The use of neural networks to predict performance of a Yankee dryer is presented. A 3-layer network with 4 inputs and 2 outputs is usd. Training is performed using back-propagation algorithm and data from a Yankee simulation program based on Karlsson and Heikkila's model. The trained network is evaluated using randomly generated test cases as input. The effect of number of training cases and hidden neurons are examined.
No takes yet. Share an insight, caveat, or question.
Huang et al. (1993) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: