ITEM METADATA RECORD
Title: Hybrid artificial Neural Network and Genetic Algorithm modelling of slag properties
Authors: Beyers, Lesley
Dutta, Abhishek
Lahiri, Sandip K.
Blanpain, Bart
Verhaeghe, Frederik
Issue Date: 2015
Publisher: GDMB Verlag
Host Document: Proceedings European Metallurgical Conference 2015 vol:2 pages:1071-1085
Conference: European Metallurgical Conference (EMC) edition:8 location:Düsseldorf, Germany date:14-17 June 2015
Abstract: Slag property data is indispensable in developing mathematical models for the kinetics and the heat, mass and fluid transport in pyrometallurgical processing. The optimisation of pyrometallurgical processes requires accurate information on the slag composition and the properties, which both vary during the process. Slag property models alleviate the amount of measurements needed to account for these changes and allow for dynamic calculation of the progress of a process, by including them in the mathematical process model.
In this study, a hybrid artificial neural network and genetic algorithm technique (ANN-GA) is used to model the density, electrical conductivity and oxidation state of multicomponent melts. The ANN-GA model was chosen because a first principle model is not available for complex industrial systems and the method can be used to circumvent limitations imposed by the assumptions in phenomenological models. The strategy uses ANN as the nonlinear process modelling paradigm, and GA for optimizing the meta-parameters of the ANN model such that an improved prediction per-formance is realised. It was found that the ANN-GA model can accurately predict the slag density (output) within experimental error and substantially improves the predictions of the electrical conductivity and oxidation state for a wide range of process conditions.
Publication status: published
KU Leuven publication type: IC
Appears in Collections:Materials Technology TC, Campus Group T Leuven
Department of Materials Engineering - miscellaneous
Technologiecluster Materialentechnologie
Sustainable Metals Processing and Recycling

Files in This Item:
File Description Status SizeFormat
Copyright Agreement_Signed.pdfSigned Copyright Agreement Published 326KbAdobe PDFView/Open Request a copy
Paper_LJBeyers_authorfinal_LIRIAS.pdfAuthor's version send to Publisher Published 267KbAdobe PDFView/Open Request a copy

These files are only available to some KU Leuven Association staff members

 




All items in Lirias are protected by copyright, with all rights reserved.