Rashid, Roslina and Jamaluddin, Hishamuddin and Saidina Amin, Nor Aishah (2004) Application of radial basis function network in modeling the tapioca starch hydrolysis. In: International Conference on Artificial Intelligence in Engineering and Technology, 2004, Kota Kinabalu, Sabah..
In recent years, there has been a resurgence of interest in the potential of artificial neural networks as a modeling tool in bioprocess. This work performed a comprehensive study on the predictive performance of Radial Basis Function (RBF) network modeling for tapioca starch hydrolysis process using ortogonal least squares (OLS) learning algorithm. The models performances are evaluated by error index values and graphical fit of train and test sets data. Detailed examinations of the effects of the performance goal (tolerence) and radius of the gaussian function on the prediction accuracy of the model were investigated. Investigation reveals that the required number of RBF centers and the error index values heavily depended on radius factor and tolerence expressed as sum of squared errors (SSE).
|Item Type:||Conference or Workshop Item (Paper)|
|Uncontrolled Keywords:||Dynamic modeling, orthogonal least squares, radial basis function, starch hydrolysis|
|Divisions:||Chemical and Natural Resources Engineering (Formerly known)|
|Deposited By:||Prof Madya Issham Ismail|
|Deposited On:||18 Dec 2007 01:27|
|Last Modified:||01 Jun 2010 03:20|
Repository Staff Only: item control page