Research Of The Device To Detect The Bacterial Population In Milk Based On Real-time Updating Of Model Parameters

In recent years,there has been a frequent occurrence of excessive levels of dairy products in China.The safety of dairy products has attracted wide attention,so it is very important to test the content of bacteria in milk.The color difference method can detect the content of bacteria in the milk and the fast speed,simple operation and low cost,which is of great significance to the safe production of dairy products.But the color difference method was used to detect the parameters of the bacteria group is affected by the moisture,nutrient content,temperature and humidity in milk.By continuously updating the model parameters and responding to changes in these factors,the model more accurately describes the relationship between the amount of bacteria and fade time.It can not only improve the accuracy of the detection of bacterial flora,but also can be applied to the detection of bacteria in related fields such as vinegar industry and wine industry.The parameters of the samples were affected by the changes of moisture,nutrients and temperature and humidity in milk.In this paper,the reproduction of the bacteria in milk and the change of the color of methylene blue in the milk were analyzed.Then the relationship between the content of bacteria and the fading time was established by using the color difference method.Finally,based on the color difference method,this paper introduced bayesian updating method to update the model parameters.The prior distribution of the non-information prior distribution as bayesian updating method is selected,the problem of the limited sample data of bacteria group content and fading time is solved,which reduces the uncertainty of the updated model.The updated and updated models are compared with the counting results of the plate counting method respectively.The results show that the relative error of the calculation results of the updated bacterial group content model is relatively small compared with that of the plate counting method.Moreover,with the increase of the data of bacteria group content and fading time,the model parameters are constantly updated,and the model is more accurate to calculate the bacteria content in milk.In the detection part,the photoelectric detection method can be used to quickly detect the bacteria content in milk,without complicated operation.The device adopts the two-channel transmission method to reduce the error caused by the change of the light source.And the transmission mode is adopted from the bottom-up transmission mode.To solve the problem of inaccuracy of the color information of the solution,the solution of the solution of milk and methylene blue mixed solution was settled.In terms of hardware,the design of light source,color glass,color sensor and display module,etc.White balance design in software.Using the device to test the content of bacteria in milk several time.The results show that the device has good stability and can quickly detect the bacteria content in milk.