A Cutting Force Prediction Model for Corner Radius End Mills Based on the Separate-Edge-Forecast Method and BP Neural Network

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1. Some long sentences have multiple nesting, which affects logical coherence. Suggest using a "topic sentence+elaboration" structure, for example, breaking down "The model, which integrates..." into "This study proposes a hybrid model (Fig. 3) that integrates... The integration approach addresses...".
2. The terminology description throughout the text should maintain consistency. For example, the full name of "corner radius end mills" appears in the abstract, and the abbreviation "CREM" is used multiple times in the main text without a clear definition. It is recommended to standardize the labeling of "Corner Radius End Mills (CREM)" when it first appears, and to use abbreviations uniformly in the future.
3. The reference format is mixed between APA and IEEE formats, such as [5] using journal abbreviations and [8] not indicating volume numbers. It is recommended that authors adopt the format template required by the journal and use tools such as EndNote to generate standardized entries.
4. The chart format is not standardized, with missing units for the coordinate axes in Figure 2 and no data source indicated in Table 3.
5. The paper does not clearly compare the fusion mechanism between traditional element force models and neural network models. Please explain the reason. Alternatively, a schematic diagram of the fusion strategy can be added to quantitatively illustrate the method for determining the weight allocation coefficient α.
6. The paper material only uses AISI 1045 steel and lacks material universality verification. Relevant literature citations can be added.
7. The discussion section is entirely qualitative in nature. It is suggested to add evaluation of quantitative results to enrich the discussion content.

Author Response

Comments 1: Some long sentences have multiple nesting, which affects logical coherence. Suggest using a "topic sentence+elaboration" structure, for example, breaking down "The model, which integrates..." into "This study proposes a hybrid model (Fig. 3) that integrates... The integration approach addresses...".

Response 1: Thank you for your valuable feedback on our manuscript. We agree with this comment. Therefore, We have revised the long sentences in the paper to enhance logical coherence. (page 1, line 13-19; page 2, line 84-88; page 3, line 110-114; page 6, line 187-188; page 9, line 110-114; page 13, line 337-338; page 14, line 348-349. )

Comments 2: The terminology description throughout the text should maintain consistency. For example, the full name of "corner radius end mills" appears in the abstract, and the abbreviation "CREM" is used multiple times in the main text without a clear definition. It is recommended to standardize the labeling of "Corner Radius End Mills (CREM)" when it first appears, and to use abbreviations uniformly in the future.

Response 2:Thank you for pointing this out. We agree with this comment. Therefore, We have annotated and uniformly adopted the abbreviation for "corner radius end mill" throughout the text. (page 1, line 10; page 3, line 134, line 138, line 139; page 4, line 150, line153; page 5, line 164, line 165, line 169, line 170, line 171, line 179; page 6, line 197; page 8, line 231, line 234; page 13, line 318, line 346; page 14, line 350, line 380)

Comments 3: The reference format is mixed between APA and IEEE formats, such as [5] using journal abbreviations and [8] not indicating volume numbers. It is recommended that authors adopt the format template required by the journal and use tools such as EndNote to generate standardized entries.

Response 3: Thank you for your valuable feedback on our manuscript. We agree with this comment. Therefore, We have modified the reference format to comply with the journal's requirementsBased on our review of relevant documentation, we believe that  "Micromachines" cited in Reference 5 corresponds to the full title of the journal rather than an abbreviated form. (page 2, line 43, line 60, line 70; page 3, line 126; page 15, line 388, line 390, line 394-395, line 398, line 399, line 403, line 406, line 425; page 16, line 432, line 443, line 469)

Comments 4: The chart format is not standardized, with missing units for the coordinate axes in Figure 2 and no data source indicated in Table 3.

Response 4: Thank you for your valuable feedback on our manuscript. We have checked the formatting of all figures in the manuscript. In our perspective, as Figure 2 serves as a schematic illustration of the tool geometry, we believe it is appropriate to omit unit notation. This paper did not include Table 3. Due to confidentiality agreement, the complete raw data cannot be disclosed. However, partial data may be made available upon request by contacting the corresponding author.(page5, figure2; page 6, figure 3; page 8, figure 5; page 10-11, figure 9; page 11, figure 10)

Comments 5: The paper does not clearly compare the fusion mechanism between traditional element force models and neural network models. Please explain the reason. Alternatively, a schematic diagram of the fusion strategy can be added to quantitatively illustrate the method for determining the weight allocation coefficient α.

Response 5: Thank you for pointing this out. In my opinion, traditional mechanistic force models primarily focus on the analysis of the cutting process, and a significant limitation of neural network models is their lack of interpretability regarding the decision-making process. This fundamental discrepancy makes it difficult to integrate neural network approaches into traditional mechanistic modeling frameworks. Therefore, a direct comparison of their integration mechanisms may not be necessary. In the model proposed in this study, neural networks are employed solely for curve-fitting purposes and do not engage in interpreting the physical cutting process. It is only under this specific condition that the integration of a neural network becomes feasible.

Comments 6: The paper material only uses AISI 1045 steel and lacks material universality verification. Relevant literature citations can be added.

Response 6: Thank you for your valuable feedback on our manuscript. We would like to kindly explain that the decision to use only one material in the milling tests was based on the following considerations:

1. The primary objective of this paper is to demonstrate the innovativeness of the proposed predictive model.
2. The corner radius end mill selected for this paper is specifically designed for aluminum alloys, and may not be suitable for machining other materials.

Regarding the relevant researches on milling force modeling for corner radius end mills, existing references have been included in the Introduction section. We have further refined the related descriptions to more clearly indicate the workpiece materials used in those studies.(page 3, line 140-142; page 4, line 144-147, line 150-152)

Comments 7: The discussion section is entirely qualitative in nature. It is suggested to add evaluation of quantitative results to enrich the discussion content.

Response 7: Thank you for your valuable feedback on our manuscript. We agree with this comment. Therefore, We have added commentary on the quantitative results.(page 13, line 321-331)

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The paper has potential findings for the publication. However, following few points need to be addressed:

1. What is the difference between points a and b in Figure 3 if both are for “Without considering runout”?

2. The work should include the commercial index of the tool and its manufacturer.

3. On Figure 8 the autors uses the word "Danamomter", it should be "Dynamometer." This should be corrected.

4. Machining nomenclature does not use "the lag angle" or "the immersion angle." Correct terminology must be used.

5. What is the direction of travel of the workpiece in Figure 4? If the top black arrow points in this direction, the material will move away from the cutting tool and no machining will occur.

6. In Figure 6, the "a_r" parameter is used. In the milling process, the milling width is indicated by the symbol "a_e" and not "a_r." This should be corrected.

7. The authors included test results for an aluminum alloy in their paper. Have similar tests been conducted for other materials, such as steel, nickel alloys or titanium alloys?

8. The legend in Figure 11 are illegible. This should be corrected.

Author Response

Response to Review 2 Comments:

Comments 1: What is the difference between points a and b in Figure 3 if both are for “Without considering runout”?

Response 1: Thank you for your valuable feedback on our manuscript. We sincerely apologize for your confusion caused by the oversight in our manuscript. Figure 3. b represents the case considering tool runout. We have corrected this error.(page 3, line 193-194)

Comments 2:The work should include the commercial index of the tool and its manufacturer.

Response 2: Thank you for your valuable feedback on our manuscript. We agree with this comment. We have added information regarding the cutting tool in our manuscript.(page 9, line 260-262; page 10, line 266; page 14, line 380)

Comments 3: On Figure 8 the autors uses the word "Danamomter", it should be "Dynamometer." This should be corrected.

Response 3: Thank you for your valuable feedback on our manuscript. We agree with this comment. We have corrected this error.(page 10, figure 8)

Comments 4: Machining nomenclature does not use "the lag angle" or "the immersion angle." Correct terminology must be used.

Response 4: Thank you for your valuable feedback on our manuscript. Based on our review of relevant literature, we have revised "the lag angle" to "the radial lag angle" and "the immersion angle" to "the axial immersion angle" to align with standard terminology.(page 5, line 173, line 176)

Comments 5: What is the direction of travel of the workpiece in Figure 4? If the top black arrow points in this direction, the material will move away from the cutting tool and no machining will occur.

Response 5: Thank you for pointing this out. The workpiece feed direction is consistent with the rotational direction of the tool, as indicated by the black arrow. This machining method is known as down milling. Although the phenomenon of the tool apparently pushing the material away does occur during down milling, it does not cause the machining process to halt. The primary reason is that this phenomenon is instantaneous and localized, whereas the overall machining process is dynamic and macroscopic.

Comments 6: In Figure 6, the "ar" parameter is used. In the milling process, the milling width is indicated by the symbol "ae" and not "ar." This should be corrected.

Response 6: Thank you for your valuable feedback on our manuscript. We agree with this comment. Therefore, We have thoroughly reviewed the entire manuscript and made revisions specifically addressing this issue.(page 8, line 230, equation 9, equation 10)

Comments 7: The authors included test results for an aluminum alloy in their paper. Have similar tests been conducted for other materials, such as steel, nickel alloys or titanium alloys?

Response 7: Thank you for your valuable feedback on our manuscript. We would like to kindly explain that the decision to use only one material in the milling tests was based on the following considerations:

1. The primary objective of this paper is to demonstrate the innovativeness of the proposed predictive model.
2. The corner radius end mill selected for this paper is specifically designed for aluminum alloys, and may not be suitable for machining other materials.

Regarding the relevant researches on milling force modeling for corner radius end mills, existing references have been included in the Introduction section. We have further refined the related descriptions to more clearly indicate the workpiece materials used in those studies.(page 3, line 140-142; page 4, line 144-147, line 150-152)

Comments 8: The legend in Figure 11 are illegible. This should be corrected.

Response 8: Thank you for your valuable feedback on our manuscript. We have readjusted the relevant legend in Figure 11.(page 12, figure 11)

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The paper has been supplemented and revised according to the suggested revisions, and the existing issues have also been resolved.

Author Response

We are truly grateful for your time and the insightful comments you provided during the review process. Your suggestions were immensely helpful in improving the quality of our manuscript.

Reviewer 2 Report

Comments and Suggestions for Authors

I would like to thank the authors for significantly improving the work.

However, not all of the reviewer's comments were addressed:
On page 10, figure 8, the word "Danamomter" was not corrected.

After appropriate corrections, the article will be ready for publication.

Author Response

Comment: On page 10, figure 8, the word "Danamomter" was not corrected.

Response: Thank you for your valuable feedback on our manuscript. Upon reviewing the entire manuscript, we confirm that the word "danamomter" was not used. After consulting the equipment manual and relevant literature, we believe that "dynamometer" is the more appropriate and standard term.

We note that in Figure 8, the misspelled word "Dynamomter" appeared. Therefore, we have corrected it to "Dynamometer" .(page 10, figure 8)

Author Response File: Author Response.pdf