Powered by Blucher Proceedings

Proceedings of the Ironmaking, Iron Ore and Agglomeration Seminars


ISSN 2594-357X

Title

APPLIED MULTIVARIATE ANALYSIS FOR SINTER FEO PREDICTION

APPLIED MULTIVARIATE ANALYSIS FOR SINTER FEO PREDICTION

Authorship

DOI

10.5151/2594-357X-39579

Downloads

51 Downloads

Abstract

This work is one of the results of a successful proof of concept, with the objective to analyze and model the iron ore sintering process on a pilot plant scale, aiming at predicting the final FeO content in the sinter, as a function of the ore blend, fuel, fluxes and other process parameters. The model was developed with real data from a sintering pilot plant, considering around 300 tests with different iron ore mixtures. Multivariate analysis and machine learning techniques were applied, and a final mathematical model with R² greater than 0.92 was obtained, confirming the strength of the proposed methodology.

 

This work is one of the results of a successful proof of concept, with the objective to analyze and model the iron ore sintering process on a pilot plant scale, aiming at predicting the final FeO content in the sinter, as a function of the ore blend, fuel, fluxes and other process parameters. The model was developed with real data from a sintering pilot plant, considering around 300 tests with different iron ore mixtures. Multivariate analysis and machine learning techniques were applied, and a final mathematical model with R² greater than 0.92 was obtained, confirming the strength of the proposed methodology.

Keywords

Machine Learning; Sintering process; Iron ore, Sinter feed

Machine Learning; Sintering process; Iron ore, Sinter feed

How to cite

Nakandakari, Bianca; Paulo Lopes; Domingues, Alei. APPLIED MULTIVARIATE ANALYSIS FOR SINTER FEO PREDICTION, p. 282-295. In: 9th Brazilian Ore Agglomeration Symposium, São Paulo, 2023.
ISSN: 2594-357X, DOI 10.5151/2594-357X-39579