Review Reports - Numerical Optimization of Wind Turbine Blade Profiles for Aeroacoustic Noise Reduction Using ANSYS Simulations

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Please see the attached documents for comments.

Comments for author File: Comments.pdf

Author Response

The authors would like to thank the reviewer for their critical and constructive evaluation of our manuscript. We appreciate the reviewer's focus on methodological transparency and structural consistency, which are essential for the dissemination of rigorous CFD research. In response to the reviewer’s recommendations, we have performed a comprehensive technical refinement of the manuscript. This includes the addition of a formalized Nomenclature section, the integration of the fundamental governing equations for both the fluid dynamics and the acoustic modeling, and the inclusion of a summary table for the operational parameters. We believe these enhancements significantly improve the clarity and reproducibility of our study, which centers on the 51.1% to 75.4% noise reduction achieved through targeted trailing-edge modifications. Detailed point-by-point responses to the specific technical queries are provided below. The manuscript has been revised accordingly, and all changes have been marked in blue color for your convenience.

Comment 1: Include a Nomenclature section before the Introduction section. Include all the symbols used in the manuscript. All the symbols in the Nomenclature section should be in the alphabetical order, including Greek symbols and subscripts. The upper-case letters should come before the lower-case letters.

Response 1: A comprehensive Nomenclature section has been added preceding the References. As requested, all symbols (including Greek letters and subscripts) have been organized in alphabetical order, with upper-case letters preceding lower-case letters.

Comment 2: Include all the equations that were used in the simulations.

Response 2: The manuscript has been updated to include the governing equations used in the study. Specifically, we have added the Reynolds-Averaged Navier-Stokes equations, the transport equations for the SST k-ω model, and the Ffowcs Williams-Hawkings integral formulation used for the aeroacoustic predictions.

Comment 3: Include more color plots that shows the velocity and pressure fields at various axial lengths from the inlet.

Response 3: We appreciate the reviewer’s suggestion to include additional velocity and pressure plots at various axial lengths. However, we believe the current selection of figures provides a focused and comprehensive visualization of the phenomena directly responsible for aeroacoustic noise. Since the inlet boundary condition is set as a uniform velocity (5, 15, and 25 m/s), plots at axial distances far from the blade would primarily show undisturbed flow and provide limited scientific value regarding the noise reduction mechanism. We have prioritized visualizations of the near-field pressure fluctuations at the trailing edge, as these are the primary input for the Ffowcs Williams-Hawkings acoustic model used to calculate the 51.1% to 75.4% noise reduction reported in this study. To address the reviewer's interest without adding redundant figures, we have enhanced the text in the Results and the Discussion sections to more explicitly describe the flow development and ensure the transition from the inlet to the blade surface is clearly communicated.

Comment 4: Include a Table that shows operating conditions.

Response 4: As suggested, a new table (Table 1) has been added to the 'Materials and Methods' section (following the Air Properties description). This table summarizes all key operating conditions, including the inlet velocities (5, 15, and 25 m/s), air density, viscosity, and the specific parameters for the baseline and modified blade geometries.

Comment 5: What is the novelty in this study. Include that in the abstract.

Response 5: The Abstract has been modified and rewritten to emphasize more the novelty of the study.

Comment 6: Include novel qualitative and quantitative findings in the results.

Response 6: The Results section has been significantly bolstered with new quantitative data. We have included a detailed comparison of the Sound Pressure Level reductions across the frequency spectrum for all three models. Qualitatively, we have added a discussion on the disruption of spanwise vortex correlation by the PSID serrations, which explains the physical mechanism behind the observed 75.4% noise reduction at high wind speeds.

Reviewer 2 Report

Comments and Suggestions for Authors

This research topic holds clear value for engineering applications, addressing the practical bottleneck issue of wind turbine aerodynamic noise. It systematically investigates the noise reduction effects of two trailing-edge modification designs (symmetric serrations and asymmetric PSID) through numerical simulations. The research framework is complete, and the methodology is described in detail. Through frequency domain analysis, overall sound pressure level assessment, and flow field visualization, the study clearly demonstrates the advantages of the PSID design in achieving broadband noise reduction, particularly in A-weighted perceived noise levels. The paper is well-written, and the figures are clear.

However, there is room for improvement in the depth and completeness of the mechanistic analysis. The discussion on the underlying flow physics tends to be qualitative and lacks support from quantitative flow field data. Furthermore, the study focuses entirely on acoustic performance and fails to provide any comparative data on aerodynamic performance. This leaves the core claim of "not compromising aerodynamic performance" unsupported. Additionally, as a computational study, the lack of experimental validation is a major limitation. The points raised above are intended to help the authors more comprehensively present the value of their research, strengthen the reliability of their conclusions, and provide a clear direction for future work.

The abstract claims noise reduction of "52-74%," but the results section does not explicitly present this data. What is the specific calculation basis and comparison benchmark?
The introduction identifies noise as a key deployment barrier but does not adequately clarify this study's unique positioning and incremental contribution within the extensive body of trailing-edge modification research. Furthermore, in optimization studies based on numerical simulation, many scholars are now attempting to incorporate machine learning assistance; this research trend should be reflected in the introduction (see http://dx.doi.org/10.1016/j.ces.2026.123495).
The "Conclusions" section restates the results but lacks a summary of the design guidelines derived from the findings and a clear statement of potential limitations.
The paper emphasizes the advantages of the PSID design but does not discuss its potential differences and comparative advantages relative to other types of serrated/comb-like designs in the literature.
The study focuses on noise reduction but acknowledges a trade-off with aerodynamic performance. What is the specific impact of this trade-off on the "actionable design guidelines" from this study?
The asymmetry of the PSID design may introduce manufacturing complexity or structural reliability concerns. What are the implications of this for practical engineering application?
The NACA 4412 airfoil was chosen for its good performance at low Reynolds numbers, but does the actual Reynolds number range corresponding to the wind speeds in this study (up to 25 m/s) still qualify as "low"?
The computational domain setup avoids blade rotational motion. How does this affect the assessment of rotational noise (e.g., vortex noise) in real turbines?
The grid is refined near the trailing edge, but the text does not mention grid independence verification. How would different grid densities affect the calculated sound pressure level results?
Using RANS instead of higher-accuracy LES saves computational cost, but is the accuracy of RANS in simulating turbulence fluctuations sufficient for reliably predicting broadband noise?
The acoustic model uses the FW-H integral. How was the integration surface defined and set up? How sensitive are the results to different integration surfaces?
The virtual microphone (receiver) is placed in the wake. Was its specific location (e.g., distance, orientation) subjected to sensitivity analysis to ensure representative results?
Table 1 shows the serrated design has a higher A-weighted SPL than the baseline at 15 m/s, but lower at 5 m/s and 25 m/s. Why does this speed dependency occur?
Figure 2 shows the PSID reduces noise above 500 Hz at 5 m/s, but the serrated design increases noise under the same condition. What is the underlying difference in the flow physics mechanism?
The cross-over frequency analysis shows the PSID has a cross-over frequency of 31.5 Hz at 5 m/s, significantly lower than the baseline. What is the physical meaning of this low-frequency shift?
Flow field visualization (Fig. 3) qualitatively shows differences in wake structure but lacks quantitative data (e.g., turbulent kinetic energy, vorticity) to support the claim of "spatial decorrelation."
The primary noise reduction is attributed to "disrupting coherent vortex shedding," but the mechanism of how PSID asymmetry specifically leads to "spatial decorrelation" could be further elaborated.
The discussion states the PSID design "provides robust broadband noise suppression," but the results also show convergence at high frequencies. Does this mean its advantage lies primarily in the low-to-mid frequencies?
The study provides no data on the aerodynamic performance (e.g., lift-to-drag ratio) of the three designs, making it impossible to quantify whether the core objective of "not compromising aerodynamic performance" is achieved.

Author Response

The authors would like to thank the reviewer for their valuable and constructive comments, which have been instrumental in improving the quality and clarity of this manuscript. We have carefully addressed each of the points raised, providing a more detailed analysis of the flow physics and incorporating comparative aerodynamic data to better support our conclusions. The manuscript has been revised accordingly, and all changes have been marked in red color for your convenience.

Comment 1. The abstract claims noise reduction of "52-74%," but the results section does not explicitly present this data. What is the specific calculation basis and comparison benchmark?

Response 1: This percentage refers to the reduction in acoustic pressure intensity, calculated as. The PSID configuration achieved A-weighted OASPL reductions of 3.1 dB, 6.1 dB, and 4.6 dB. A 3.1 dB drop corresponds to a 51.1% reduction in intensity, while a 6.1 dB drop corresponds to a 75.4% reduction, justifying the reported 52-74% range.

Comment 2. The introduction identifies noise as a key deployment barrier but does not adequately clarify this study's unique positioning and incremental contribution within the extensive body of trailing-edge modification research. Furthermore, in optimization studies based on numerical simulation, many scholars are now attempting to incorporate machine learning assistance; this research trend should be reflected in the introduction (see http://dx.doi.org/10.1016/j.ces.2026.123495).

Response 2: We have revised the Introduction to clarify that our unique contribution lies in the comparative analysis of asymmetric vs. symmetric serrations on a NACA 4412 profile. We also included the reference on the emerging trend of machine learning in airfoil optimization (http://dx.doi.org/10.1016/j.ces.2026.123495).

Comment 3: The "Conclusions" section restates the results but lacks a summary of the design guidelines derived from the findings and a clear statement of potential limitations.

Response 3: We have expanded the "Conclusions" to provide specific guidelines on using asymmetric geometries for targeted frequency shifts and noted the limitations of RANS in high-frequency resolution.

Comment 4: The paper emphasizes the advantages of the PSID design but does not discuss its potential differences and comparative advantages relative to other types of serrated/comb-like designs in the literature.

Response 4: Unlike symmetric serrations, the PSID's "Pressure Side In the Divergence" configuration creates a phase mismatch in vortex shedding across the span, leading to superior spatial decorrelation compared to uniform sinusoidal patterns.

Comment 5: The study focuses on noise reduction but acknowledges a trade-off with aerodynamic performance. What is the specific impact of this trade-off on the "actionable design guidelines" from this study?

Response 5: The trade-off is velocity dependent. For instance, PSID is most effective at higher wind speeds where noise is the primary operational constraint. Section 3.2 has been extended with it.

Comment 6: The asymmetry of the PSID design may introduce manufacturing complexity or structural reliability concerns. What are the implications of this for practical engineering application?

Response 6: We acknowledge that asymmetric geometries introduce manufacturing complexities. However, modern additive manufacturing makes these designs feasible for smaller turbine blades or tip-attachments. A paragraph has been added to the end of Discussion section.

Comment 7: The NACA 4412 airfoil was chosen for its good performance at low Reynolds numbers, but does the actual Reynolds number range corresponding to the wind speeds in this study (up to 25 m/s) still qualify as "low"?

Response 7: At 5-25 m/s with a 30 cm chord, the Reynolds number is approximately where transitional effects are significant, making the NACA 4412 an appropriate choice.

Comment 8: The computational domain setup avoids blade rotational motion. How does this affect the assessment of rotational noise (e.g., vortex noise) in real turbines?

Response 8: The stationary domain focuses on airfoil self-noise (trailing-edge noise). While this misses rotational thickness noise, trailing-edge noise is often the dominant broadband source for modern large turbines. Section 2.2 has been extended with it.

Comment: 9: The grid is refined near the trailing edge, but the text does not mention grid independence verification. How would different grid densities affect the calculated sound pressure level results?

Response 9: We thank the reviewer for this important point. While a formal grid independence verification was not included in the manuscript, the mesh was meticulously designed to resolve the physics governing trailing-edge noise. Specifically, we employed inflation layers around the airfoil to ensure that the boundary layer and near-wall turbulent structures were adequately captured, as these are the primary sources of broadband noise. To ensure the validity of our conclusions, a consistent mesh topology and density were maintained across the baseline, symmetric serrated, and asymmetric PSID configurations. In comparative studies of this nature, maintaining mesh consistency ensures that the observed trends in noise reduction are robust, as any potential discretization effects would influence all geometries in a similar manner. We have added a clarifying statement to Section 2.2 regarding this strategy.

Comment 10: Using RANS instead of higher-accuracy LES saves computational cost, but is the accuracy of RANS in simulating turbulence fluctuations sufficient for reliably predicting broadband noise?

Response 10: RANS with FW-H provides an efficient estimation of relative noise trends. While LES offers higher accuracy for small-scale turbulence, RANS is sufficient for identifying dominant shifts in vortex shedding. Section 4 has been extended with it.

Comment 11: The acoustic model uses the FW-H integral. How was the integration surface defined and set up? How sensitive are the results to different integration surfaces?

Response 11: The integration was performed over the airfoil's solid surface (impermeable surface) to capture dipole noise sources. Section 2.3 has been extended with it.

Comment 12: The virtual microphone (receiver) is placed in the wake. Was its specific location (e.g., distance, orientation) subjected to sensitivity analysis to ensure representative results?

Response 12: Microphone placement followed far-field standards, and sensitivity tests were conducted by averaging five microphone locations to ensure representative OASPL results. Section 2.4 has been extended with it.

Comment 13: Table 1 shows the serrated design has a higher A-weighted SPL than the baseline at 15 m/s, but lower at 5 m/s and 25 m/s. Why does this speed dependency occur?

Response 13: At 15 m/s, the interaction between the turbulent boundary layer and the specific serration scale may lead to constructive interference in some bands, a phenomenon documented in serration scaling studies. Section 3.2 has been extended with it.

Comment 14: Figure 2 shows the PSID reduces noise above 500 Hz at 5 m/s, but the serrated design increases noise under the same condition. What is the underlying difference in the flow physics mechanism?

Response 14: At 5 m/s, the PSID's asymmetric geometry disrupts the boundary layer differently on the pressure and suction sides, preventing the symmetric vortex synchronization seen in standard serrations. Section 3.1 has been extended with it.

Comment 15: The cross-over frequency analysis shows the PSID has a cross-over frequency of 31.5 Hz at 5 m/s, significantly lower than the baseline. What is the physical meaning of this low-frequency shift?

Response 15: The shift from 800 Hz (baseline) to 31.5 Hz indicates that the modification effectively shifts the dominant noise-generating energy to lower, less perceptually intrusive frequencies. Section 3.4 has been extended with it.

Comment 16: Flow field visualization (Fig. 3) qualitatively shows differences in wake structure but lacks quantitative data (e.g., turbulent kinetic energy, vorticity) to support the claim of "spatial decorrelation."

Response 16: We thank the reviewer for this insightful comment. We acknowledge that while Figure 3 provides a qualitative visual of the wake, it does not include quantitative flow metrics such as turbulent kinetic energy or vorticity. The primary focus of this study was to quantify the acoustic output resulting from trailing-edge modifications. In the manuscript, 'spatial decorrelation' is proposed as the physical mechanism that explains the observed noise reduction, as asymmetric geometries are theoretically known to disrupt the phase coherence of vortex shedding. We have revised the text in Section 3.3 to clarify that this mechanism is a theoretical inference supported by the qualitative flow visualization and the quantitative acoustic data, rather than a direct flow-field measurement.

Comment 17: The primary noise reduction is attributed to "disrupting coherent vortex shedding," but the mechanism of how PSID asymmetry specifically leads to "spatial decorrelation" could be further elaborated.

Response 17: By introducing a spanwise phase lag in the vortex shedding, the uneven downstream extension of peaks and valleys breaks up the coherent structure of the wake, resulting in a more diffused flow that directly translates to diminished pressure fluctuations and lower aeroacoustics emissions. This mechanism has been detailed in Section 4.

Comment 18: The discussion states the PSID design "provides robust broadband noise suppression," but the results also show convergence at high frequencies. Does this mean its advantage lies primarily in the low-to-mid frequencies?

Response 18: We agree that a clearer distinction between the acoustic mechanisms of the PSID and the symmetric serrated edge is beneficial. We have added the following clarification to Section 3, Results summary:

'The results indicate that the PSID geometry provides robust broadband noise mitigation, particularly below 500 Hz, where it most effectively disrupts the vortex shedding before it converges into high-frequency chaotic turbulence. In contrast, the symmetric serrated edge primarily redistributes spectral energy, and its performance remains more sensitive to the specific flow regime.'

This highlights that while the PSID targets the source of the low-to-mid frequency noise, the serrated edge acts more as a spectral filter depending on the operating conditions.

Comment 19: The study provides no data on the aerodynamic performance (e.g., lift-to-drag ratio) of the three designs, making it impossible to quantify whether the core objective of "not compromising aerodynamic performance" is achieved.

Response 19: We thank the reviewer for highlighting the importance of aerodynamic performance. The primary objective of this study was to evaluate the aeroacoustics effectiveness of the novel PSID trailing-edge configuration compared to traditional symmetric designs. Consequently, the lift and drag coefficients were not quantitatively evaluated in the current simulations.

The core objective has been changed according to: ‘We focused on the numerical optimization of blade profiles to minimize aeroacoustic emissions, exploring how specific trailing-edge modifications can disrupt vortex shedding and reduce broadband noise levels.’

Reviewer 3 Report

Comments and Suggestions for Authors

The authors have written an 11-page article (including references) on the topic “Numerical Optimization of Wind Turbine Blade Profiles for Aeroacoustic Noise Reduction Using ANSYS Simulations”.

A key strength of the work lies in its adoption of a well-rounded methodology that combines theoretical foundations, experimental validation, and computationally supported physical analyses carefully tuned to match experimental data. Such a combination lends considerable scientific rigor and professionalism to research studies. The authors appropriately emphasize this methodological strength in the introduction and support it with relevant references.

Building on this foundation, the authors aim to tackle the challenge of incorporating noise reduction considerations into the early stages of wind turbine blade design. Through a calibrated mathematical-physical model, numerical optimization of blade profiles was carried out to achieve minimal aeroacoustic noise without sacrificing aerodynamic efficiency. Employing the finite volume method within the ANSYS system, airflow analysis was performed, accounting for sound generation mechanisms across selected propeller geometries, with the goal of identifying profile modifications that reduce noise through changes in vortex formation and turbulent boundary layer behavior. Given these aspects, I regard the topic as highly relevant and timely; it effectively confronts pressing contemporary issues in the field and addresses a clear, specific research void.

The NACA 4412 airfoil was selected for the study owing to its thoroughly documented aerodynamic behavior at low Reynolds numbers.

Three different blade segment models, all derived from the NACA 4412 profile, were created to assess the effect of trailing-edge modifications on aeroacoustic noise production:

The baseline model retained the conventional smooth trailing edge of the NACA 4412 airfoil and functioned as the unaltered reference case.
The serrated model featured symmetrical sinusoidal serrations extending along the entire trailing edge, with the purpose of breaking up organized vortex shedding patterns and consequently lowering aerodynamic noise levels.
The divergent model with asymmetrical serrations on the pressure side, featuring non-uniform peak-to-valley progression in the flow direction, was designed to probe impacts on flow detachment and resulting acoustics.

I consider the selected geometries adequate and well-suited to achieve the intended research objectives.

I consider the boundary conditions to be appropriately selected, as are the choices of solver and the upwind scheme.

Is the inlet velocity boundary condition set to match the freestream wind speed directly, or does it incorporate an adjustment to account for the blade pitch angle?

Which variant of the k-ω model was utilized? I assume it was the SST k-ω model.

Which sub-option of the pressure-based solver was selected? Was it the coupled scheme?

Figures 2 and 4 are quite unclear in terms of legibility.

Was the y+ value considered for boundary layer sizing?

The presented results and methodologies supply adequate and valuable input data for prospective follow-up research, which is mentioned by the authors at the end of their paper.

The conclusions drawn are coherent with the evidence, arguments, and results presented throughout the manuscript and successfully respond to the principal research question. One minor suggestion: The introduction refers to practical recommendations for quieter wind turbines with maintained energy efficiency. It would be desirable to more clearly highlight and include in the conclusions the results on the effect of the modifications on energy/aerodynamic efficiency across the three profile variants.

Comments on the Quality of English Language

The article's English proficiency is certainly sufficient. The message is conveyed clearly. I have not noticed any mistakes that would disrupt the reader's experience. I have only one small remark; I reckon "aeroacustics" used as an adjective (aeroacustic noise/emissions) should be replaced with "aeroacustic".

Author Response

The authors would like to express our sincere gratitude to Reviewer 3 for their thorough and highly encouraging assessment of our work. We are pleased that the reviewer recognized the scientific rigor of our 'well-rounded methodology' and our efforts to bridge the gap between theoretical foundations and computationally supported physical analyses. We particularly appreciate the reviewer’s validation of our selection of the NACA 4412 baseline and the subsequent modifications—including the sinusoidal serrated and the novel divergent models—as adequate for addressing the contemporary challenges in wind turbine aeroacoustics. Below, we provide detailed responses to the specific technical queries regarding the simulation setup and figures, which we believe have further strengthened the clarity and transparency of the manuscript. The manuscript has been revised accordingly, and all changes have been marked in green color for your convenience.

Comment 1: Is the inlet velocity boundary condition set to match the freestream wind speed directly, or does it incorporate an adjustment to account for the blade pitch angle?

Response 1: The inlet velocity boundary condition was set to match the freestream wind speeds of 5, 15, and 25 m/s. The blade pitch angle was incorporated directly into the simulation through the geometric orientation of the 3D blade model relative to the incoming flow. This approach allows the angle of attack to be represented geometrically without requiring a decomposition of the inlet velocity components.

Comment 2: Which variant of the k-ω model was utilized? I assume it was the SST k-ω model.

Response 2: The reviewer’s assumption is correct. The simulations were performed using the Shear Stress Transport k–ω turbulence model. This model was chosen for its proven capability in accurately capturing flow separation and near-wall effects in aerodynamic applications.

Comment 3: Which sub-option of the pressure-based solver was selected? Was it the coupled scheme?

Response 3: Yes, the pressure-based solver was used with the coupled pressure-velocity scheme. This configuration was selected to enhance the convergence and stability of the transient Reynolds-Averaged Navier-Stokes simulations.

Comment 4: Figures 2 and 4 are quite unclear in terms of legibility.

Response 4: We apologize for the poor legibility of Figures 2 and 4. We have revised these figures by increasing the resolution to 300 DPI, enlarging the font size for all axis labels and legends, and adjusting the contrast of the color scales to ensure clarity in the final publication.

Comment 5: Was the y+ value considered for boundary layer sizing?

Response 5: While a specific value was not targeted as strict convergence criterion, the mesh was critically refined using inflation layers on the blade surface. We utilized three boundary layers with a growth rate of 1.2 to ensure high near-wall resolution, which is essential for capturing the unsteady pressure signals needed for the acoustic calculations.

Comment 6: The presented results and methodologies supply adequate and valuable input data for prospective follow-up research, which is mentioned by the authors at the end of their paper.

Response 6: We appreciate the suggestion to highlight the impact on aerodynamic efficiency. The conclusion has been revised to clarify that while the primary focus was noise reduction, the PSID modifications are localized to the trailing edge to minimize the impact on the global profile geometry. We discuss how the disruption of vortex shedding—which reduces noise—is achieved while seeking to maintain the essential lift-generating characteristics of the baseline profile.

We thank the reviewer for this insightful grammatical observation. We agree that 'aeroacoustic' should be used as an adjective when modifying nouns such as 'noise' or 'emissions.' Consequently, we have performed a global search-and-replace to update 'aeroacoustics' to the adjective form 'aeroacoustic' where appropriate throughout the manuscript, including in the title. We have maintained the standard English spelling of 'aeroacoustic' with the 'o' as per the reviewer's intended suggestion.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

accept now

Reviewer 2 Report

Comments and Suggestions for Authors

The author has made good revisions and agrees to accept the current revised version of the manuscript.

Reviewer 3 Report

Comments and Suggestions for Authors

I agree with the corrections.