Buoyant Flow and Thermal Analysis in a Nanofluid-Filled Cylindrical Porous Annulus with a Circular Baffle: A Computational and Machine Learning-Based Approach

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Technical Comment on Approach:

While the research question is valid, the approach needs further improvement. In the manuscript, the authors conducted CFD, which seems accurate—so why pursue the ML approach? If you performed CFD, which is computationally intensive, it leads to accurate results. So, the question is: Why do we need an ML model with slightly lower accuracy compared to CFD? The use of ML should be clearly justified.
Also, we read through the methodology: "The primary motivation is to develop reliable and accurate predictive models to support thermal equipment designers." It's unclear how designers can use the ML model. The authors only provided a table on statistical comaprision (table 4).  In the current manuscript, it seems that ML is briefly attached to the main discussion with one table (Table 4) to increase the level of novelty, which is not a professional decision. Also, the type of inputs, the number of data (training set, test set, etc.) is not justified for ML models.  The methodology for ML is very broad and generic, explaining that some preprocessing was done.

Comment on the Format:
There appears to be a typo in Fig. 7's x-label, as it shows two 10e-1 entries.

Author Response

Please see attached file.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

In this paper, the behavior of buoyancy-driven convection and heat transfer in the annular region of a nanofluid-filled porous cylinder is investigated, with a special focus on the annular region with a circular baffle. A combination of computational fluid dynamics (CFD) and machine learning (ML) was used to analyze the effects of various parameters on fluid flow and heat transfer. The research provides theoretical support for designing efficient heat exchangers and cooling systems, and demonstrates the potential of machine learning in complex fluid dynamics problems. By optimizing the baffle design and nanofluid characteristics, the performance of the thermal management system can be significantly improved.

1.The article assumes that nanofluids and porous media are in a state of local thermal equilibrium, which may not be true for some high RA numbers or fast transient processes. Have the effects of local non-thermal equilibrium been considered?

2.Line23:“Thoroughly” should be removed.

3.In Fig1(a, b) the words are too small to be legible, and we think the label should be made larger.

4.Whether the top and bottom lines of a three-line chart should be thicker than the middle line?

5.The left image in Fig2 is not clearly labeled.

6. In Fig. 7, the number fonts should be the same as in Fig. 5 and Fig. 6, and the scale direction should be the same.

Comments on the Quality of English Language

The use of professional terms is appropriate and the fluency is high, but there is still room for improvement

Author Response

Please see attached file.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This paper presents a parametric study of a saturated nanofluid (NF) flowing through a porous annulus fitted with an inner‐wall circular hot baffle. However, I fail to see the point of subsequently applying various machine‐learning techniques. For such a relatively straightforward problem, whenever you need results under new operating conditions, it would be far more efficient simply to rerun the simulations rather than build surrogate models.

Moreover, the authors’ comparison of different algorithms strikes me as superficial. They report that neural networks underperform compared to Gradient Boosting, yet give no details about how either model was implemented—no hyperparameter settings, training procedures, data splits, or validation strategies—so the assertion lacks any solid justification. Finally, I recommend spelling out all acronyms in both the title and conclusions rather than relying on abbreviations alone.

Author Response

Please see attached file.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

 

Author Response

We thank the reviewer for all his efforts and the positive evaluation of our work.