Using Numerical Analysis to Design and Optimize River Hydrokinetic Turbines’ Capacity Factor to Address Seasonal Velocity Variations

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The novelty of the approach is notable due to the interchangeable rotor design, but it could be better emphasized:

 

While the paper explains the advantages of two-blade rotors and highlights their adaptability, there is limited comparison to previous rotor designs and cost-effectiveness benchmarks.

The paper should address dynamic performance under real-world conditions, such as river debris and turbulence, which are not sufficiently covered.

 

The paper reviews a broad set of studies related to hydrokinetic turbines and CFD modeling but it needs to include some recent advancements in Analysis and optimization techniques. such as:

" Numerical analysis of topology-optimized cold plates for thermal management of battery packs"

" Numerical and experimental study of a wearable exo-glove for telerehabilitation application using shape memory alloy actuators"

 

 

The methodology is well-structured, integrating CFD simulations and FEA stress analyses, but it requires additional validations and robustness tests:

 

Strengths:

 

Detailed rotor design process (page 5, Figure 1) and optimization methods using BladeGen and ANSYS CFX​.

Clear breakdown of computational domains and mesh independence studies (pages 10–12, Figures 5 and 6)​.

Limitations:

 

Parameter Sensitivity Analysis: The effects of parameter variations (e.g., blade angles, material properties) on performance are not fully explored.

 

Scalability Tests: Tests are limited to static flow conditions, with no evaluations under dynamic river fluctuations or obstructions.

Author Response

The novelty of the approach is notable due to the interchangeable rotor design, but it could be better emphasized:

Response 1: Thank you for pointing this out. The novelty of study revised in conclusion section in following:

“The novelty of this approach lies in several key improvements. First, the analysis shows how to achieve efficiencies of 43% to 45% using CFD. This is achieved through optimization of blade geometry and aerodynamic parameters, ensuring maximum energy capture under varying flow conditions. Second, the system can effectively almost double the CF compared to existing RHKT systems that no other approach can compare. This improvement in CF is critical for enhancing energy production in low-flow conditions, where exciting RHKT systems are underperforming in winter. Additionally, the design features a 2-blade rotor, which reduces both costs and complexity during deployment and retrieval. By removing the need for a shroud and gearbox, the system becomes lighter, more reliable, and has fewer components, allowing for easier access and servicing with boats available in remote communities. With three rotors designed, the system allows rotors to exchange, depending on the seasonal water velocity. This design is especially beneficial for remote communities, where deployment and maintenance can be managed using local boats, eliminating the need for specialized technicians. The approach is more suited for smaller turbines”.

This change can be found in the Conclusions on page 29, paragraph 2, and line 722 of the revised manuscript.

While the paper explains the advantages of two-blade rotors and highlights their adaptability, there is limited comparison to previous rotor designs and cost-effectiveness benchmarks.

Response 2: Thank you for pointing this out. There are no real-world two-bladed river hydrokinetic turbines available for comparison, which highlights the novelty of this study. However, existing turbines are three-bladed, and Smart Hydro is one of the few companies that manufacture horizontal river hydrokinetic turbines (RHKT). We compared our method (two-bladed) with the real-world performance of three-bladed RHKT.

The paper should address dynamic performance under real-world conditions, such as river debris and turbulence, which are not sufficiently covered.

 Response 2: Thank you for pointing this out. By eliminating the shroud, which is one of the existing turbine issues, we aim to address debris. For turbulence, we have considered turbulence intensity, as mentioned in section 2.3.1, Preprocessing.

The paper reviews a broad set of studies related to hydrokinetic turbines and CFD modeling but it needs to include some recent advancements in Analysis and optimization techniques. such as:

"Numerical analysis of topology-optimized cold plates for thermal management of battery packs"

"Numerical and experimental study of a wearable exo-glove for telerehabilitation application using shape memory alloy actuators"

Response 3: Thank you for your suggestions. While the two approaches you mentioned are valuable, they are not directly relevant to our study of river hydrokinetic turbines. Our focus is on fluid dynamics and rotor optimization for energy conversion, which differs from thermal systems and actuation.

There are various methodological approaches to turbine design, but the key objective is achieving a high-efficiency turbine. For the first time, we have achieved 45% efficiency, marking a significant milestone. This represents the true novelty of our work, as it enables the manufacturing of a highly efficient turbine, which are not suitable for practical manufacturing applications.

 The methodology is well-structured, integrating CFD simulations and FEA stress analyses, but it requires additional validations and robustness tests:

Response 3: For validation of our results, we compare numerical results with experimental data

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript presents a novel and practical approach for improving the performance of river hydrokinetic turbines by developing and optimizing a system of interchangeable rotors to adapt to seasonal river velocity variations. The study is well-conceived and addresses a significant challenge in the field of renewable energy, particularly for remote communities that rely on consistent energy production. The following detailed comments and suggestions are provided to help improve the quality of the manuscript:

1. If available, the authors need to mention the specific seasonal improvement in the capacity factor or the relative increase compared to a fixed-rotor design in the abstract.
2. In the introduction, the authors mentioned the problems of low capacity factor and high costs in RHKT. While these are significant issues, the introduction would benefit from a clearer articulation of how seasonal rotor swapping addresses these gaps compared to existing solutions. Therefore, the authors could consider briefly comparing your proposed method with existing techniques aimed at improving CF (e.g., variable pitch control, power electronics) and highlighting why their approach is more feasible or cost-effective.
3. The authors mentioned Smart Hydro's example, which adds valuable context. However, the statement "even if our team modified this turbine to successfully operate in winter conditions" lacks sufficient detail. Therefore, the authors need to briefly explain what modifications were made and how they improved the turbine's winter performance.
4. The concept of rotor swapping is core to the study but requires further explanation. It is unclear whether the swapping process is entirely manual or could be semi-automated and how often this swapping would occur in practice (e.g., seasonally or more frequently). Therefore, the authors need to provide more details on the operational feasibility of rotor swapping, including how long it takes, the required manpower, and any potential logistical challenges.
5. The authors mention BEM, vortex wake models, and CFD in the introduction section; however, the rationale for selecting CFD is not sufficiently detailed. Therefore, the authors need to add a sentence explaining why CFD is preferred in this study.
6. The authors need to add a brief sentence explaining why these specific values (96% and 97%) for efficiencies (ηG and ηM) were selected or referencing similar studies where such assumptions were made.
7. The authors mentioned that the rotor spins counterclockwise. While this detail is important, the reason for choosing this rotational direction is not explained. Therefore, the authors need to clarify whether this design choice was made based on the testing setup, fluid flow characteristics, or other factors.
8. The authors discussed the vortices and pressure fluctuations on the blade surface, which is insightful. However, this section lacks quantitative validation or reference to specific turbulence models that capture these effects. Therefore, the authors need to mention whether additional turbulence models (such as LES or DES) were considered for capturing vortex dynamics or explain why the SST model was deemed sufficient.
9. The authors need to briefly explain why five segments were chosen (e.g., to balance computational efficiency with design accuracy) and how the shape of each segment is optimized using the velocity triangle method.
10. The authors mentioned that RHKT does not have a standardized airfoil topology. This is an important point, but it would be helpful to clarify how the airfoil profiles were selected or customized for this study.
11. The authors mentioned that the blockage ratio is critical. The chosen range of 0.03 to 0.05 should be justified with references or prior research. Hence, the authors need to explain why this range was selected and how it impacts the simulation accuracy.
12. The authors need to briefly mention the rationale for choosing a transient rather than a steady-state model and emphasize the importance of capturing unsteady effects in hydrokinetic turbines.
13. The authors describe the boundary conditions (inlet velocity, turbulence intensity, and outlet pressure) well. However, specifying the freestream velocity range and how a turbulence intensity of 5% was determined (e.g., based on river flow conditions) would improve clarity.
14. Although the SST model is widely used, a brief justification of its selection over other models (e.g., k-ε or k-ω) would improve the section's depth.
15. In the solver section, the authors need to consider briefly mentioning whether the simulation results will be validated with experimental data or prior literature results to strengthen confidence in the solver's accuracy.
16. The text of FEA Mechanical Analysis (Section 2.4) lacks details on the specific load cases considered in the FEA simulations (e.g., static loads, dynamic loads, or varying flow conditions). Hence, the authors need to Include this information.
17. The authors need to consider elaborating on why PEEK composites were selected over other common turbine materials, such as aluminum or standard polymer composites.
18. In the results and discussion section, the authors present initial efficiencies for the three rotors (11.5%, 13.6%, and 13.3%), followed by a detailed description of the optimization process. However, the initial efficiencies seem low compared to typical hydrokinetic turbines. While this emphasizes the importance of optimization, explaining why the initial design yielded such low values would be beneficial. Therefore, the authors need to briefly explain factors contributing to the low initial efficiency, such as suboptimal blade angles or pressure distributions.
19. The reported efficiencies of 45.10%, 43.27%, and 43.42% after optimization are commendable and within a reasonable range for hydrokinetic turbines. However, the authors need to discuss whether these efficiencies meet or exceed those of similar designs in the literature. This would help place the results in context and demonstrate the novelty of the proposed design.
20. The authors clearly explained the limitations of a single rotor in maintaining consistent energy production, especially during low-velocity periods. The sharp drop in energy delivery during winter months (from 4,423 kWh to 680 kWh) illustrates the problem. While the analysis highlights the seasonal energy drop, mentioning the specific period used for the calculation (e.g., monthly or quarterly) would be helpful to clarify how the energy output was determined.
21. While the authors highlight key results in the conclusion section, it would be beneficial to explicitly emphasize the novelty of the proposed approach compared to existing RHKT systems.

 

 

Author Response

Comments 1: If available, the authors need to mention the specific seasonal improvement in the capacity factor or the relative increase compared to a fixed-rotor design in the abstract.

Response 1: Thank you for pointing this out. I agree with this comment. Therefore, I have added the following sentence in the abstract to clarify the seasonal improvement in the capacity factor: “Results show that rotor interchangeability significantly enhances turbine capacity factor, increasing it from 52% to 92% by adapting to river seasonal velocity changes”. This change can be found in the abstract on page 1, paragraph 1, and line 23 of the revised manuscript.

Comments 2: In the introduction, the authors mentioned the problems of low capacity factor and high costs in RHKT. While these are significant issues, the introduction would benefit from a clearer articulation of how seasonal rotor swapping addresses these gaps compared to existing solutions. Therefore, the authors could consider briefly comparing your proposed method with existing techniques aimed at improving CF (e.g., variable pitch control, power electronics) and highlighting why their approach is more feasible or cost-effective.

Response 2: Thank you for pointing this out. We address your comment in the following subsections.

To clarify: The second paragraph of the introduction highlights the issue with the RHKT by referencing concerns raised by other researchers and industries. In the subsequent pages, we detail our approach to addressing low velocity and cost concerns.

For instance, in the Capacity Factor (CF) section, we elaborate on the low-velocity issue, as outlined following:
"Most RHKT manufacturers rate turbines based on performance at approximately 3 m/s flow velocity, which is high for most rivers and can decrease significantly seasonally. Since power output scales with the cube of flow velocity, this rating can mislead expectations. For example, a turbine rated at 5 kW in a 3 m/s flow would produce only 1.5 kW in a 2 m/s flow and 0.18 kW in a 1 m/s flow [3]. Thereby seasonal velocity variation has a strong impact on the CF. For small RHKT, it is possible to seasonally change the rotor diameter to better match the generator size."

We also highlighted how our approach is cost-effective in the following:

"The study presents three key contributions:

1. Enhancing CF and simplifying system design: Improving RHKT's low- CF by using seasonally adjustable rotor sizes, eliminating shrouds, and incorporating two-bladed rotors to simplify near-shore deployment to lower costs. These changes reduce installation and maintenance costs, boost microgrid revenues, and simplify deployment—elimination of the gearbox and having RHKT designs have their rotor be easily changed to adapt to seasonal velocity variations improves the economics."

Overall, by the end of the introduction, we provide a brief explanation of the current issue and a comparison between the current and proposed approaches for the reader.

The improvements do not exclude any other improvements or approaches.  Its just that no other improvement can increase the CF as much when there are seasonal flow variations.

Comments 3: The authors mentioned Smart Hydro's example, which adds valuable context. However, the statement "even if our team modified this turbine to successfully operate in winter conditions" lacks sufficient detail. Therefore, the authors need to briefly explain what modifications were made and how they improved the turbine's winter performance.

Response 3: Thank you for pointing this out. What we mean is that our approach can address the issues faced by Smart Hydro in winter conditions. The explanation of how this is achieved is detailed in the following pages, where we discuss capacity factor (CF), cost, and weight.

Comments 4: The concept of rotor swapping is core to the study but requires further explanation. It is unclear whether the swapping process is entirely manual or could be semi-automated and how often this swapping would occur in practice (e.g., seasonally or more frequently). Therefore, the authors need to provide more details on the operational feasibility of rotor swapping, including how long it takes, the required manpower, and any potential logistical challenges.

Response 4: Thank you for pointing this out. We agree with this comment. Therefore, I have replaced the following sentence in conclusion section: “A 2-blade rotor was used to lower cost and simplify deployment, retrieval and interchanging rotors” with “Additionally, the design features a 2-blade rotor, which reduces both costs and complexity during deployment and retrieval. By removing the need for a shroud and gearbox, the system becomes lighter, more reliable, and has fewer components, allowing for easier access and servicing with boats available in remote communities. With three rotors designed, the system allows rotors to exchange, depending on the seasonal water velocity. This design is especially beneficial for remote communities, where deployment and maintenance can be managed using local boats, eliminating the need for specialized technicians. The approach is more suited for smaller turbines.” This change can be found in the conclusions on page 29, paragraph 2, and line 728 of the revised manuscript.

It would never be done semiautomatically.  The turbines needs to be designed for easy rotor swapping.

Comments 5: The authors mention BEM, vortex wake models, and CFD in the introduction section; however, the rationale for selecting CFD is not sufficiently detailed. Therefore, the authors need to add a sentence explaining why CFD is preferred in this study.

Response 5: Thank you for pointing this out. We agree with this comment. Therefore, we have revised the following sentence for clarification: “Versteeg and Malalasekera [17] highlighted significant advancements in Computational Fluid Dynamics (CFD), noting its growing importance for designing fluid systems in aerospace, automotive, and energy industries. Despite costs, CFD's ability to shorten lead times, simulate complex scenarios, and offer detailed insights has made it indispensable. CFD offers superior accuracy in capturing complex flow dynamics, including turbulence, unsteady effects, and fluid-structure interactions, which are critical for hydrokinetic turbine design. Unlike BEM and vortex wake models, CFD can simulate the intricate behavior of the fluid flow around the rotor, especially under varying seasonal conditions and non-ideal flow scenarios.” This change can be found on page 3, paragraph 3, and line 125 of the revised manuscript.

Comments 6: The authors need to add a brief sentence explaining why these specific values (96% and 97%) for efficiencies (ηG and ηM) were selected or referencing similar studies where such assumptions were made.

Response 6: Thank you for pointing this out. I agree with this comment. Consequently, I have included the reference for ηG and ηM.

Comments 7: The authors mentioned that the rotor spins counterclockwise. While this detail is important, the reason for choosing this rotational direction is not explained. Therefore, the authors need to clarify whether this design choice was made based on the testing setup, fluid flow characteristics, or other factors.

Response 7: Thank you for pointing this out. We respectfully disagree with this comment. The direction in which the rotor spins, whether clockwise or counterclockwise, is a design choice and does not significantly affect energy capture. The primary consideration is that when generating geometry in BladeGen, it is important to follow a consistent direction, either left to right or right to left. This direction should also be accounted for when defining boundary conditions for structural analysis. The testing setup is not relevant in this context.

Comments 8: The authors discussed the vortices and pressure fluctuations on the blade surface, which is insightful. However, this section lacks quantitative validation or reference to specific turbulence models that capture these effects. Therefore, the authors need to mention whether additional turbulence models (such as LES or DES) were considered for capturing vortex dynamics or explain why the SST model was deemed sufficient.

Response 8: Thank you for your comment. We agree that turbulence modeling is critical for accurately capturing vortex dynamics and pressure fluctuations on the blade surface. In this study, the SST k-ω model was selected due to its proven capability to effectively capture the boundary layer behavior and shear stress transport in rotor flows, which are key for the analysis of hydrokinetic turbines. The SST model has shown reliable performance in simulating turbulence effects, particularly in the near-wall regions and areas with adverse pressure gradients, which are important for our application.

While Large Eddy Simulation (LES) or Detached Eddy Simulation (DES) models offer higher fidelity in capturing turbulence at finer scales, they were not employed in this study due to the significantly higher computational cost and longer simulation times required. For our design scope and the computational resources available, the SST model provided a sufficient balance between accuracy and efficiency. We have updated the manuscript to clarify this rationale and the suitability of the SST model for capturing vortex dynamics in hydrokinetic turbine design. We have updated the manuscript to clarify this rationale and the suitability of the SST model.  “The Shear-Stress Transport (SST) model is applied in the CFD simulation. This step captures critical factors such as flow variations, wake interactions, and turbulence effects that can significantly impact rotor performance. Given the scope and objectives of this study, the SST model is sufficient to accurately simulate the key phenomena affecting hydrokinetic turbine design. Advanced turbulence models like Large Eddy Simulation (LES) or Detached Eddy Simulation (DES) are typically used for capturing turbulence at finer scales or in highly complex flow scenarios. However, for the current application, the SST model provides an effective balance between accuracy and computational efficiency, as it can resolve the primary flow features without the need for the additional computational cost associated with LES or DES. Once the solver has completed the simulation, the results undergo post-processing and validation to ensure the accuracy of the analysis and meet the design goals.” This change can be found on page 4, paragraph 2, and line 182 of the revised manuscript.

Comments 9: The authors need to briefly explain why five segments were chosen (e.g., to balance computational efficiency with design accuracy) and how the shape of each segment is optimized using the velocity triangle method.

Response 9: Thank you for pointing this out. We respectfully disagree with the number of sections in CFD results efficiency.

In BEM, the blade is divided into discrete sections along its span, with aerodynamic forces calculated at each section based on local flow conditions. The accuracy of BEM depends on the number of sections used to approximate overall blade performance, But in CFD, the entire blade geometry is simulated, with flow interactions resolved through mesh quality and fluid dynamics equations. The number of sections in BladeGen (CAD) is primarily used to define the blade's shape and design intent. Once CFD begins, mesh resolution and turbulence models become crucial for accuracy. Adding more sections in BladeGen does not inherently improve CFD results unless they capture geometric variations.

Regarding the use of sections, our methodology includes two phases. First, we employ a calculation approach to start with BladeGen, followed by CFD for numerical analysis and optimization. BladeGen requires initial data, provided by velocity triangle calculations, to create the initial geometry, though this often results in lower efficiency. This approach of using sections differs from the BEM methodology. It only serves to define the first and last angles of each layer, which is necessary for CAD software (BladeGen).

In regard to your comment about how the shape of each segment is optimized using the velocity triangle method: We only use the velocity triangle method to calculate the initial data for BladeGen to define the starting geometry. However, this method is not used for optimization in subsequent steps.

In our design, we used 50 points divided into five layers, each containing 10 points. Table 3 provides details for only the first and last points of each layer. By manually adjusting these 50 points, the chord length and angle were modified simultaneously in BladeGen. During the optimization phase, each rotor underwent over 700 iterations of geometry refinement and CFD steady-state simulations to achieve optimal efficiency. After achieving maximum efficiency, we conducted transient simulations to ensure the results were consistent and reliable.

For further clarification, I have added the following to the manuscript: “In this study, the number of sections used in BladeGen does not affect the accuracy of the CFD results. In BEM method, forces and torques are calculated at discrete sections based on local flow conditions, and the accuracy depends on the number of sections used to approximate the overall blade performance. In contrast, CFD simulates the entire blade geometry and resolves flow interactions based on mesh quality and fluid dynamics equations. The sections in BladeGen are primarily used to define the blade’s shape and design intent, while the mesh resolution and turbulence models play a critical role in determining CFD accuracy.

To initiate the design process in BladeGen, preliminary calculations using the velocity triangle method were performed to approximate the first and last points of each layer, ensuring the desired aerodynamic performance. These initial values, as presented in Table 3, served as the starting point but were not optimized.

During the optimization phase, the design was refined using 50 points distributed across five layers, with 10 points per layer. Each point within a layer was iteratively adjusted throughout the optimization process. This phase involved over 700 iterations of geometry regeneration and CFD simulations, ultimately resulting in an optimized blade design with improved performance and efficiency.” This change can be found on page 10, paragraph 1, and line 323 to 330, and paragraph 2, line 336 to 345 of the revised manuscript.

 

Comments 10: The authors mentioned that RHKT does not have a standardized airfoil topology. This is an important point, but it would be helpful to clarify how the airfoil profiles were selected or customized for this study.

Response 10: Thank you for pointing this out. We agree with your comment, but as explained in the previous response and outlined in the methodology, we calculated the first and last points of each layer using the velocity triangle method. Additionally, it's important to understand how BladeGen operates. The developers of this software focused on enhancing optimization capabilities, which makes our study somewhat different from most of the literature I've reviewed. Many studies focus on simulations, using a specific airfoil section for CFD analysis and comparing it with other airfoil shapes or the behavior around some standard airfoil. The goal of our study, however, is to achieve maximum efficiency, rather than determining which standard NACA airfoil is better. Our primary focus is on minimizing aerodynamic losses to develop an optimized blade design. I have added more explanation for the reader in the previous comment: “Unlike wind turbines, RHKT do not have a standardized airfoil topology due to the highly site-specific nature of their design, influenced by factors such as flow velocity and depth. In ANSYS BladeGen, the blade Beta angle is a key parameter that dynamically adjusts the blade geometry, including the chord length and thickness, to meet the specified aerodynamic and hydrodynamic design requirements. This functionality allows for efficient customization of the blade profile to suit varying flow conditions.” To initiate the design process in BladeGen, preliminary calculations using the velocity triangle method were performed to approximate the first and last points of each layer, ensuring the desired aerodynamic performance. These initial values, as presented in Table 3, served as the starting point but were not optimized.

During the optimization phase, the design was refined using 50 points distributed across five layers, with 10 points per layer. Each point within a layer was iteratively adjusted throughout the optimization process. This phase involved over 700 iterations of geometry regeneration and CFD simulations, ultimately resulting in an optimized blade design with improved performance and efficiency.” This change can be found on page 10, paragraph 2, and line 336 to 345 of the revised manuscript

Comments 11: The authors mentioned that the blockage ratio is critical. The chosen range of 0.03 to 0.05 should be justified with references or prior research. Hence, the authors need to explain why this range was selected and how it impacts the simulation accuracy.

Response 11: Thank you for pointing this out. We agree with your comment and have added the relevant reference to support our choice. Additionally, I have explained why this specific number was selected. The following information substantiates this decision: “When creating the river channel box, considering the blockage ratio is crucial. The blockage ratio, which is the ratio of the rotor area to the channel area, typically falls within the range of 0.03 to 0.05. The blockage ratio of 0.03 is selected [20]. A decrease in the blockage ratio increases the volume of water in the model, leading to more accurate simulation results.”

 This change can be found on page 11, paragraph 2, and line 372 of the revised manuscript.

Comments 12: The authors need to briefly mention the rationale for choosing a transient rather than a steady-state model and emphasize the importance of capturing unsteady effects in hydrokinetic turbines.

Response 12: Thank you for pointing this out. We agree with your comment, I have added more explanation: “As these expressions are for transient simulations, they provide critical insights into the turbine's efficiency and overall functionality under varying operating conditions, which is essential for accurately capturing unsteady effects in hydrokinetic turbines. Unlike steady-state models, which assume constant operating conditions, transient models can account for the fluctuations and dynamic changes in flow, torque, and blade performance that are typical in real-world hydrokinetic environments. As depicted in Figure 7(b), solver control is configured to ensure the accuracy and stability of the simulation results with residual value of 1.0 ×. A second-order backward Euler transient scheme is employed, which is well-suited for time-dependent simulations. Each time step involves 15 iterations, striking a balance between computational efficiency and result accuracy. This robust solver configuration allows the transient behavior of the turbine to be captured effectively, ensuring the reliability of the simulation outcomes.”  This change can be found on page 14, paragraph 1, and line 415 of the revised manuscript

Comments 13: The authors describe the boundary conditions (inlet velocity, turbulence intensity, and outlet pressure) well. However, specifying the freestream velocity range and how a turbulence intensity of 5% was determined (e.g., based on river flow conditions) would improve clarity.

Response 13: Thank you for pointing this out. I have added more explanation: “In high-turbulence cases, like rotating gears or turbines, turbulence intensity ranges from 5% to 20%. Medium-turbulence occurs in systems like large pipes or ventilation, with a range of 1% to 5%, while low-turbulence is seen in air over vehicles, with intensity under 1%. For RHKT, a turbulence intensity of 5% represents the typical operating environment, capturing the turbulence present without overestimating its effects [21]. This choice balances accuracy with computational efficiency. For the subsonic inlet flow, boundary conditions are defined with an average freestream velocity of 1.6, 2.2 and 2.8 m/s and a turbulence intensity of 5% to reflect realistic flow conditions.” This change can be found on page 14, paragraph 4, and line 438 to 443 of the revised manuscript

Comments 14: Although the SST model is widely used, a brief justification of its selection over other models (e.g., k-ε or k-ω) would improve the section's depth.

Response 14: Thank you for pointing this out. We agree with your comment, I already talked about Menter study, but I have added more explanation: “The SST k-ω model is employed because it is widely recognized as one of the most effective turbulence models for turbomachinery applications. Its strength lies in its ability to accurately manage both near-wall and free-stream flow conditions, making it particularly suitable for simulations performed using ANSYS CFX.

 Menter [21] highlighted that while the k-ε model is less sensitive to arbitrary free-stream boundary conditions, its performance in near-wall regions, particularly for boundary layers under adverse pressure gradients is inadequate. To overcome this limitation, Menter proposed a hybrid approach that combines the strengths of both models. This approach involves (i) transforming the k-ε model into a k-ω formulation near the wall and (ii) employing the standard k-ε model in fully turbulent regions far from the wall. The Reynolds stress computation and the k-equation align with Wilcox's original k-ω model, while the ε-equation is reformulated as an ω-equation by substituting ε = kω. Based on these insights, we selected the SST k-ω model for our simulations, as it effectively handles both the near-wall and free-stream flow regions, making it particularly suitable for turbomachinery applications.” This change can be found on page 16, paragraph 2, and line 468 to 471 of the revised manuscript

Comments 15: In the solver section, the authors need to consider briefly mentioning whether the simulation results will be validated with experimental data or prior literature results to strengthen confidence in the solver's accuracy.

Response 15: Thank you for pointing this out. We agree with your comment, I have added the following paragraph: “To ensure the accuracy of the solver, we first compared the simulation results with experimental data from Stephanie Ordonez-Sanchez [23], which provided a benchmark for the power coefficient. The comparison showed that the maximum power coefficient occurred at a tip speed ratio (TSR) of 4, which fully aligned with our expectations and validated the solver's reliability. This validation, alongside prior literature, strengthens confidence in the accuracy of our simulation results and the chosen modeling approach.” This change can be found on page 16, paragraph 3, and line 472 to 477 of the revised manuscript

Comments 16: The text of FEA Mechanical Analysis (Section 2.4) lacks details on the specific load cases considered in the FEA simulations (e.g., static loads, dynamic loads, or varying flow conditions). Hence, the authors need to Include this information.

Response 16: Thank you for pointing this out. We agree with your comment, but We mentioned this point in section 2.4.2. Preprocessing, where the load is directly imported from the CFD results.

Comments 17: The authors need to consider elaborating on why PEEK composites were selected over other common turbine materials, such as aluminum or standard polymer composites.

Response 17: Thank you for pointing this out. We agree with your comment, but as mentioned in Section 2.4.1, we selected PEEK based on Kalkanis' research, which highlights that composite materials are commonly used in turbines due to their excellent mechanical properties and lightweight nature. Kalkanis emphasizes the importance of composite materials for turbines because of their superior weight-to-stiffness ratio, which is essential for optimal turbine blade performance. Furthermore, as outlined by Patnaik and Amar, PEEK was chosen for its compatibility with the required functionality in turbine applications, particularly when fiber loading ranges from 30% to 40%. This aligns with their findings that such fiber-reinforced composites offer enhanced performance in demanding environments. Additionally, nano-sized particle fillers, as mentioned by Patnaik and Amar, can further improve the overall performance of these materials.

Aluminum is not commonly used for manufacturing turbine blades, especially in high-performance applications like hydrokinetic or gas turbines, due to its relatively low strength at elevated temperatures and susceptibility to corrosion. However, aluminum can be used in certain components of wind turbines, such as casings, housings, or low-load parts, because it is lightweight and has good machinability.

For turbine blades, materials like PEEK composites, carbon fiber composites, stainless steel, and titanium are more commonly chosen due to their higher strength-to-weight ratios, corrosion resistance, and ability to withstand the mechanical stresses of water flow over long periods. These materials offer better mechanical properties, including high strength resistance and durability under cyclic loading conditions, making them more suitable for the harsh operating conditions in turbines.

 

Comments 18: In the results and discussion section, the authors present initial efficiencies for the three rotors (11.5%, 13.6%, and 13.3%), followed by a detailed description of the optimization process. However, the initial efficiencies seem low compared to typical hydrokinetic turbines. While this emphasizes the importance of optimization, explaining why the initial design yielded such low values would be beneficial. Therefore, the authors need to briefly explain factors contributing to the low initial efficiency, such as suboptimal blade angles or pressure distributions.

Response 18: Thank you for pointing this out. We agree with your comment. However, it is important to note that only advanced software, such as SIEMENS HEEDS, can provide an optimized blade on the first try. Other methodologies, including the approach used in our study, require an optimization phase to refine the design. I have added further explanation to clarify this point: ” The initial analysis of the three rotor configurations yielded efficiencies of 11.5%, 13.6%, and 13.3% for Rotors 1, 2, and 3, respectively, These values are typical of early-stage turbine designs, reflecting the reliance on calculation methods and simplified assumptions, such as initial blade angles and pressure distributions. Consequently, the initial rotor designs are suboptimal, which is common in the preliminary phases before optimization processes are applied." This change can be found on page 18, paragraph 3, and line 536 to 540 of the revised manuscript.

 

Comments 19: The reported efficiencies of 45.10%, 43.27%, and 43.42% after optimization are commendable and within a reasonable range for hydrokinetic turbines. However, the authors need to discuss whether these efficiencies meet or exceed those of similar designs in the literature. This would help place the results in context and demonstrate the novelty of the proposed design.

Response 19: Thank you for pointing this out. We agree with your comment. We have added further explanation to clarify this point: ” Compared to similar designs in the literature, Tan et al. [10] achieved an efficiency of approximately 34% with a two-bladed horizontal-axis turbine, while Muratoglu and Yuce [11] reported a 43% efficiency for a three-bladed river hydrokinetic turbine. The efficiencies obtained in this study are highly competitive, underscoring the potential of the proposed design to enhance hydrokinetic energy conversion efficiency.” This change can be found on page 19, paragraph 2, and line 566 of the revised manuscript

Comments 20: The authors clearly explained the limitations of a single rotor in maintaining consistent energy production, especially during low-velocity periods. The sharp drop in energy delivery during winter months (from 4,423 kWh to 680 kWh) illustrates the problem. While the analysis highlights the seasonal energy drop, mentioning the specific period used for the calculation (e.g., monthly or quarterly) would be helpful to clarify how the energy output was determined.

Response 20: Thank you for pointing this out. I would like to clarify that our study is based on monthly calculations, as you suggested. I have provided a more detailed explanation for better clarity and I have replaced the following: “For instance, energy delivery drops sharply from 4,423 kWh during high-velocity seasons to just 680 kWh in low-velocity periods, such as winter months” with  ”For instance, energy delivery gradually decreases from February (825.4 kWh) to January (680.1 kWh). As shown in Table 10, a significant drop is observed when comparing June (4,423 kWh) with January (680.1 kWh), highlighting the sharp decline in energy production from high-velocity summer seasons to low-velocity winter periods.” This change can be found on page 25, paragraph 2, and line 642 to 646 of the revised manuscript

Comments 21: While the authors highlight key results in the conclusion section, it would be beneficial to explicitly emphasize the novelty of the proposed approach compared to existing RHKT systems.

Response 21: Thank you for pointing this out. I have now addressed this and made the necessary revisions for improved clarity. I replace the following: “A 2-blade rotor was used to lower cost and simplify deployment, retrieval and interchanging rotors. Our approach requires a RHKT that is designed that allows simple rotor exchange.” with” The novelty of this approach lies in several key improvements. First, the analysis shows how to achieve efficiencies of 43% to 45% using CFD. This is achieved through optimization of blade geometry and aerodynamic parameters, ensuring maximum energy capture under varying flow conditions. Second, the system can effectively almost double the CF compared to existing RHKT systems that no other approach can compare. This improvement in CF is critical for enhancing energy production in low-flow conditions, where exciting RHKT systems are underperforming in winter. Additionally, the design features a 2-blade rotor, which reduces both costs and complexity during deployment and retrieval. By removing the need for a shroud and gearbox, the system becomes lighter, more reliable, and has fewer components, allowing for easier access and servicing with boats available in remote communities. With three rotors designed, the system allows rotors to exchange, depending on the seasonal water velocity. This design is especially beneficial for remote communities, where deployment and maintenance can be managed using local boats, eliminating the need for specialized technicians. The approach is more suited for smaller turbines. “This change can be found on page 29, paragraph 2, and line 722 to 736 of the revised manuscript