摘要
利用遗传算法(GA)提取茶叶的近红外吸收特征波长的方法,研究建立了绿茶水分和氨基酸的近红外分析模型,并对波长选择前后两种成分的模型进行了比较分析。结果表明,经遗传算法波长选择后,简化了分析模型,同时模型的稳健性增强。氨基酸预测集的均方根误差(SEP)减小82.1%,水分预测集的均方根误差减小(SEP)76.6%,它们在波长选择前后对应的分析波长数之比分别为995∶7和1990∶33。
Two near IR models for the analysis of moisture and amino acid in green tea were established.The selection of the characteristic near IR absorption wavelengths for the analysis was based on the genetic algorithms.The models for the moisture and amino acid fore-and-aft the wavelength selection were compared.The results showed that the genetic algorithms was very effective in selecting the characteristic wavelength.The root mean square error of the predication for moisture and amino acid models were reduced by 82.1% and 76.6%,respectively and the ratio of wavenumbers fore-and-aft selection were 995∶7 and 1990∶33,respectively.
出处
《分析测试学报》
CAS
CSCD
北大核心
2007年第5期679-681,685,共4页
Journal of Instrumental Analysis
基金
浙江省科技计划项目资助(2006C2144)
浙江省教育厅科研项目资助(20050381)
关键词
绿茶
近红外光谱
遗传算法
波长选择
Green tea
Near infrared spectroscopy（NIR）
Genetic algorithms（GA）
Wavelength selection
