Research On Detection Methods Of Meat Freshness

The meats will spoilage during the storage, the current methods for detecting the meat freshness mainly are sensory methods and chemical methods. However, these methods have a highly request for the samples and testing operations and time-consuming, also is not conductive to actual production.This paper is based on the characteristic curve of meat pressure to determine the freshness of the meat, trying to find a quick and simple method to detect the freshness of meat. Main contents and conclusions are as follows:1. Measured the pressure characteristic curve of beef, chicken and pork using universal testing machine and analyze the pressure curve, we calculated mechanical parameters related to the pressure curve of the meat, these including the most vigorous F, rebound Displacement S, external force actingΦ₂, acting resilientΦ₁,power consumption thanΦ₁/Φ₂.2. Analysis the relationship between every mechanical parameter of beef, pork, chicken and the storage time, TVB-N. Respectively, we get the regression polynomial regression equation using the mechanical parameters of beef, pork, chicken and volatile basic nitrogen (TVB-N) values .3. Taking pork as the testing sample, using partial least squares elasticity to analyze the relationship between pressure mechanical parameters and the freshness, we get the regression equation. Using the regression equation to predict the value of TVB-N, we have a good result via error analysis which the result is 0.09%, indicating that these results are satisfactory. Therefore it can be concluded as follows: partial least squares method can be well applied and the elastic parameters of pork volatile basic nitrogen (TVB-N) value of the regression analysis.4. Taking pork as the testing sample, we put in use the BP neural network between pork freshness and pressure mechanical parameters and establish a nonlinear forecasting model. The establishment of the BP neural network was trained, back-contracting and tested. The original contract rate of the sample is 93.7% and the detection accuracy rate is 87.5%, indicating that neural network can be applied to predict pork freshness. 5. Researched on the different effect of the meat plastic parameters under different experimental conditions. We select a different degree of compression pressure and compression rate curve for the determination of test. Through the various mechanical parameters analysis of variance, we have the conclusion like this: the compression displacement significantly affect F,Φ₁ andΦ₂, affect the S, but not very significantly to theΦ₁/Φ₂; the compression speed affect F,Φ₂,Φ₁ and S, but not as significantly toΦ₁/Φ₂.6. We apply the electronic nose system bionic fresh pork to detect the freshness of the pork. First of all, samples for BP neural network was trained and then to return to negotiations. In the sampling of the inspection, we get the accuracy of the testing results. And the results are like this: the return negotiation was 96.4%, the accuracy of new sample ratewas 92.8%, showing that the experimental system can distinguish under different time rangeof pork.