Title

MODELING PHYSICAL PROPERTIES OF STEEL SLAG BASED ON NEURAL NETWORKS PART 2: GLASS TRANSITION TEMPERATURE

MODELING PHYSICAL PROPERTIES OF STEEL SLAG BASED ON NEURAL NETWORKS PART 2: GLASS TRANSITION TEMPERATURE

Authorship

ANJOS, PATRICK QUEIROZ DOS; GRILLO, FELIPE FARDIN; MACHADO, MARCELO LUCAS PEREIRA; QUARESMA, LUCAS DE ALMEIDA

ANJOS, PATRICK QUEIROZ DOS

I am this author

GRILLO, FELIPE FARDIN

I am this author

MACHADO, MARCELO LUCAS PEREIRA

I am this author

QUARESMA, LUCAS DE ALMEIDA

I am this author

DOI

10.5151/2594-5300-40944

Downloads

Abstract

THE GLASS TRANSITION TEMPERATURE IS ONE OF THE MOST IMPORTANT PHYSICAL PARAMETERS IN RELATION TO THE CRYSTALLINE BEHAVIOR OF MATERIALS AND HAS INFLUENCE UM PHYSICAL PROPERTIES IN MOLD SLAG AND FE-CR-BASED SLAGS. HENCE THE GLASS TRANSITION TEMPERATURE IS AN IMPORTANT PARAMETER IN STEEL SLAGS. THE SCIGLASS DATABASE WAS USED TO PROVIDE STEEL SLAG DATA WITH THE SIO2-CAO-AL2O3-MGO-FEO-NA2O-K2O-LI2O-B2O3-BASED SLAGS AND GLASS TRANSITION TEMPERATURE SYSTEM. THE CHEMICAL COMPOSITION DATA WERE CONVERTED INTO THE DEGREE OF DEPOLYMERIZATION PARAMETERS AND SUBSEQUENTLY RELATED TO EACH GLASS TRANSITION TEMPERATURE. THE MODELING WAS CARRIED OUT USING NEURAL NETWORKS BY VARYING WIDTH AND DEPTH USING A LINEAR COMBINATION BETWEEN DIFFERENT CENTRAL MOMENTS AS A REFERENCE OF EFFICIENCY. THE CHOSEN NEURAL NETWORK HAD A WIDTH OF 14 AND A DEPTH OF 10 (14-10). SENSITIVITY ANALYSIS WAS PERFORMED DEMONSTRATING CONSISTENCY WITH THE LITERATURE AND STATISTICAL EVALUATIONS WERE PERFORMED TO DEMONSTRATE THE EFFICIENCY OF THE NEURAL NETWORK 14-10 WHICH DEMONSTRATED BETTER EVALUATIONS IN RELATION TO DIFFERENT LITERATURE EQUATIONS.

THE GLASS TRANSITION TEMPERATURE IS ONE OF THE MOST IMPORTANT PHYSICAL PARAMETERS IN RELATION TO THE CRYSTALLINE BEHAVIOR OF MATERIALS AND HAS INFLUENCE UM PHYSICAL PROPERTIES IN MOLD SLAG AND FE-CR-BASED SLAGS. HENCE THE GLASS TRANSITION TEMPERATURE IS AN IMPORTANT PARAMETER IN STEEL SLAGS. THE SCIGLASS DATABASE WAS USED TO PROVIDE STEEL SLAG DATA WITH THE SIO2-CAO-AL2O3-MGO-FEO-NA2O-K2O-LI2O-B2O3-BASED SLAGS AND GLASS TRANSITION TEMPERATURE SYSTEM. THE CHEMICAL COMPOSITION DATA WERE CONVERTED INTO THE DEGREE OF DEPOLYMERIZATION PARAMETERS AND SUBSEQUENTLY RELATED TO EACH GLASS TRANSITION TEMPERATURE. THE MODELING WAS CARRIED OUT USING NEURAL NETWORKS BY VARYING WIDTH AND DEPTH USING A LINEAR COMBINATION BETWEEN DIFFERENT CENTRAL MOMENTS AS A REFERENCE OF EFFICIENCY. THE CHOSEN NEURAL NETWORK HAD A WIDTH OF 14 AND A DEPTH OF 10 (14-10). SENSITIVITY ANALYSIS WAS PERFORMED DEMONSTRATING CONSISTENCY WITH THE LITERATURE AND STATISTICAL EVALUATIONS WERE PERFORMED TO DEMONSTRATE THE EFFICIENCY OF THE NEURAL NETWORK 14-10 WHICH DEMONSTRATED BETTER EVALUATIONS IN RELATION TO DIFFERENT LITERATURE EQUATIONS.

Keywords

GLASS TRANSITION TEMPERATURE; STEEL SLAG; NEURAL NETWORKS; PHYSICAL PROPERTIES

Glass transition temperature; Steel Slag; Neural networks; Physical Properties.