Research On The Origin Traceability Of Silkworm Cocoons Based On Stable Isotope Fingerprints

As a great contribution of the working people in ancient my country,silk has been endowed with special artistic and cultural values,and the origin of silk has also attracted the attention and attention of the world.It is urgent to establish a method system to trace the origin of silkworm cocoons and related silk fabrics.Since modern research,stable isotope technology is one of the more effective methods for traceability of origin.In the field of origin research of silkworm cocoons and silk fabrics based on stable isotope fingerprints,the most important research goal is to distinguish and determine whether there are differences in the stable isotope ratios in cocoons from different sources or places of origin.This thesis takes the cocoons from four regions of Zhejiang,Shandong,Hubei,and Sichuan as the research object.Based on the stable isotope technology,the carbon,nitrogen,hydrogen,and oxygen isotope ratios in the four regions are measured and analyzed,combined with a variety of chemometrics.Methods Identify and analyze the differences of cocoons from different producing areas,and establish the relevant index system and technical system for tracing the origin of silkworm cocoons.The main findings are as follows:A total of 64 samples of silkworm cocoons were collected from four regions.The results of stable isotope determination after pretreatment showed that theδ¹³C,δ¹⁸O,δ¹⁵N,andδ²H change ranges of the cocoons in the four provinces were different.The average range ofδ¹³C is-27.55‰to-24.24‰,the average range ofδ¹⁵N is 2.45‰to7.98‰,the average range ofδ²H is-87.98‰to-54.03‰,and the average range ofδ¹⁸O is 16.81‰to 23.90‰.The stable isotope determination values of silkworm cocoons in different regions show different distribution ranges,which preliminarily shows certain regional differences.In the significance analysis based on single factor index,Shandong’sδ¹³C value is-24.91‰,which is significantly lower than the other three provinces,Hubei’sδ¹³C value is the highest at-26.57‰;Shandong’sδ¹⁵N is 6.17‰,which is significantly higher than the other three provinces.In the region,theδ²H of Sichuan is significantly higher than that of Hubei and Shandong,but there is no significant difference compared with Zhejiang.Theδ¹⁸O of Hubei,Sichuan,and Zhejiang is significantly different,and the value ofδ¹⁸O of Sichuan is the highest at23.23‰.Collect samples of silkworm cocoons from the same place of production in Hubei or Zhejiang at different times.After pre-treatment,the stable isotope was determined.The results show that the sampling time has an effect on the stable isotope content of the cocoons in the comparison of the same variety in the same production area in different sampling years.,But at the same time affected by the species of silkworm cocoons,it can be seen from the comparison of Hubei and Zhejiang that the difference in isotope may be related to the local climate in different sampling years,that is,the temperature of the mulberry leaf growth environment changes in different years,and the atmospheric precipitation is also different.This may be the possible cause of the discrepancy.The cluster analysis results show that the clustering results of a single index sample are consistent with the actual origin of the sample,but at the same time,there are cases where the stable isotope parts of different regions are mixed together in the cluster analysis.Cluster analysis has certain feasibility for the application of the visible stable isotope index of silkworm cocoon samples to the identification of silkworm cocoon producing area.But it needs to be analyzed and judged in combination with other chemometric analysis methods.In the construction of the 3-layer BP neural network,the stable isotope detection result is extracted as the input vector of the first layer.After training,verification and testing,the results of the training group and the test group are that the correct rate of the training set is 76.63%,the recall rate is 73.08%,and the F1-score score is 74.81%;the correct rate of the test set is 91.67%,and the recall rate is 91.67%.33.33%,F1-score score is 48.89%.The values of the training set are all greater than 70%,but less than 80%,indicating that the network results have not reached the best condition.Due to the lack of regions in the sample size of the test set,the overall index value is falsely high.Finally,FISHER discriminant analysis was used to analyze the differences ofδ¹³C,δ¹⁵N,δ¹⁸O,andδ²H of cocoons from different origins,and most of them reached a significant level;the coupling effect of multiple indicators was better than that of single isotope indicators,and the results of stepwise discriminant analysis showed that with the increase of coupling indicators,The initial discrimination rate and cross discrimination rate of the place of origin are gradually increased;finally,a traceability model is established using 4 indicators ofδ¹³C,δ¹⁸O,δ¹⁵N,andδ²H.The initial discrimination rate of the model is 84.5%,and the cross discrimination rate is 77.6%.The discriminant function of the traceability model is:Y（Hubei）=-1746.21-115.787δ¹³C+13.333δ¹⁵N+0.168δ²H+18.048δ¹⁸O,Y（Shandong）=-1589.42-108.987δ¹³C+13.689δ¹⁵N-0.027δ²H+18.155δ¹⁸O,Y（Sichuan）=-1762.933-115.363δ¹³C+12.51δ¹⁵N+0.311δ²H+19.954δ¹⁸O,Y（Zhejiang）=-1748.463-115.439δ¹³C+13.224δ¹⁵N+0.217δ²H+18.794δ¹⁸O.The results of this study provide a basis for the identification of the origin of silkworm cocoons..