MODELING A WOOD-CHIP REFINER USING ARTIFICIAL NEURAL NETWORKS

The mechanism of the wood-chip refining process is still being studied , and no thorough model has yet been developed. Neural networks can be an attractive alternative to mathematical modeling of complex process es if a sufficient amount of input-output data is available. This arti cle examines the use of a feed-forward neural network to model a wood- chip refiner. The network's predicted outputs compared faborably with industrial refiner data. It is also shown that the network structure c an be modified to optimize refiner operation and product quality. Adva ntages and disadvantages of applying neural-networks models to simulat e and optimize industrial processes are discussed.