Study On The Detection Of Prohibited Ingredients In Cosmetics By Three-dimensional Fluorescence Spectroscopy

With the development of The Times,people’s demand for cosmetics is getting higher and higher,and its safety has attracted the attention of the majority of consumers.The use of banned ingredients in cosmetics has caused harm to consumers,such as skin diseases.In this paper,three dimensional fluorescence spectrum technology combined with stoichiometry is used to detect several banned substances in cosmetics,which is of great significance to ensure the safety of cosmetics market and the rights and interests of consumers.The main research work of this paper are as follows:(1)In order to solve Rayleigh scattering problem in fluorescence spectral data,the defect data reconstruction iterative algorithm combined with principal component analysis(MDR-PCA)was applied to deal with it.In order to solve the problem of noise interference in the fluorescence spectral data,the CEEMDAN method was applied to deal with it,which laid a foundation for the subsequent fluorescence spectral analysis.(2)In order to detect hydroquinone in cosmetics,the experimental system of hydroquinone in different substrate cosmetics was designed.Combined with the regression performance of support vector machine,three optimization algorithms including chicken optimization algorithm(CSO),cuckoo search algorithm(CS)and particle swarm optimization algorithm(PSO)were used to optimize the parameters of support vector machine.And the results show that CSO-SVM algorithm is more suitable for the quantitative determination of hydroquinone in complex cosmetics.(3)In order to detect sex hormones in cosmetics,an experimental system was designed to add mixed sex hormones to cosmetics.The fluorescence characteristics of estrone,estradiol and estriol in single component and mixed component were analyzed.In view of the serious overlapping problem of fluorescence characteristic peaks,combined with the classification performance of support vector machine,the improved chicken flock optimization algorithm(ICSO),CSO and PSO optimization support vector were used for qualitative detection of sex hormones in cosmetics.It is proved that ICSO-SVM algorithm is more suitable for qualitative detection of the mixed components of estrone,estradiol and estriol in cosmetics.(4)In order to detect the banned substance quinolone antibiotics in cosmetics,an experimental system of adding standard mixed antibiotics in cosmetics was designed.The fluorescence characteristics of enrofloxacin and ciprofloxacin in single component and mixed component were analyzed.To solve the overlapping problem of fluorescence characteristics,the second order correction algorithm of parallel factor analysis(PARAFAC)was applied to achieve the qualitative and quantitative detection of quinolone antibiotics in cosmetics.