Numerical Study on the Power Efficiency and Flow Characteristics of a New Type of Wind Energy Collection Device

Round 1

Reviewer 1 Report

The paper explores the performance of the Invelox wind turbine system and proposes an improved design, which provides a speed ratio increase near 42% at the Venturi tube region. To that aim, RANS numerical simulations of the problem are conducted. The topic constitutes a problem of interest and is worth of study. Nevertheless, there are still some issues to address before it is recommended for publication.

Main comments:

1) The analysis of the flow around the Invelox should be conduced under realistic conditions. In this regard, the consideration of a uniform velocity profile is not appropriate in the case at hand, given the size of the system. Thus, the study should be conducted employing only realistic velocity profiles and the conclusions should be based on the results obtained from those computations.

2) From fig. 16, the simulations performed using the ABL velocity profile consider the system placed 12 m above the ground floor. Again, these simulations should be conducted considering the system placed at the ground, where the velocity affecting the outlet region is lower.

3) The authors propose a modification of the system that affects both the size and structural loads acting on it. Please, clearly explain such modifications, the change in size, the total height of the structure, the structural modifications needed to cope with the new design, etc, as well as their implications.

Also, explain the possible (if any) drawbacks of the new design.

Minor comments:

4) Equations 1 and 2 should be those of an incompressible flow, please modify these eqs. accordingly.

5) Transient simulations should be conduced, later presenting averaged results. In that regard, PISO algorithm should be used.

6) Please, use the same contour levels in figs. 8 to 14 for the sake of clarity.

7) From contours in fig. 10, it would seem that the grid resolution yields some spurious results. Please, improve the quality of the mesh where necessary. Furthermore, a more extensive and comprehensive analysis of the flow field (pressure and velocity) should be conduced, as well as of the flow modifications induced by the installation of the shield, e.g. flow detachment, discharge pressure, among others.

8) Please, use different line styles to better identify each one.

9) Please, revise the text, since a few typos have been detected. For example:

“sine”, line 75.

“…values for this study is given…”, line 204.

Author Response

Dear reviewer:

We are very grateful to your comments for the manuscript. All your suggestions are very important and have important guiding significance for our thesis writing and scientific research work. According with your advice, we amended the relevant part in manuscript. All your questions were answered. Details please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

1- Introduction are not complete and many important papers in the field are missed. For instance you can see: Study of Horizontal Axis Tidal Turbines (HATT) Performance and Investigation on the Optimum Fixed Pitch Angle Using CFD: Case study of Iran, International Journal of Numerical Method for Heat and Fluid Flow, 2019, DOI (10.1108/HFF-05-2019-0447)

2- There are different places that sentences are misleading, for instance " the conventional energy on the earth is drying up " which is not true at least not for the current century. Or, they mentioned "To solve the problem of yaw and high starting wind speed of horizontal axis wind turbine, a wind turbine system called Invelox is developed ", is this the only reason for development of such wins turbines.

3- Equations 1 and 2 are wrong! Either write the NS equation and explain the RANS averaging, or write the whole equation in RANS averaged form.

4- Why use the RANS and more specifically k-omega. First it is shown in "Direct Numerical Simulation of flow instabilities over Savonius style wind turbine blades, Renewable Energy 2017 (105) 374--385." that  RANS prediction for lift and drag coefficient can be underestimated by a factor of 2. Second, k-omega model is not that much reliable in such problems.

5- According to authors the flow is fully turbulent, but I don't see the boundary layer mesh in figure 4 which are inevitable in the simulation of turbulent flow.

6- What are the y+ and u_tau for your simulations?

7-"To ensure the convergence, the residuals for all solved equations were set as 10^-5". What convergence you are talking about?  CFD 101, residual converged, doesn't mean results are converged!

8- Average velocity is a very loose parameter to study the convergence. Please also compare other relevant parameters that are more sensitive to the mesh resolution.

9- more physical discussions are needed for the results.

Author Response

Respected reviewer:

We are very grateful to your comments for the manuscript. All your suggestions are very important and have important guiding significance for our thesis writing and scientific research work. According with your advice, we amended the relevant part in manuscript. All your questions were answered. Details please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Authors significantly, improve the quality of manuscript. there are 2 points that are not well considered. so, I suggest a minor modification in

1- equations 1 and 2 in which you can find typo. for instance:S_i should S_ij, or \bar{u')=0 therefore can not appear in formula:

2- Authors mentioned Y+ in the response letter. I think I t is good to present for some of the cases in the manuscript.