In this study,the prediction of moisture and fat contents of fried battered and breaded fish nuggets（BBFNs） was performed through combined response surface methodology（RSM） and artificial neural network（ANN）. RSM was utilized to collect the experimental data to establish the ANN,which the process parameters of fried BBFNs（ratio of xanthan gum to soybean fiber,drying time of BBFNs,soybean oils with different quality,frying temperature and time） and moisture and fat contents in the crust were used as the input and output,respectively. The training set was used for model fitting and the test set was used to evaluate the generalization ability of the model. The results showed that there was a significantly negative correlation between moisture and fat contents in the crust of fried BBFNs. The moisture and fat contents in the crust were significantly affected by the ratio of xanthan gum to soybean fiber,drying time of BBFNs,soybean oils with different quality,and frying temperature.However,frying time had a significant influence on moisture content in the crust,and slight influence on fat content was presented. The model was notably fitted and had the good approximation ability. Furthermore,R values for the test set of moisture and fat contents were 0. 911 and 0. 943,respectively,and there were lightly absolute errors between predicted values and actual values（range from 0. 028% to 5. 408%）,suggesting accuracy prediction of moisture and fat contents in the crust of fried BBFNs with ANN.
Journal of Wuhan Polytechnic University
Artificial neural network model
battered and breaded fish nuggets