Study of grass carp (Ctenopharyngodon idellus) quality predictive model based on electronic nose

An electronic nose based quality predictive model of grass carp (Ctenopharyngodon idellus) stored at 277 K temperature was proposed in this paper. The changes of sensor array response to samples were caused by the new-generated gas species released by microbial propagations. Principal component analysis method discriminated fresh grass carp samples from medium samples and aged samples. Stochastic resonance signal-to-noise ratio maximums distinguished fresh, medium, and aged grass carp samples successfully. The quality predicting model was developed based on signal-to-noise ratio maximums non-linear fitting regression. Validating experiments demonstrated that the predicting accuracy of this model was 87.5%. This method presented some advantages including easy operation, quick response, high accuracy, good repeatability, etc. This method is promising in aquatic food products quality evaluating applications.