Detection method for analyzing white spirit by combining ion mobility spectrometry fingerprint spectrum with neural network

Abstract

The invention belongs to the technical field of white spirit detection, and particularly relates to a detection method for analyzing white spirit by combining an ion mobility spectrometry fingerprint spectrum with a neural network, which comprises the following specific steps: S1, establishing an artificial neural network; s2, liquid sampling is carried out and sent into an ion mobility spectrometer to be detected, so that the detected white spirit is analyzed, after the liquid sample is detected through the ion mobility spectrometer, the fingerprint spectrum of the detected sample is obtained through detection, and when the fingerprint spectrum of the detected sample is obtained, the fingerprint spectrum of the detected sample is detected through the ion mobility spectrometer. Establishing a baijiu ion mobility spectrometry fingerprint spectrum database; s3, after the Baijiu ion mobility spectrometry fingerprint database is established by the ion mobility spectrometer, direct sample injection of the Baijiu sample is realized, sample pretreatment is not needed, when the Baijiu is tested by the ion mobility spectrometer, the sampling amount is small, the time consumption is short, an analysis result can be obtained in 2 minutes, the result is reliable, the operation is simple and convenient, the instrument performance is stable, and the automation degree is high.

Description

technical field [0001] The invention relates to the technical field of liquor detection, in particular to a detection method for liquor by combining ion mobility spectrum fingerprints with neural network analysis. Background technique [0002] Liquor, distilled wine made from grains and grains, using Daqu, Xiaoqu or bran koji and distiller's yeast as saccharification starter, cooking, saccharifying, fermenting, and distilling, also known as shochu, Laobaigan, Shaodaozi, etc., liquor It is colorless or slightly yellow and transparent, with a pure and pure smell, sweet and refreshing in the mouth, and has a high alcohol content. After storage and aging, it has a compound fragrance mainly composed of esters. It uses koji and distiller's yeast as saccharification starter. Starch or sugar as raw materials, all kinds of wine brewed through cooking, saccharification, fermentation, distillation, aging and blending, but strictly speaking, blended wine made from edible alcohol and edi...

AI Technical Summary

Problems solved by technology

[0003] The existing liquor quality detection schemes are generally divided into two types, one is judged by professional wine tasters, and the other is analyzed by gas chromatography coupled with mass spectrometry, but the training period of professional wine tasters is very long and takes a long time Accumulation, subjectivity exists at the same time, and the main problem of gas chromatography-mass spectrometry detection is that too many substances are obtained. Generally, there are more than 100 flavor substances in liquor, and there are too many variables, so the data cannot be analyzed effectively. For this reason, We propose a detection method of ion mobility spectrometry fingerprint combined with neural network analysis of liquor