Designing of intelligent expert control system using petri net for grinding mill operation
ICAI'05/MCBC'05/AMTA'05/MCBE'05 Proceedings of the 6th WSEAS international conference on Automation & information, and 6th WSEAS international conference on mathematics and computers in biology and chemistry, and 6th WSEAS international conference on acoustics and music: theory and applications, and 6th WSEAS international conference on Mathematics and computers in business and economics
| Hi-index | 0.00 |
The paper proposes a new supervised training algorithm for feed-forward neural networks. Instead of applying single valued input- output information at a time , multi-valued information in the form of a K - dimensional vector (K1) are applied to each node of the input - output layer. Weights are adjusted using gradient decent-approximation method in order to minimize the sum-squared error value at each node of the output layer. The training algorithm has been studied for wide range of input-output value and gives worthy results specially when the output vector is small enough compared to the input vector. The paper suggests a judicious method for choosing bias component of the sigmoidal activation function used in the training algorithm.