Sponsored Research (Hyundai Steel)
Research Motivation
Electric Arc Furnace (EAF) has recently gained attention due to its eco-friendly, scrap-recycling approach and its potential for carbon-reduction. The type of special steel producted depends on its carbon content of the molten steel which must be therefore controlled. This research aims to develop a decision support system for predicting the carbon content using data-driven modeling to reduce carbon emissions and power consumption during the EAF process.
Key Idea
This research proposes a methodology that includes four different steps: 1) collect raw data and perform data pre-processing (e.g., outlier detection) using statistical methods to obtain refined data, 2) select features based on domain knowledge, 3) develop a machine learning based regression model, divided into low carbon steel group and high carbon steel group, and 4) identify correlations between input variables and predicted output values with sensitivity analysis.
Overview of the proposed methodology
Results
The results show that the predicted carbon contents are within a 0.05% margin of error compared to the actual values for 99% of the low carbon steel data and 97% of the high carbon steel data. Also, it indicates that the carbon and oxygen added after first temperature measurement, and the carbon content at first temperature measurement had the greatest impact on the final carbon content. Carbon content at second temperature measurement tends to increase as carbon content at first temperature measurement and carbon added after first temperature measurement increase, but decrease as oxygen added after first temperature measurement increases.
Publications
[1] Applied Energy (Under review)
Commentaires