作者: J. Lozano , J.P. Santos , J. Gutiérrez , M.C. Horrillo
DOI: 10.1016/J.SNB.2007.04.018
关键词: Electronic nose 、 Artificial intelligence 、 Wine 、 Pattern recognition (psychology) 、 Pattern recognition 、 Chromatography 、 Probabilistic logic 、 Purge and trap 、 Mathematics 、 Artificial neural network 、 Principal component analysis 、 Sampling (statistics)
摘要: A comparison among several sampling systems usually employed in an electronic nose is performed this paper order to improve the performance of instrument for wine discrimination. Three different methods have been studied: static headspace with dynamic injection (HS), purge and trap (P&T) solid-phase micro-extraction (SPME). These noses developed discriminate five Spanish wines coming from grape varieties elaboration processes. Linear techniques as principal component analysis (PCA) nonlinear ones probabilistic neural networks (PNN) used pattern recognition. Results show that best discrimination achieved P&T SPME, although highest response sensors obtained by HS method.