Aerodynamic Characteristics Analysis of Rectifier Drum of High-Speed Train Environmental Monitoring Devices

Round 1

Reviewer 1 Report

In my opinion Forces should be given non-dimensional form.

Author Response

Thanks for your professional advice. The aerodynamic forces and moments have been nondimensionalized based on the dynamic pressure of free stream and feature size of the devices (e.g. aera and length of the base). The coefficients are defined as follows:

where ρ=1.225kg·m³, V=110m/s, S_c=0.149m² and L_c=0.5m.

Then results in Table 2, as well as some analysis in the text, have been replaced with aerodynamic coefficients.

Date of this reply

26 May 2023

Author Response File: Author Response.docx

Reviewer 2 Report

This study focuses on investigating the aerodynamic characteristics of three types of rectifier drums attached to high-speed trains using CFD simulations. The numerical model employs three-dimensional NS equations for the compressible flow over the train for subsonic flow. The IDDES turbulence model was combined with a transition model to capture large separations from the vehicle. While authors mostly focus on the turbulence model in the paper, details of the numerical model such as numerical schemes for convective and diffusive fluxes and shock-capturing schemes were not discussed. Details of the code or software used for the numerical approximations of the governing equations were not given. Another important concern is the validation of the numerical model.

Typos, grammatical errors, and poor language used in the paper make it difficult to read the paper overall.

Author Response

Thanks for your thorough review and professional advice. The replies have been conducted as follows in two parts:

1. 1. A brief introduction of numerical schemes for convective and diffusive fluxes used in present paper has been supplemented in section 2.4. We adopted the Bounded Central Differencing (BCD) scheme to discretize the flux terms which using a blended coefficient φtoadjust the numerical accuracy and stability. In present simulation, the coefficient φ is set to 0.15 to reduce numerical dissipation.

2. 2. A case of square cylinder unsteady flow has been adopted for validation, seen in Section 2.7. The computational setup, wake flow of Q iso-surface, and time-averaged results along centerline have been described in the manuscript.

3. 3. We have double checked the manuscript and some mistakes have been corrected, some English expressions have also been improved.

Date of this reply

26 May 2023

Author Response File: Author Response.docx

Reviewer 3 Report

The topic addressed in the article is of scientific and pertinent interest. The submitted manuscript is well organized. I have positive opinion about the publication of the article; however, some correction must be done, and others clarified.

1.    Some small mistakes must be corrected.

2.    The manuscript must be reviewed and the English improved.

2.    In my opinion, a nomenclature should be added at the beginning of the document. Some of the equations are well known and do not need to appear in the text. If included the meaning of each of the components or variables should also be included.

3.     The meaning of acronyms must be mentioned in the text as soon as they appear for the first time. For example, EMU or IDDES.

4.        Please clarify / correct the legend of Figure 9 (a) and (b). Seems there is a mistake!

5.        The numerical results make sense physically. However, a very important step of the modeling process is its validation by comparison with experimental results. In my opinion, a weakness of the study is that it is purely numerical. The authors do not confront numerical results with any experimental measurements. Why you didn't submit at least the model of one of the device's configurations to experimental tests in a wind tunnel.

1.    Some small mistakes must be corrected.

2.    The manuscript must be reviewed and the English improved.

Author Response

Authors’ Reply

Thanks for your thorough review and professional advice. The replies have been conducted as follows:

1. We have double checked the manuscript and some mistakes have been corrected, some English expressions have also been improved.
2. 2. The nomenclature seems not to be inconsistent with the template of MDPI. Though we supplement detailed explanations for all variables that first appear in the text.
3. 3. All acronyms first mentioned in the text have been explained.
4. 4. The mistakes of the legend in Figure 9 (Figure 12 in revised version) has been corrected. A similar mistake has also been corrected in Figure 14 (Figure 17 in revised version).
5. 5. Unfortunately, no experimental measurements have been done in a wind tunnel yet. We have provided a case of unsteady wake flow behind a square cylinder for validation of the numerical models.

Numerical Validation

The case of the square cylinder flow at high Reynolds number was adopted as an validation for the numerical method. The Reynolds number is 2.2×10⁴ based on the side length of the square, and the Mach number is 0.2. A set of grid contains 2.8 million cells, which is refined near the wall to ensure the y⁺≈1 for the first grid spacing. The physical time step is 0.005s and the Courant number during the inner iteration is 10. Figure 3 illustrates the computational field and the grid distribution.

Figure 3. Computational field (left) and grid distribution (right) of the square cylinder flow

The Q criterion is commonly used to depicted the coherent structures of the intricate wake flow. The detailed definition has been given in reference [24]. Figure 4 gives the Q iso-surface distribution with non-dimensional Q=0.45. The large-scale, three-dimensional vortex can be seen shedding alternately in the wake. The quantitative results have been provided and compared with the experiment results in Figure 5, including the time-averaged streamwise velocity and the Reynolds stress. The simulation results agree well with the experimental ones, concluded that the numerical model of IDDES gives higher numerical accuracy comparing to the RANS model.

Figure 4. Q isosurface of the square cylinder flow

Comments on the Quality of English Language

1. Some small mistakes must be corrected.

Answer：We have double checked the manuscript and some mistakes have been corrected

2. The manuscript must be reviewed and the English improved.

Answer：We have submited the revised manuscript to the institution recommended by the MPDI for English improved.

Author Response File: Author Response.docx