Study on the Multicomponent Quantitative Analysis Using Near Infrared Spectroscopy Based on Building Elman Model
LIU Bo-ping1,2,QIN Hua-jun3,LUO Xiang2,CAO Shu-wen3,WANG Jun-de1*
1. College of Chemical Engineering, Nanjing University of Science and Technology, Nanjing 200014,China 2. Analytical and Testing Center of Jiangxi Province, Nanchang 330029, China 3. Key Laboratory of Food Science of MOE, Nanchang University, Nanchang 330047, China
摘要: 研究了Elman神经网络(反馈神经网络,Recurrent Network)在近红外光谱定量分析中的应用。以饲料样品为实验材料,采用Elman网络建立了饲料中苯丙氨酸(Phe)、赖氨酸(Lys)、酪氨酸(Tyr)和胱氨酸(Cys)四种氨基酸含量的近红外光谱定量分析模型。用偏最小二乘法(partial least squares ,PLS)将原始数据压缩为主成分,取前3个主成分的12个吸收峰值输入Elman网络,网络中间层神经元个数为47。Elman网络模型对样品4个氨基酸含量的预测决定系数(r2)分别为0.960,0.981,0.979,0.952。表明所建Elman网络预测模型通过近红外光谱能够较准确预测饲料中苯丙氨酸、赖氨酸、酪氨酸和胱氨酸四种氨基酸的含量,为通过近红外光谱技术进行多组分定量分析提供了新思路。
关键词:近红外光谱;Elman网络;偏最小二乘法;多组分定量分析
Abstract:The present paper introduces an application of near infrared spectroscopy(NIRS) multi-component quantitative analysis by building a kind of recurrent network(Elman)model. Elman prediction model for phenylalanine(Phe), lysine(Lys), tyrosine(Tyr) and cystine(Cys) in 45 feedstuff samples was established with good veracity. Twelve peak value data from 3 principal components straight forward compressed from the original data by PLS were taken as inputs of Elman, while 4 predictive targets as outputs. Forty seven nerve cells were taken as hidden nodes with the lowest error compared with taking 43 and 45 nerve cells. Its training iteration times was supposed to be 1 000. Predictive correlation coefficients by the model are 0.960, 0.981, 0.979 and 0.952. The results show that Elman using in NIRS is a rapid, effective means for measuring Phe, Lys, Tyr and Cys in feedstuff powder, and can also be used in quantitative analysis of other samples.
刘波平1,2,秦华俊3,罗香2,曹树稳3,王俊德1* . 基于Elman网络的近红外光谱技术多组分定量分析研究[J]. 光谱学与光谱分析, 2007, 27(12): 2456-2459.
LIU Bo-ping1,2,QIN Hua-jun3,LUO Xiang2,CAO Shu-wen3,WANG Jun-de1* . Study on the Multicomponent Quantitative Analysis Using Near Infrared Spectroscopy Based on Building Elman Model . SPECTROSCOPY AND SPECTRAL ANALYSIS, 2007, 27(12): 2456-2459.
[1] CHENG Zhong, CHEN De-zhao(成 忠, 陈德钊). Spectroscopy and Spectral Analysis(光谱学与光谱分析),2006, 26(6): 1046. [2] HE Yong, LI Xiao-li, SHAO Yong-ni(何 勇, 李晓丽, 邵咏妮). Spectroscopy and Spectral Analysis(光谱学与光谱分析),2006, 26(5): 850. [3] ZHANG Lin, ZHANG Li-ming, LI Yan, et al(张 琳, 张黎明, 李 燕, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析),2006, 26(4): 620. [4] Yande Liu, Yibin Ying. Postharvest Biology and Technology, 2005, 37: 65. [5] Meher L C, Sagar D V, Naik S N. Renewable and Sustainable Energy Reviews, 2006, 10(3): 248. [6] McClure W F, Crowell B, Stanfield D L, et al. Journal of Near Infrared Spectrosc. 2002, 10(3): 177. [7] Inmaculada G M,Noelia A G, José Miguel G C. Talanta, 2006, 69(3): 706. [8] LI Yan, SUN Xiu-yun, WANG Jun-de(李 燕, 孙秀云, 王俊德). Spectroscopy and Spectral Analysis(光谱学与光谱分析),2000, 20(6): 773. [9] YANG Jian-gang(杨建刚). Applied Tutorial of Artificial Neural Networks(人工神经网络实用教程). Hangzhou: Zhejiang University Press(杭州: 浙江大学出版社), 2001. [10] Rasit Kker. Information Sciences, 2006, 176: 1781. [11] JIA Chun-yu, LIU Hong-ming. Journal of Iron and Steel Research, International, 2006, 13(1): 31. [12] Lin Chi-ming, Huang Jih-jeng, Mitsuo Gen, et al. Applied Mathematics and Computation, 2006, 175(2): 1139. [13] Rasit Kker. Engineering Applications of Artificial Intelligence, 2005, 18(6): 685. [14] CAO Yue, ZHANG Cheng-liang(曹 跃,张成良). Journal of Huhan University of Technology(武汉理工大学学报),2006,28(2):49. [15] ZHANG Ya-ou, L Tian-sheng, DU Jian-fu(张亚欧,吕恬生,杜建福). Journal of System Simulation(系统仿真学报),2006,18(2):309. [16] GAO Ling, SHI Jun-xian, REN Shou-xin(高 玲, 石俊仙, 任守信). Spectroscopy and Spectral Analysis(光谱学与光谱分析),2006, 26(1): 117. [17] LI Chun-yan, ZHU Dong-hua(林春燕, 朱东华). Computer Application(计算机应用),2006, 26(2): 476. [18] CHEN Feng, HU Shang-xu, YU Meng-huai(陈 锋, 胡上序, 俞蒙槐). Chemical Journal of Chinese Universities(高等学校化学学报), 2000, 21(2): 193.