Developing a generalised neural-fuzzy hydrocyclone model for particle separation

Development of a neural-fuzzy model for an operational hydrocyclone is reported in this paper. The model integrates the benefits of the artificial neural network (ANN) and the fuzzy-logic techniques. It preserves the generalisation capability of an ANN, while expressing the final model in fuzzy rules. These rules can be modified and examined by the user. This will in turn control the interpretation ability of the system. Results from a case study have shown that the new proposed neural-fuzzy hydrocyclone model produces comparable results as those from the ANN model but with an added advantage of the use of linguistic fuzzy rules