Research On Spectral Quantitative And Qualitative Analysis Model

Study on the Spectrometry and Quantitative Model of Chemical Measurement.Taking dairy products as research objects,the quality control and safety of milk powder is used to identify melamine in milk powder in combination with medium infrared spectroscopy,using ultraviolet spectroscopy,combined with the main mass parameters of milk powder.The combined pattern of infrared spectroscopy(FTIR)method in Fourier transform was used to rapidly identify melamine in milk powder.This method overcomes the limitations,complexity and subjectivity of the spectral alignment identification.The use of non-related linear discriminating analysis(ULDA)methods successfully identified melamine doped in milk powder,identifying sensitivity and specificity of 100%.Then,the ULDA method is used to characterize the infrared spectrum of milk powder,and the relationship between retaining variables and identification accuracy is investigated.ULDA can still realize the identification of pure milk powder and doped 0.01 ‰ melamine milk powder in the number of small variables.Therefore,a pattern recognition method based on mid-infrared spectroscopy to quickly and accurately identify low-content melamine in milk powder is proposed.It provides an effective way for the identification and quality control of adulteration of milk powder.It can also be extended to the identification of the authenticity of other foods.The component content of milk powder is an important evaluation index for the quality and nutritional value of milk powder.The UV spectroscopy combined with a multivariate calibration regression model is used to simultaneously predict the content of the four main quality parameters of protein,fat,carbohydrate and energy in milk powder.After obtaining the spectrum of the milk powder solution sample,the partial least squares(PLS)method is used to establish a prediction model.In the end,a number of prediction results with good quality parameters were obtained,the root mean square error of prediction(RMSEP)was low,and the correlation coefficient between the predicted value and the reference value was high(greater than 0.98).The result is better than the quantitative result directly using the working curve method.In addition,compared with the multiple linear regression(MLR)model,the PLS model also obtained better results.Therefore,ultraviolet spectroscopy combined with partial least squares is an effective,fast and simple "green" technology that can simultaneously detect the content of the main components in milk powder.It can be applied to the quality control of milk powder,and further research is expected to be extended to the quantitative analysis of other components of milk powder and other food quality parameters.