Characterization of the Changes in Opened Sufu Bottles During Storage with Mathematical Model
JingJing Liang, Dawei Li, Ruiqin Shi, Jie Wang, Yanli Ma and Ke Xiong
DOI : 10.3844/ajbbsp.2018.285.297
American Journal of Biochemistry and Biotechnology
Volume 14, Issue 4
Artificial Neural Network (ANN) is a type of nonlinear empirical model, which can clarify the complex relation between the inputs and outputs that allows it to approximate any nonlinear function for making predictions. The objective of this study is to monitor the Biogenic Amines (BAs) content and selected physicochemical properties of sufu (a traditional Chinese fermented soybean product) along time. Simultaneously, based on initial values, a grey model and an ANN were developed to predict the influence of storage process parameters on the quality changes during storage. Results revealed that the total amounts of BAs in newly opened bottles of white, red and grey sufu were 419.61, 311.52 and 603.10 mg kg-1, respectively, no sufu samples posed the total biogenic amines tolerance level (over 1000 mg kg-1). Results showed that slight changes in the individual BAs were detected at 4°C, 15°C, 25°C and the formation of BAs was promoted at 35°C in grey sufu. Furthermore, grey model was developed with average relative errors within ±7%, the statistical parameters (R2) of pH, water activity and amino nitrogen was all above 0.90. In the ANN, the number of neurons in the hidden layer was optimized, ten neurons revealed a positive correlation between the values obtained experimentally and those predicted values (R2 = 0.99). So ANN with highest R2 was selected to predict biogenic amines and our results demonstrated that grey sufu were not edible on the 25th days at 4°C (BAs? 1000 mg kg-1) and it would be better if white and red sufu are consumed within 40 days. We envision that our works can be used for proving a reference for consumers and offer new perspectives by mathematical model to avoid difficult, costly and time-consuming quality inspection, particularly in the field of storage.
© 2018 JingJing Liang, Dawei Li, Ruiqin Shi, Jie Wang, Yanli Ma and Ke Xiong. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.