Reducing EAF Carbon Emissions Through Input and Slag Analysis
Journal of Petroleum Technology · 01.01.2026 · 78 (01) · 252-267
DOI: 10.2118/tr/0426 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Petroleum Technology, online erschienen am 01.01.2026. Verfasst von Quantum Zhuo und 1 weiteren Autor:innen. Die Arbeit nennt 15 Literaturangaben.
Abstract (Originalsprache)
Five data sets, each containing a year’s worth of electric arc furnace (EAF) steel production, were analyzed for electricity consumption trends using a literature multilinear regression (MLR) model, a modified MLR model with engineered variables, and a machine learning (ML) model utilizing XGBoost and Shapley analysis. While the literature model provided a stronger fit to data, the engineered model resulted in more physically probable results and is recommended due to a wider applicability. The results of the ML analysis provided an auxiliary analysis that confirmed several accuracies in the engineered model; however, the results exposed some weaknesses of the model, namely its sensitivity to outlier data points. It is determined that the analysis of EAF electricity usage is most effectively done by utilizing the MLR with engineered variables in conjunction with an ML analysis.
Zitieren
APA
Zhuo, Q., & Pistorius, P. C. (2026). Reducing EAF Carbon Emissions Through Input and Slag Analysis. Journal of Petroleum Technology, 78(01), 252-267. https://doi.org/10.2118/tr/0426
Vancouver
Zhuo Q, Pistorius PC. Reducing EAF Carbon Emissions Through Input and Slag Analysis. Journal of Petroleum Technology. 2026;78(01):252-267. doi:10.2118/tr/0426
BibTeX
@article{Zhuo2026,
author = {Zhuo, Quantum and Pistorius, P. Chris},
title = {Reducing EAF Carbon Emissions Through Input and Slag Analysis},
journal = {Journal of Petroleum Technology},
year = {2026},
volume = {78},
number = {01},
pages = {252--267},
doi = {10.2118/tr/0426}
}
EUVIMED