A Study On White Liquor Adulteration Detection Based On Portable Spectrometer

As a unique white liquor in China,white liquor is always in a leading position.With more and more white liquor brands and the ingredients gets more complex,there are now various brands of fake white liquors on the market,Including mixing liquor with water and then selling it,and even some of them mix the white liquor and low-cost but extremely harmful industrial alcohols.However,most of the current detection of white liquor adulteration in China only stays at large-scale laboratory equipment testing and external packaging traceability,and cannot be tested immediately and effectively.Therefore,this liquor detection mode needs to be improved.With the development of science and technology,near-infrared spectroscopy detection technology has become more and more perfect.This technology can quickly and non-destructively detect samples,and it is simple to process sample data.It has very important applications in food adulteration detection.However,one of the problems of the near-infrared spectroscopy detection instrument is that it is not convenient to operate.This article aims to solve the accuracy problems of the portable near-infrared spectroscopy detection and construct a perfect and convenient detection system.The specific research is as follows:(1)Research on Type Identification of White Liquor Adulteration Based on Portable Near Infrared Spectrometer: Doing the out-of-bottle detection of liquor mixed with water and industrial alcohol by portable near-infrared spectrometer DLP NIRscan Nano.Preprocessing for the noise of the sample spectrum,and classified the two adulterated types of liquor,Comparative analysis of the analysis results of the Support Vector Machine(SVM)model and the analysis results of the Extreme Gradient Boosting(XGBoost)model.After preprocessing the data,Compare the accuracy of the two classification models.Finally,the classification model that is most suitable to solve the problem of liquor adulteration is selected.The final result shows that the results obtained by XGBoost are more accurate.(2)Research on the Identification of White Liquor Adulteration Degree Based on Portable Near Infrared Spectrometer: In order to improve the systematic detection of white liquor adulteration,first,carry out accuracy research on different pre-processing combinations,the preprocessing combination of SG+SNV is selected for fitting processing.Then,back propagation(BP)neural network and partial least squares(Partial Least-Square method,PLS)are used to predict the adulteration degree of liquor.This research first further improved the degree of non-destructive testing,Secondly,on the basis of ensuring that the equipment is small and portable,the detection accuracy is ensured through algorithms.(3)Development of White Liquor Adulteration Detection APP Based on Portable Spectrometer: On the basis of the types and degrees of adulteration that have been obtained in the previous two chapters,this study built an Io T system platform for liquor adulteration detection.First set up the server,the user uses Android to establish a connection with the spectrometer via Bluetooth to control the spectrometer to scan data,then connect the Android phone to the server through the network and upload the data to the server,the analysis algorithm in the server is obtained from the analysis of the first two chapters.After the server analyzes the type and degree of adulteration,send the results back to the client through the network for display,Through this system,a complete detection platform for liquor adulteration has been completed.