Machine Learning Strategies for Control of Flotation Plants

摘要: Abstract Although flotation processes are notoriously difficult to model from first principles, knowledge-based systems can be used great advantage monitor and control plants, provided that process knowledge captured effectively on the plant. By making use of machine learning techniques features surface froths cells construct representations behaviour a Two probabilistic decision tree methods backpropagation neural net were all equally capable classifying different at least as well human expert. Explicit trees derived, relating froth characteristics structures. Relatively sharply clustered Sammon maps structures obtained, allowing good visualisation multidimensional data.