Quantitative analysis of fatty acid value during rice storage based on olfactory visualization sensor technology

Abstract

Fatty acid value is one of the important indicators to judge whether aging occurs during rice storage. This study innovatively proposes to use a homemade olfactory visualization sensor to realize the quantitative determination of fatty acid value during rice storage. First, based on GC–MS analysis results, 14 porphyrins and one pH indicator were preliminarily selected to prepare an olfactory visualization sensor array for obtaining odor information of rice samples of different storage periods. Then, the ant colony optimization combined with back propagation neural networks (ACO-BPNN) was used to optimize the color components of the preprocessed array difference image to obtain the best combination of color components. Finally, a back propagation neural networks (BPNN) model was established based on the optimized color component combination, and the number of hidden layer nodes of the network was optimized by using leave-one-out cross-validation (LOOCV). The results showed that the best BPNN model constructed on the color component numbers 41, 11 and 34, and the optimal topology of the network was 3-12-1. The correlation coefficient (RP) and root mean square error of prediction (RMSEP) of the best BPNN model were 0.9715 and 0.4310 mg/100 g, respectively. The overall results demonstrate that it is feasible to use the homemade olfactory visualization sensor to detect fatty acid values during rice storage. In addition, the ACO-BPNN can find the best color components combination, which can effectively improve the detection accuracy and stability of the BPNN model, and also reduce the cost of preparing the olfactory visualization sensor array.