Fracture Mechanism Analysis and Design Optimization of a Wheelset Lifting Mechanism Based on Experiments and Simulations

Round 1

Reviewer 1 Report

In the manuscript, the authors analyze the fracture mechanism of wheelset lifting. A macro examination, scanning electron microscope examination and a metallographic examination were carried out, and then a dynamic stress test in order to accurately diagnose the occurring damage. Next, using FE simulations, it was confirmed that the main cause of the damage was resonance. Optimization of the parameters was proposed to avoid damage. Double-layer Kriging surrogate model and improved NSGA-II were used.

The work proposes a solution of the considered problem that may have practical significance.

Questions:

• How were the optimization calculations performed? Was special software used?
• Did the authors try other optimization algorithms? Why was this one chosen? There is no such comment in the paper.
• Do the authors see the possibility of conducting an experiment for optimized parameters of the analysed wheelset lifting? This would be a good confirmation of the theoretical research carried out.

 

Additionally, the authors should also correct or explain:

• Line 22 - the abbreviation NSGA II appears for the first time and is not explained
• Line 161 - the abbreviation EDX is not explained
• Line 200 - no explanation of AW0 abbreviation
• Line 235 - in the formula there is σ and in the explanations n - please unify it
• Line 299 - it would be good to briefly explain what this algorithm is about (here or in the Introduction)
• Line 319 – ORE B12/RO17 - it would be good to write what it is
• Line 347 – r₃ is not in the formulas (4) and (5)
• Equation (7) – no explanation of p_c
• Equation (8) – no explanation of p_m
• Equations (11) and (12) – no explanation of u_i; in the formula there is Δ_j and in the explanations Δ_i - please unify it
• Equations (13) - no explanation of σ_σm, σ_σa, μ_σm and μ_σa

Finally, in the reviewer’s opinion, the manuscript may be considered for publication in the Machines after revision made by authors in according to the comments raised in the previous paragraphs.

Author Response

Please see the attachment.

Author Response File: Author Response.doc

Reviewer 2 Report

Review Comments on the Manuscript Submitted to “Machines”

Manuscript ID Machines-1716376
Title Fracture mechanism analysis and optimization design of wheelset lifting based on experiment and simulation

In the current study, material tests and dynamic stress experiments are combined with finite element (FE) simulations to reveal the fracture mechanism of the wheelset lifting. A structural optimization design scheme based on the double-layer Kriging surrogate model is proposed. An efficient fatigue robust optimization design method based on the two-layer Kriging surrogate model and improved NSGA-II is proposed to improve the original design scheme. The proposed optimization design method is found to be the most effective method, and the optimized structural scheme is verified by comparative analysis. Some useful conclusions are drawn.

The reviewer recommends the publication of the manuscript based on the abovementioned contributions, but with the following issues for the authors to further elaborate on:

1) since the cracks are under mix-mode loadings, how to judge when a crack will start propagating? What is the crack propagation criterion the current research is using?

2) Is any plastic deformation considered? In elastic-plastic stress (strain) investigations, when there is plastic stress/deformation involved, the simulation results are loading-path dependent. For example: you add tensile loading first then shear loading, or shear loading first then tensile loading, the results will be substantially different. Please clarify this point in your simulation. The authors may consider adding the related references: “On the plastic zone size and crack tip opening displacement of a Dugdale crack interacting with a circular inclusion”, ACTA MECHANICA, Vol. 210, No.
3-4, 305 – 314, 2010.

3) The authors state “An evident difference between the actual and standard stress state is observed for dynamic stress tests and FE simulations that the vibration fatigue occurred at the point of maximum stress is the main cause of the fracture”. What is the reasoning behind this result?

Author Response

Please see the attachment.

Author Response File: Author Response.doc

Reviewer 3 Report

This study intended to improve the structural design of a wheelset lifting part by using the failure analysis, FEA, and optimization method. There FEA needs to be reorganized due to the lack of much important information listed below. The “fracture mechanism” the authors claimed is not clearly presented, instead, they provided some failure analysis. The conclusion from the study seems not persuasive because the stress results in the “optimized” design are not much different from that of using the original design.

Detailed questions and concerns are as follows:

1. Line 28, “the most effective method” is not appropriate.
2. Line 92, the claim “since engineers have few tools to conduct fatigue optimization in an easy way” needs to be revised.
3. The authors mentioned many times “a comprehensive analysis” in the manuscript, e.g., on lines 100, 136, 193, 499, etc. The question is, there are many research and analysis methods that can be used to reveal the fracture mechanism, why the authors claim the “comprehensive analysis” is necessary?
4. Line 102, please delete “To avoid that”
5. Line 122, the claim “which not only analyzes the fracture cause in terms of material science but also determines the type of follow-up experiments and simulations.” is not clear and needs further explanations.
6. For the first paragraph on page 5, does Fig 4 support the fracture mechanism analysis? If not, what is it for?
7. Please add the scale bar to Fig 6.
8. Line 185, “the fracture has the characteristics of bidirectional multi-source”, please mark it in the figure.
9. On line 196, “may be related to” is inappropriate.
10. Line 230, “Recommendations for fatigue design of welded joints and components are employed for damage calculation”, did the authors use the same material as that used in the reference? If not, how can the authors use the same damage rule and S-N curve? If yes, please explain.
11. Line 238, what does “stress damage” mean?
12. For section 4 FE simulation, what material properties were used? And loading and boundary conditions? Yielding/failure criteria? Line 291, please specify what is the “maximum stress”
13. Line 307, “However, L1 , L2 , L4 , L5, and T1 have the greatest impact on its fatigue performance in terms of the mechanism analysis.”, why?
14. In Fig 17 and Fig 18, the stress results of the “optimized design” are not much different from that of the original design, does it mean there is not much improvement after all the optimizations in section 5? The results are not persuasive, please further explain in more detail.
15. Line 500, please delete “micro”, or just include all fracture analysis.
16. Finally, the writing style needs to be fully revised. Both grammar and spelling checks may help.

Author Response

Please see the attachment.

Author Response File: Author Response.doc

Round 2

Reviewer 3 Report

This reviewer's questions have been answered in the revised manuscript.