| Dairy products are easily contaminated by microorganisms due to their complex production processes,which can cause harm to human health.Its safety and quality are related to public health safety.Enterprises should manage and control the risks of microbial contamination in a timely manner and strictly supervise the production process to ensure that the concentration of microorganisms in dairy products meets the requirements of national standards.However,the current detection time of the microbial detection method specified in the national standard is more than 24 hours by the plate counting method,which cannot meet the repetitive detection requirements of enterprises that should be checked and checked at every step.The ultraviolet spectroscopy has the advantages of fast,simple,low cost,etc.,and is more suitable for online detection in actual production situations.However,dairy products contaminated by microorganisms are often not a single bacterial group.The background material of dairy products is complicated.When the tested bacteria and interfering bacteria exist at the same time,the ultraviolet-visible spectrum of the tested bacteria has red shift,overlap,and spectral scattering problems.The characteristics of the tested bacteria are difficult to separate,and the concentration detection model is inaccurate.Therefore,this paper selects raw milk as the main raw material in dairy products as the representative of the detection background,and the E.coli with the highest detection rate is selected as the detection object.The yeasts that are widely present in cow milk are Interfering bacteria,UV-Vis spectroscopy characteristics separation and concentration detection research have provided an important theoretical basis and reference basis for the actual promotion and application of UV-Vis spectroscopy technology in dairy safety detection,which is of great benefit to the development of dairy safety.The specific research of this article is as follows:The principle of quantitative detection of ultraviolet-visible spectroscopy,the cause of characteristic absorption,and the photochemical characteristics of the measured substance are theoretically analyzed.E.coli in water,Saccharomyces cerevisiae in water,mixed two bacteria in water,E.coli in milk and interfering yeast yeast The UV-Vis spectrum acquisition experiment of bacterial mixture,analyze the information and characteristics of its spectral data.According to the results of the microbial experiment in water and the two-dimensional correlation principle,a plan for the characteristic separation experiment of E.coli in milk was designed.The concentration of E.coli was introduced as a system disturbance factor.Under the premise of a fixed concentration of interfering bacteria and yeast,the large intestine was changed in a gradient Bacillus concentration as an external disturbance factor of the two-dimensional spectrum designed a collection experiment of ultraviolet-visible spectrum data of the sample to be tested,obtained effective ultraviolet spectrum data information,and carried out an experiment of yeast characteristic separation.In view of the scattering problems in the spectral data,the overlap of multiple component features and the redshift of E.coli features,the standard normal change method is used to correct the errors caused by scattering,and the two-dimensional correlation analysis method is used to realize the feature separation of E.coli and yeast.A prediction model of E.coli concentration based on BP neural network was established.The results are:the degree of fit R~2=0.9556,the mean square error MSE=0.0498,and the prediction time is 20s.The results show that this method can eliminate the interference of other bacterial species and meet the needs of detection. |
PDF Full Text Request