Modeling of Alkali and Temperature Profiles in Continuous Kamyr Digester using Neural Networks

M. Tervaskanto, T. Ahvenlampi, R. Rantanen, and U. Kortela (Finland)


Neural networks, principal component analysis, pulpquality control, alkali profile


In this study the alkali and temperature profiles of the industrial continuous Kamyr digester are identified using the experimental methods. Neural networks (NN), principal component analysis (PCA) and least squares regression (LS) are used. The estimated variables are utilized in the Kappa number prediction model. The gained results were accurate and the models can be used in the on-line quality control of the continuous digester. The analytical redundancy can be achieved using the models in parallel with the on-line alkali and temperature measurements.

Important Links:

Go Back

Rotating Call For Paper Image