Application Of Chemometrics In Food Preservatives

The article based on the knowledge and methods of chemometrics and chemical structure of the network to build food additives static parameters of the molecular structure of the network for the first time, studied and completed quantitative structure-activity relationship (QSAR) of analysis in certain food additives (QSAR), proposed a new method quantitative structure-activity relationship (QSAR) analysis in food additives, in order to provide a scientific basis for the prediction and guide the development of non-toxic or low toxic green food additives.It also utilized more advanced extreme learning machine (ELM) to detect and study content of some food preservatives in food, the use of characteristic spectral bands in the ultraviolet and visible analysis of the range, This is for the use of common instruments and sample spectra without separation and detection of food preservatives completed provides a fast, simple, inexpensive (as now many detection methods are used expensive HPLC) method.The main contents and results of this study are as follows:1、Molecular structure network static parameters for food preservatives quantitative structure-activity/property relationship studiesThe paper utilize the network of scientific theory and method, the atom that composed the molecule as a network node, a planar structure of molecular is passed by the edges connecting these nodes together to construct the molecular structure of the network; the18network static characteristic variables that calculate the average degree, the average path length, etc the molecular structure of the network as QSAR candidate arguments. In order to better carry out QSAR analysis, using BP artificial neural network variable from18candidates were selected from the active contribution of the proximity of the center of a large maximum, etc.4independent variables, then with support vector regression (SVR) and BP neural network on the molecular structure of the network of4independent variables and the rat by oral LD50for QSAR studies respectively. Experimental results show that:support vector machine regression that based on the molecular structure of the network static parameters model has good predictive ability. This new method that applying for network static parameter of molecular structure establish model has certain application prospect for the same (or more) property/activity of a class of substances QSAR analysis and research.2、Application of Extreme Learning Machine (ELM) to detect the content of multiple preservatives simultaneously in foodContent of food preservatives for food safety has been a focus of the study, excessively use will cause some degree of damage for body. In this experiment, computer technology combined with the use of extreme learning machine UV-visible spectrophotometer to predict the content of multiple preservatives in food. The experiment proposed to extract characteristic spectral data segment as extreme learning machine training and testing samples from UV-visible spectra curves of the experiment, in order to achieve a variety of food samples preservative content testing using extreme learning machine. From the research results can be seen:Extreme Learning Machine for some foods Sodium benzoate content to predict its decisive factor R2=0.99998, and its Mean Squared Error, MSE=2.4025e-009, compared to the use of BP neural network R2=0.91643, MSE=0.072245, extreme learning machine has a better explanation generalization ability, That illustrates the extreme learning machine (ELM) has better precision and accuracy, while its measurement speed, stability and other performance than BP neural network has a more significant improvement. It also provides valuable learning and reference for multiple food related preservatives or other additives the content of prediction and detection...