The prediction of rolled product properties by means of correlation with as cast structure and rolled product, plant process variables, plant process variables including modelling of segregation


The main objective of this RFCS project was to develop a method (or methods) of predicting the rolled product quality from data from the as-cast structure as well as caster operational (including casting machine condition) data.

Models of macrosegregation in the as-cast structure have been developed to support the prediction modelling methods developed within the project. Neural network, fuzzy logic and regression analysis have been used to derive predictive models for rolled product quality.

Plant data have been collected by the industrial partners and distributed to all partners for subsequent analysis.

Start Year: 
End Year: 
Contact Person: