Figure 2 -proposed On-line System Layout Controller Current Model Model Up-to-date Validation Model Model Builder Data Validation

reduce the moisture content of the pulp coming out of the presses. This will be done by obtaining, from the models, ratios to run the presses at and running the presses at those ratios. Results will be obtained during the next sugar processing campaign at the end of this year. Data analysis, making use of graphic tool XVGR available on X-terminals, will enable to asses the optimisation system efficiency. It will also facilitate definition of a control strategy for a planned on-line optimisation system. Experiments are now being conducted using evolutionary strategies, Rechenberg (5), on the same data to determine whether or not a slightly stronger search technique will give any improvement over the performance of the genetic algorithm on this application. All further results of the investigation will be published in following papers. REFERENCES 1. Holland J H (1975) " Adaptation in Natural and Artificial Systems ". 2. Fogarty T C (1989) " An Incremental Genetic Algorithm for Real-time Learning " in 'Proceedings of the 6th International Workshop on Machine Learning', pp 416-419. 3. Fogarty T C (1988) " Rule-based optimisation of combustion in multiple burner furnaces and boiler plants " in Engineering Applications of Artificial Intelligence, vol:1, iss:3, p.203-9. 4. Fogarty T C (1989) " Learning new rules and adapting old ones with the genetic algorithm " in 5. Rechenberg (1973) " Evolutionsstrategie: Optimierung technischer Systeme nach Principien der biologischen Evolution " 6.