缩-RBF网络同时测定润滑油中铁、铜、锌

将小波包压缩-RBF网络方法用于润滑油中铁铜锌三组分分光光度同时测定该方法采用小波包函数对光谱数据进行压缩处理,用较大的小波色系数构成新的校正集和预测集代替原始的校正集和预测集,然后用RBF网络进行数据解析。研究表明,用bior2．4小波包处理原始测定数据,最佳小波基用logenerge熵标准,选择适当阈值将变量数由46个压缩成25个（压缩比为0．54）,整体预测效果最好。将此方法用于合成样预测．预测结果与实际浓度的相对误差绝对值在2．50％～9．40％之间;用于实际润滑油样品中铁、铜和锌的同时测定,解析值与原子吸收法（AAS）的测定值的相对误差绝对值作3．55％～7．86％之间。应用结果令人满意。     

小波包变换潜变量回归分辨重叠的紫外光谱

采用小波包变换潜变量回归(WPLVR)方法,同时测定联苯、苯酚和邻苯二酚。该法结合小波包变换和潜变量回归改进除噪质量。通过最佳化,选择了小波函数及小波包分解水平(L)。编制了两个程序PWPLVR和PFTLVR进行WPLVR和付立叶变换潜变量回归(FTLVR)法计算。试验结果表明WPLVR法是成功的且优于FTLVR法。     

用分段训练的Elman递归神经网络滤除色谱数据中的噪声

分析了色谱信号噪声的分形特征,采用分段训练的Elman递归神经网络除噪.实验结果表明,此方法在滤除噪声时具有一般神经网络和中值滤波器所没有的优点,精度明显提高.     

Simultaneous Determination of Vitamin B Complex Using Wavelet Neural Network

A simultaneous determination of four components of B‐group vitamin, using a novel wavelet‐based neural network (WNN), combined with correlation coefficient and standard deviation approach for wavelength selection, was reported in this work. Eleven representative wavelength points were selected from each original UV spectrum, based on correlation coefficients and standard deviations of the observed data. A family of wavelet basic functions built from Morlet wavelet was adopted to improve the transfer quality of output data and solve the problems of training difficultly involved in neural networks. The predicted results, with fitting correlation coefficients (R = 0.9998–0.9999) and rooted mean squares errors (RMS = 0.0578–0.1478), are satisfactory.     

人工神经网络催化动力学光度法同时测定硒和铁

在酸性介质中,硒(Ⅳ)和铁(Ⅲ)同时催化H_2O_2氧化二甲基黄褪色反应.研究发现:两者对H_2O_2氧化二甲基黄褪色反应的催化作用不具有加和性.根据这一现象,结合人工神经网络,建立了一种新的不经分离可同时测定硒(Ⅳ)和铁(Ⅲ)混合物中硒(Ⅳ)和铁(Ⅲ)各自含量的分析方法.方法应用于合成样品及烟草样品中硒(Ⅳ)和铁(Ⅲ)的测定,结果满意.     

小波变换用于植物激素重叠色谱峰中组分的定量分析

根据小波变换将信号按频率大小分解的性质，作者提出了一种用小波变换直接提取重叠色谱峰中单一组分信息的方法，并将其应用于植物激素混和体系重叠峰的解析与定量分析。结果表明：单组分的色谱信号经小波变换分解后仍能保持原有的线性关系，用小波变换从重叠信号中提取的单一组分信息也具有良好的线性关系。     

Rapid Determination of Enantiomeric Excess of Tert-Butoxycarbonyl (BOC-Protected) Amino Acids Based on Infrared Spectra Technique with Optimal Wavelet Packet Transform Decomposition Frequency Band

A method was employed to determine enantiomeric excess (ee) value of chiral tert-butoxycarbonyl (Boc-protected) amino acids in a rapid way by using an infrared spectroscopy technique combined with wavelet packet transform (WPT) and least squares support vector machines (LS-SVM). Infrared spectral data were decomposed by using WPT algorithm. Simulated annealing (SA) algorithm was then used to search the optimal decomposed frequency band that had the greatest contribution to the quantitative analysis of ee values. As a result, the band (7, 34) with 34 variables and the band (5, 1) with 116 variables were determined as the optimal ones for the determination of Boc-protected proline and alanine, respectively. The selected variables in the optimal band were used as the inputs of LS-SVM models. The spectral variables selected by the WPT-SA method had lower predicting errors than full range spectra and the spectral variables selected by some traditional variable selection methods. Reasonable good results with root mean-square error of prediction (RMSEP) of 7.51 and 3.80 were obtained for the determination of ee values of two Boc-protected amino acids, showing that it is possible to rapidly determine ee values of amino acids by using IR spectroscopy rapidly.     

线性神经网络应用于维生素B族4组分同时测定

报道了一种基于卡尔曼滤波的线性神经网络用于B族维生素混合组分的定量同时测定,用相关系数和标准偏差从原始紫外光谱数据中挑选11个波长点供线性神经网络处理,以极少的迭代次数达到较高的精度,预测结果较好,相关系数大于0.99983,标准偏差0.11301.     

基于小波神经网络的新型算法用于化学信号处理

基于紧支集正交小波神经网络的构造思想,用具有紧支集的Ｂ－样条函数的伸缩和平稳替代小波函数,提出了一种新型算法,并将其应用于化学信号的处理,实现了信号的压缩和滤噪,适应小波神经网络相比,学习速度得到了大幅度的提高。     

离散小波特征提取及人工神经网络分类法的傅里叶变换红外光谱法识别鳞毛蕨科3种植物

利用水平衰减全反射-傅里叶变换红外光谱法测定了3种药用鳞毛蕨科植物贯众、阔鳞鳞毛蕨和变异鳞毛蕨根部的FT-IR。运用基于离散小波多分辨率分析法对FT-IR吸收较为相似的3种药用蕨类植物根的FT-IR进行特征提取。选择第4、5分解层数的特征向量,进行人工神经网络(Artificial neural network,ANN)训练;再用训练出来的网络对不同产地的3种药用蕨类植物根所得FT-IR小波提取的特征向量进行分类。通过对240个不同样本的预测,说明能够采用基于FT-IR-离散小波特征提取及人工神经网络分类法对同科3种药用蕨类植物根的FT-IR进行识别。     

小波神经网络在紫外光谱识别中的应用研究

介绍用于信号识别的小波神经网络的结构和算法，并将其用于酪氨酸，二羟基苯丙氨酸和色氨酸的紫外光谱识别。在波小神经网络中，采用Ｍｏｒｌｅｔ母波波和一维搜索变步长共轭梯度优化方法。结果表明，小波神经网络对于光谱间的细微结构差别具有很好的识别能力。     

用小波变换确定电位滴定终点

将小波变换在信号和图象处理过程中边缘检取的思想方法引入到化学中，以确定电位滴定的终点，能直接对测量的数据进行处理，不会引入人为的误差因素。计算精度可达实验精度的下限，且处理结果的判断非常简便，用本方法对实验数据进行处理，获得满意的结果。     

小波神经网络用于色谱数据的滤噪与压缩

介绍了小波神经网络（Ｗａｖｅｌｅｔ Ｎｅｕｒａｌ Ｎｅｔｗｏｒｋ）的结构和算法，并将其应用于色谱数据的滤噪与压缩。在小波神经网络中，采用了Ｂｕｂｂｌｅ母小波和一维搜索变步长共轭梯度优化方法。结果表明，小波神经网络将原始数据压缩至１：２０时，仍能得到很好的重建谱图。重建的色谱图不仅能保留原始的谱图特征而且有光滑滤噪的效果。对于色谱数据的存储与平衡滤噪具有重要意义。     

动力学电位法测定硼

本文根据硼酸，苯乙醇酸，孔雀绿形成三元离子缔合型配合物的反应，用孔雀绿离子选择性电极监测反应速率，在最佳实验条件下，用固定时间法，电位变化与硼量在０．２－５．０μｇ／ｍＬ范围呈线性关系，用於测定标准钢样和玻璃标样中的硼，结果与标准值十分接近。     

小波变换与神经网络结合用于示波计时电位测定

小波变换与神经网络的结合是化学计量学中一个十分活跃的研究领域。作者介绍了小波变换与神经网络的两种结合方式——辅助式结合和嵌套式结合,并将二者用于单组分示波计时电位信号定量计算的比较。结果表明：辅助式结合训练次数少,预测误差小;小波神经网络计算的自动化程度高,计算结果明显优于传统的人工神经网络。     

Application of Artificial Neural Network in Oscillographic Chronopotentiometry

研究了误差反传神经网络和小波变换在示波计时电位测定中应用的可能性。对Pb