Neural network prediction of ascorbic acid degradation in green asparagus during thermal treatments

Authors:
Hong Zheng;Shuangshuang Fang;Heqiang Lou;Yong Chen;Lingling Jiang;Hongfei Lu
Affiliations:
College of Chemistry and Life Science, Zhejiang Normal University, Jinhua 321004, China;College of Chemistry and Life Science, Zhejiang Normal University, Jinhua 321004, China;College of Chemistry and Life Science, Zhejiang Normal University, Jinhua 321004, China;College of Chemistry and Life Science, Zhejiang Normal University, Jinhua 321004, China;College of Chemistry and Life Science, Zhejiang Normal University, Jinhua 321004, China;College of Chemistry and Life Science, Zhejiang Normal University, Jinhua 321004, China
Venue:
Expert Systems with Applications: An International Journal
Year:
2011

Citing 3
Cited 1

Multilayer feedforward networks are universal approximators

Neural Networks
Application of neural networks to an emerging financial market: forecasting and trading the Taiwan stock index

Computers and Operations Research - Special issue: Emerging economics
A novel method for shelf life prediction of a packaged moisture sensitive snack using multilayer perceptron neural network

Expert Systems with Applications: An International Journal

Automatic recognition of quarantine citrus diseases

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	12.05

Visualization

Abstract

An artificial neural network was developed to predict the kinetics of ascorbic acid loss in green asparagus during thermal treatments and the model was trained using a back-propagation algorithm. The results indicate that the optimal ANN models consisted one hidden layer and the optimal number of neurons in the hidden layer was 24, 26, 26 and 18 for bud, upper, middle and butt segments of asparagus, respectively. The ANNs could predict the kinetic parameters of ascorbic acid degradation in asparagus with an MSE of 1.3925 and MAE 0.5283 for bud segment, MSE 2.4618 and MAE 0.6436 for upper segment, MSE 0.8985 and 0.4258 for middle segment and MSE 0.2707 and MAE 0.1883 for butt segment. In addition, the correlation coefficients between experimental k, t"1"/"2 or D-value and predicted values were greater than 0.99 in all cases. Therefore, ANN offers a simple, quick and convenient means of the kinetic parameters prediction in chemical kinetics.