Optimization of eigenvalue selection in Chinese liquors discrimination based on electronic nose

Abstract

An eigenvalue selection optimization method for Chinese liquors discrimination using electronic nose (e-nose) was proposed in this paper. Experiments on 13 Chinese liquors were conducted and e-nose responses were recorded. Eigenvalues of e-nose responses were selected and used in principal component analysis (PCA). The Chinese liquors were partially discriminated. The e-nose responses were analyzed by stochastic resonance method and systematic output signal-to-noise ratio eigenvalues were selected for PCA analysis. Experimental results demonstrated that the eigenvalue optimization improved the Chinese liquors discrimination results. This discussed method provided a promising way for food quality and safety examinations based on artificial olfactory system.