A Study On Milk Powder With Prohibited Additions Based On THz Spectrometric Analysis

With the development of economic and the progression of society,People pay more and more attention to food security.But the current food security situation is not optimistic.kinds of food security problems appear continuously.Food safety is closely related to our lives.It not only affects human health,but also affects the development of economic and social stability.So,governments have invested enormous human and material resources to monitor food safety issues.At present,the detection of food quality is also mainly rely on the use of biochemical methods,even though these methods can work effectively,but generally speaking,these methods are complex testing procedures,detection cycle is long,high detection costs.The terahertz spectrum detection technology as a new type of spectral detection technology in today's society has been development rapidly.It is widely used in public safety,cosmic information detection,communications,environmental testing,food and drug safety supervision and other fields.This method needs a short cycle with high sensitivity and no damage to samples.It belongs to non-destructive testing method,so it can make up for the lack of other means of detection.In this paper,the terahertz time-domain spectroscopy technique was used to detect the pure milk powder,Benzoic acid,potassium sorbate,milk powder containing benzoic acid and potassium sorbate,The frequency spectrum is obtained by Fourier transform of the obtained time domain spectrum;The absorption spectrum and the refractive index spectrum of the samples were obtained by the optical parameter formulations.The experimental results show that these samples have different characteristics in the effective frequency range of terahertz.Through the analysis of these different characteristics,the qualitative analysis of different samples was achieved.Based on the analysis of the optical constants of the data obtained by terahertz detection on different samples,A neural network model for the determination of benzoic acid and potassium sorbate in milk powder was established by using reverse propagation(BP)neural network.Through proper training and learning,it was found that the presence of benzoic acid,potassium sorbate Method,compared with the analysis method of the absorption curve,the method is more intuitive,accurate and fast.In the actual food safety testing,we not only do not qualitative testing,often also need to carry out quantitative testing,with a view to assess the extent of harm to the human body.Therefore,based on the qualitative analysis of the benzoic acid and potassium sorbate in the samples,the additive in the sample was quantitatively analyzed by linear regression of the least squares algorithm(PLS),and the concentration of the samples And the absorption coefficient of the relationship between the model.It can provide the theoretical basis for the detection of benzoic acid and potassium sorbate in milk powder.