摘要
建立了用偏最小二乘(partial least squares,PLS)与人工神经网络(artificial neural networks,ANN)联用对饲料样品同时测定水分、灰分、蛋白质、磷含量的预测校正模型。光谱数据用二阶微分及标准归一化处理(SNV),用PLS法将原始数据压缩提取前10个主成分与2个特征峰值作为12个输入向量,采用单隐层的反向传播人工神经网络(Back-Propagation Network,BP),确定中间层的神经元个数为23,初始训练迭代次数为1000。PLS-BP模型对样品四个组分含量的预测决定系数(r2)分别为:0.9950,0.9980,0.9990和0.9670;样品平行扫描光谱预测值的标准偏差分别为:0.02774,0.04853,0.03292和0.02204。
Partial least squares （PLS） and artificial neural networks （ANN） prediction model for four components of feedstuff has been established with good veracity and recurrence. The spectra put into the model should be processed by second derivative and standard normal variate （SNV）. Ten principal components compressed from original data by PLS and two peak values were taken as the inputs of Back-Propagation Network （BP）, while four predictive targets as outputs, according to Kolmogorov theo- rem and experiment, and twenty three nerve cells were taken as hidden nodes. Its training iteration times was supposed to be 10 000. Prediction deciding coefficient of four components by the model are 0. 995 0, 0. 998 0, 0. 999 0 and 0. 967 0, while the standard deviation of an unknown sample scanned parallelly are 0. 027 74, 0. 048 53, 0. 032 92 and 0. 022 04.
出处
《光谱学与光谱分析》
SCIE
EI
CAS
CSCD
北大核心
2007年第10期2005-2009,共5页
Spectroscopy and Spectral Analysis
基金
教育部南昌大学食品科学重点实验室开放基金项目(NCU200404)
江西省星火计划项目(2005年)资助
