Simulation of onion dehydrationusing neural newtworks.
2007
Addo, A., Department of Engineering, Kwame Nkrumah University of Science and Technology, Kumasi. | Bart-Plange, A., Department of Engineering, Kwame Nkrumah University of Science and Technology, Kumasi.
A neural network model was developed to predict moisture content of onion slices during dehydration. Five input data: hot-air temperature, hot-air relative humidity, hot-air velocity, onion slice thickness and time were the input of the neural network. After trials with different networks using experimental data reported by Akbari et al. (2001), three-layer feedforward backpropagation network with 3 nodes in the hidden layer was selected. The predicted data compared well with the experimental data for different dehydration conditions. The root E mean square error (RMSE) values varied from 0.87 to 10.98 and the coefficients of determination were 0.98 D and 0.99. The NN has the potential as real-time online process control of optimum product quality during dehydration.
Show more [+] Less [-]AGROVOC Keywords
Bibliographic information
This bibliographic record has been provided by Council for Scientific and Industrial Research, Institute for Scientific and Technological Information