Next Article in Journal
Interfacial Microstructure and Cladding Corrosion Resistance of Stainless Steel/Carbon Steel Clad Plates at Different Rolling Reduction Ratios
Next Article in Special Issue
Machinability Assessment and Multi-Objective Optimization of Graphene Nanoplatelets-Reinforced Aluminum Matrix Composite in Dry CNC Turning
Previous Article in Journal
Study on Near-Net Shape Forging of Large Marine Crank Throws
 
 
Article
Peer-Review Record

Feature Engineering to Embed Process Knowledge: Analyzing the Energy Efficiency of Electric Arc Furnace Steelmaking

Metals 2025, 15(1), 13; https://doi.org/10.3390/met15010013
by Quantum Zhuo 1, Mansour N. Al-Harbi 2 and Petrus C. Pistorius 1,*
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Reviewer 3: Anonymous
Reviewer 4:
Metals 2025, 15(1), 13; https://doi.org/10.3390/met15010013
Submission received: 4 November 2024 / Revised: 19 December 2024 / Accepted: 26 December 2024 / Published: 28 December 2024

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The subject is interesting and it well suits in this special issue. However, considering metallurgist readers, when the authors have focused at the computational methods, they have neglected to describe the metallurgical start points. These shortages and some notes are listed below.

The metallic charge was mainly DRI. Should you give more data about the DRI, average composition, metallization, C-content, etc. and variation range. You could organize the input data for the model calculations in a Table showing the input items and their typical values/ranges. Also the data of the furnace(s), size, type, etc. are necessary for clarity. 

p.2 line 83: median; is it really median value or mean value?

p.2 line 91: After removal of the outliers, 699 entries remained. How many values initially or were removed?

Table 1 and the text; in which stage of the smelting process the slag samples were taken? That is essential as the “FeO” contents look exceptionally high. When summarizing the percentages in Table 1 the results is appr. 95%. What is the rest?

Figure 1 and the text; although common, such acronyms like RMSE should be explained.

Ref. 6 Köhle et al is a kind of reference point for the research and for results evaluation. Therefore, it would be useful to show the “Köhle´s equation”.

Figure 3; why did you take the temperatures 1200, 1600 and 1632 C for gas, steel, and slag, respectively? In 3b would it be better to write CO2 + 2H2O on the right side (you have CO2+H2O)?

p. 8 line 231: (variables are shown in decreasing order of their effect on electricity consumption). The plot reveals that the total carbon input, injected oxygen and slag volume have the largest effect on electricity consumption…Comment: as there is no scale on y-axis, the order (from top to bottom) is only a qualitative signal.

One your result was that increasing carbon addition increases electricity, which is a questionable relation. Generally, the reason for carbon + oxygen injection is to speed up smelting and decrease electricity consumption. (Of course, foaming slag is important as well). Your observation might be connected to “too high carbon burden,” low metallization/high unreduced “FeO” in DRI. You should try to find and explain this issue, e.g., based on the "data Table" proposed above. The 2nd conclusion could be elaborated after the adjustments requested above. 

Comments for author File: Comments.pdf

Author Response

Please see the attached table with responses.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript entitled “Feature Engineering to Embed Process Knowledge: Analyzing the Energy Efficiency of Electric Arc Furnace Steelmaking ” is partly in line with the Metals journal. This article is based on computer modeling and is not supported by experiments.

 

However, the presented topic can be potentially interesting, before publication, the article requires some changes as follows:

·       The references in all of this article should be changed. Firstly there has to be the reference and the next point.

·       Abstract: The most important results should be presented.

·       Introduction (last paragraph): explain the novelty of the provided research.

·       Methods - the chapter requires significant supplementation. There is a lack of information about the methodological approach and the description of particular data, including inputs.

·       Discussion – a lack of proper scientific discussion with up-to-date literature.

·       Conclusion: This is required to be structured and measurable results should be included.

·       References: small amount of scientific articles with the last period.

Author Response

Please see the attached table with responses.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This fits into the scope of metals. Please find additional comments to further improve the quality of this work below:

-I suggest presenting more quantitative results in the abstract already. Same in the conclusions.

-You can also be more specific here. I.e. when you say it is a “large plant” you could provide the annual output...

-The information here is interesting. I feel you use a lot of ways to present a relatively simple dataset, but you can of course do this.

-I would strongly suggest language editing. The English is fine but there are really a lot of typos and formatting issues with the manuscript.

I would also suggest including a discussion section. Your results data is very abstract even though you start with a very general introduction. In a discussion section you could circle back to this general introduction and discuss the main finding of this work and how it will be relevant for industry.

 

Author Response

Please see the attached table with responses.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Feature Engineering to Embed Process Knowledge: Analyzing the Energy Efficiency of Electric Arc Furnace Steelmaking is very important paper imetallurgy of iron and steel. Minor improvements are required!

Line 12, 13, 14: The data was used to test different approaches to quantifying the effects of process conditions on specific electricity consumption (kWh per tonne of crude steel). In previous work, inputs such as the proportion of DRI, fluxes, natural gas and oxygen were linearly correlated with the specific electricity consumption (in which temperature interval?)

Line 38, 39: The aim of the work presented here was to test two different approaches to quantifying the main process variables that affect the electricity consumption in EAF steelmaking (using green hydrogen or carbon?).

Line 101:Table 1. Average slag composition (mass percentages) produced at temperature of…..

Line 176: a  1 °C increase in tapping temperature (in which temperature interval?)

Line 241; The Shapley dependence plots (Figure 6) further clarify the modeled effects of the  different input variables (such as….)

Conclusion:

Line 286, 287: The analysis demonstrates that a machine-learning  method can reveal significant and physically plausible trends when coupled with Shapley  analysis (in which temperature interval? And which temperature of formed gases?)

General question:

What are limitations  of multiple linear regression for calculation  of electricity consumption!

Author Response

Please see the attached table with responses.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have responded my comments quite well, and the supplements make the manuscript readable and easy to understand. One unclear point has remained, i.e., the response and text addition concerning "The increased electricity consumption with increased carbon input may appear surprising at first, since the carbon is oxidized in the furnace, and the oxidation of carbon by oxygen is exothermic." You have validated your conclusion based on the endothermic reaction between carbon and "FeO", although the reactions of oxygen are exothermic. Actually, the observed "cooling effect of carbon addition" must be caused by the high iron oxide ("FeO") in DRI which is then reduced in EAF. This results in lower O2:C ratio in your 80% DRI/20% scrap smelting compared to conventional EAF smelting with 100% scrap. Increased carbon input must be related to unreduced "FeO", and hence the increased electricity consumption is a consequence. In this respect the text on the top of page 8 and in Conclusions, respectively, should be modified.

 

Author Response

Please see the responses in the attached document.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The authors implement most of the comments. Slight supplementation is required in 2 points:

1) The reference style is not fully coherent with the template. Please notice where the point or comma is: 

"its carbon intensity.[2]" - incorrect

"its carbon intensity [2]." - correct 

Please apply in the whole text.

2) Discussion is still a quite generic.

Author Response

Please see the responses in the attached document.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Comments have been addressed. This is fine for me now.

Author Response

Comment 1: Comments have been addressed. This is fine for me now.

Response 1: Thank you!

Back to TopTop