Research On Detection Of Sex Hormones In Prohibited Ingredients In Cosmetics Based On Three-dimensional Fluorescence Spectroscopy

In recent years,with the improvement of the quality of life,people have begun to use cosmetics to maintain the beauty and health of their skin.However,with the surge in demand for cosmetics and the diversification of sales channels,the issue of cosmetic quality and safety has become an inevitable problem.In order to quickly and effectively achieve the effect of advertising,some manufacturers of undesirable cosmetics may excessively add restricted substances or prohibited substances which has been rigorously restrained in "Cosmetics Safety Technical Specifications",If consumers use this kind of cosmetics for a long time,it will cause the skin to be too sensitive,and even affect fertility and cancer.Therefore,the safety of cosmetics is of great significance.Our country has clearly announced the content of restricted ingredients and specific banned ingredients,sex hormones have been listed as banned ingredients in cosmetics.This article is based on three-dimensional fluorescence spectroscopy,combined with optimized support vector machine and parallel factor algorithm in second-order correction,to achieve qualitative and quantitative analysis of mixed samples.The main research contents of this paper are as follows:1.Based on the principle of fluorescence detection,for the estrone,estradiol and estriol,the feasibility of the experiment was analyzed from the perspective of chemical structure,and it was confirmed that the three sex hormones can be analyzed by three-dimensional fluorescence spectroscopy;In the experimental design stage,exploring the setting of experimental parameters and collecting the experimental samples data to draw the three-dimensional fluorescence spectrum for laying the foundation for the next work.2.The effects of scattering and noise on spectral data were studied,and the triangle interpolation method and Savitzky-Golay(SG)were applied to the pretreatment of fluorescence spectral data to remove the scattering interference and reduce the influence of noise.3.The support vector machine(SVM)is used to analyze the fluorescence spectrumdata,and the ant colony and particle swarm optimization are applied to the parameter optimization process of SVM,and the optimized support vector machine algorithm model is designed.Combined with the three-dimensional fluorescence spectroscopy method,the classification of samples to be tested and the concentration prediction are performed.4.The PARAFAC and the optimized support vector machine algorithm are combined to design a sex hormone detection model under the background of cosmetics.Aiming at the characteristic of the PARAFAC that is particularly sensitive to the number of components,the core consistency diagnostic method is used to estimate the component values.and the matrix after trilinear decomposition is input into this training model to predict the concentration of sex hormones in the background of complex cosmetics.