Towards an on-line measurement of the purity cane sugar factory juices: A neural network approach

For most semi-batch crystallization processes, it is still difficult to obtain on-line relevant information about the purity of sugar solutions. Consequently, very few studies dealing with the estimation of the purity, can be found in the literature. Therefore, the aim of this article is to present a new method of on-line measurement of the purity of cane sugar factory juices used during the crystallization. The first results reinforce the reliability of neural networks and recommend their uses to estimate the purity which is a significant parameter when dealing with crystallization supervision.