Dynamic Load Sharing Behavior for the Pitch Drive in MW Wind Turbines
Round 1
Reviewer 1 Report
The authors raised a very interesting issue. There are few studies that talk about manufacturing errors or those related to improper assembly of wind turbine components. Most of the research work takes into account errors caused by exploitation. The problem is all the more interesting as it concerns multi-stage planetary gearboxes. It is true that the authors have described the results of, for example, the displacement response in the time domain at individual gear stages, but these are poorly represented in the graphs. Perhaps it would be better to look for an independent signal measure that could be used in a practical way. What do the authors suggest and how can the errors indicated in the article be controlled (or checked for occurrence) during gearbox operation ? How can the vibration signals obtained in Figure 16 be related to the changes occurring in the gearbox ? Is it possible to deduce what the difference between theoretical and experimental calculations would be in the event of significant changes in the errors analysed ? Can general conclusions be drawn and applied in practice (on a working wind turbine gearbox) ?
Author Response
Response to Reviewer 1 Comments
Point 1: It is true that the authors have described the results of, for example, the displacement response in the time domain at individual gear stages, but these are poorly represented in the graphs. Perhaps it would be better to look for an independent signal measure that could be used in a practical way.
Response 1: Thank you from the bottom of my heat! We have measured the vibration signal of the output shaft in a test bench in a practical way. Perhaps it implement the suggest.
Point 2: What do the authors suggest and how can the errors indicated in the article be controlled (or checked for occurrence) during gearbox operation?
Response 2: Authors suggest that the eccentric error of the planetary gear and carrier should be controlled effectively. And the advice is written in the corresponding area and conclusion. These errors indicated in the article can be controlled in the design stage, and also can be controlled through load sharing measures (such as flexible pin) during gearbox operation.
Point 3: How can the vibration signals obtained in Figure 16 be related to the changes occurring in the gearbox?
Response 3: The vibration sensor test radial acceleration signal of the the output shaft. A signal processor inside the vibration sensor can calculate the velocity and displacement according to the vibration signal automatically, and then output the vibration displacement in time domain.
Point 4: Is it possible to deduce what the difference between theoretical and experimental calculations would be in the event of significant changes in the errors analyzed? Can general conclusions be drawn and applied in practice (on a working wind turbine gearbox)?
Response 4: In order to test the displacement under different loads according to other reviewers’ comment, we did the experiment again. This time the difference between theoretical and experimental calculations are small, so it may be because of the installation of the sensor. Besides, general conclusions are drawn in the last part of this paper, and they can be applied in design stage of the planetary gear transmission.
Sincerely thanks for your advice! Wish you everything goes well.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The authors are proposing a paper that aims to evaluate the load sharing between planets. I have found the article very difficult to read. Indeed, the authors are using the article “the” everywhere. However, the use of “the” implies the identification of the “object”. Indeed, there is a difference between “I have gearboxes” (I have some gearboxes) and “I have the gearboxes” (I have more than one gearbox, and we both know which gearboxes we are talking about). When the authors claim that “The planetary gear transmission is characterized by a high speed ratio, and its input torque is split into several parallel paths among the planetary gears in the same stage, so its structure is more compact than other gear transmissions.”, they are implying that there is only a unique “planetary gear transmission” and that both authors and readers know which “planetary gear transmission” they are talking about. This error affects the whole paper.
Furthermore, the authors are also using a not completely correct nomenclature. Like “central gear floating” (i.e. a “floating sun”) and “planetary gear transmission” (i.e. a “planetary gearbox”, “planetary geared transmission” is non strictly correct as the concept of planetary\epycicloidal modules already includes the idea of a geared transmission).
The novelty of the article is not clear. In the review of the state-of-the-art (which mainly covers papers from the period 2015-2017), the authors claim that “few studies have focused on the multistage planetary gear transmission”. However, those studies are nor cited nor discussed. Furthermore, it is not clear what is the difference between their proposed model and those discussed in the literature. Finally, the state of the art is very poor. There are even review papers on the topic (e.g. https://doi.org/10.3390/machines10080634).
Furthermore, the idea behind this paper is not clear. The authors claim to have modelled a multistage planetary gearbox and that all the stages are connected with each other. However, do we really need this model? Is the ith stage influencing the jth one? And, if we really need this model, why “ All planetary gears are assumed to be identical with the same mass, inertia, gear profile error and gear mesh stiffness.” when studying (and focusing) on a multistage planetary gearbox?
Some questions also arise by looking at the experimental part (which is limited to only one test condition, i.e. only one speed\torque configuration). How “The vibration displacement [is]obtained by the signal collected from the sensor” can be related to the load sharing between planets? Furthermore, also the picture is not clear, where did the sensor\\sensors have been located? Why did a “CA-YD-141” sensor, which should be an accelerometer, have been selected in order to measure the “vibration displacement”? How have the data been elaborated?
Author Response
Response to Reviewer 2 Comments
Point 1: The authors are proposing a paper that aims to evaluate the load sharing between planets. I have found the article very difficult to read. Indeed, the authors are using the article “the” everywhere. However, the use of “the” implies the identification of the “object”. Indeed, there is a difference between “I have gearboxes” (I have some gearboxes) and “I have the gearboxes” (I have more than one gearbox, and we both know which gearboxes we are talking about). When the authors claim that “The planetary gear transmission is characterized by a high speed ratio, and its input torque is split into several parallel paths among the planetary gears in the same stage, so its structure is more compact than other gear transmissions.”, they are implying that there is only a unique “planetary gear transmission” and that both authors and readers know which “planetary gear transmission” they are talking about. This error affects the whole paper.
Response 1: Thanks a million for your suggests. I am sincerely sorry. Actually, the manuscript sent to you is after the American Journal Experts polishing. We have modified this mistake and revised English in the revision paper.
Point 2: Furthermore, the authors are also using a not completely correct nomenclature. Like “central gear floating” (i.e. a “floating sun”) and “planetary gear transmission” (i.e. a “planetary gearbox”, “planetary geared transmission” is non strictly correct as the concept of planetary\epycicloidal modules already includes the idea of a geared transmission).
Response 2: We have revised these mistakes in this paper. Thank you!
Point 3: The novelty of the article is not clear. In the review of the state-of-the-art (which mainly covers papers from the period 2015-2017), the authors claim that “few studies have focused on the multistage planetary gear transmission”. However, those studies are nor cited nor discussed. Furthermore, it is not clear what is the difference between their proposed model and those discussed in the literature. Finally, the state of the art is very poor. There are even review papers on the topic (e.g. https://doi.org/10.3390/machines10080634).
Response 3: We update the introduction and the state-of-the-art mainly covers papers from the period 2017-2023. Besides, the study focused on the multistage planetary gear transmission is also cited after revision. The difference between their proposed model and those discussed in the literature is the excitation, their proposed model considered the interaction of random input speed and stochastic excitation. However, the model considers definite input speed, internal excitation and external torque.
Point 4: Furthermore, the idea behind this paper is not clear. The authors claim to have modeled a multistage planetary gearbox and that all the stages are connected with each other. However, do we really need this model? Is the ith stage influencing the jth one? And, if we really need this model, why “All planetary gears are assumed to be identical with the same mass, inertia, gear profile error and gear mesh stiffness.” when studying (and focusing) on a multistage planetary gearbox?
Response 4: we really need this model of a multistage planetary gearbox because the uneven load distribution of planet gear pair in the first-stage is more intense than other stages, and the load sharing coefficient of the internal meshing decreases with the stage. About this sentence “All planetary gears are assumed to be identical with the same mass, inertia, gear profile error and gear mesh stiffness”, I am very sorry for not writing clearly, they are assumed to be identical in the same stage, but the mass, inertia, gear profile error and gear mesh stiffness of planetary gears in different stages are not the same. We have declared this point clearly in this manuscript.
Point 5: Some questions also arise by looking at the experimental part (which is limited to only one test condition, i.e. only one speed\torque configuration). How “The vibration displacement [is]obtained by the signal collected from the sensor” can be related to the load sharing between planets? Furthermore, also the picture is not clear, where did the sensor\\sensors have been located? Why did a “CA-YD-141” sensor, which should be an accelerometer, have been selected in order to measure the “vibration displacement”? How have the data been elaborated?
Response 5: We finished the experiment once more to test the displacement under 50%,100% and 150% rated torque. Furthermore, because the load sharing coefficient could not be measured directly from the experiment, but the displacement can be tested and calculated from the dynamic load sharing model directly, so we use the displacement to verify the theoretical model. Besides, the sensor is labeled in the replaced picture in the revision. As we redo the experiment, most experiment part of this manuscript is rewritten. The accelerometer sensor tests radial acceleration signal of the the output shaft. A signal processor inside the vibration sensor can calculate the velocity and displacement according to the vibration signal automatically, and then output the vibration data in time domain. I am sorry for not writing it clearly, and now it is declared distinctly in the revision.
Thanks from the bottom of my heart for your suggests!
Wish you everything goes well.
Author Response File: Author Response.docx
Reviewer 3 Report
Dear authors, for good paper with appropriate scientific approuch you should compare 3 cases, not just one. You gave results with 1600 min-1, but there should be with for example 750 min-1 and 2500 min-1. Or you should give some other difference like in load.
Figure 1, 3, 4, 5, 14 and 15 are not clear. Please increase quality.
In Introduction, last part or as separate, please give what is novelty of this paper. Is it just a case or what?
"However, all the above studies are related to the load sharing behavior of the single-stage planetary gear system, few studies have focused on the multistage planetary gear transmission. " - I am not sure about this. There are studies that are dealing with multistage panetary gear transmission.
This paper https://www.mdpi.com/2076-3417/9/24/5501 is dealing with this problematic.
References are not the newest. Please update the references. Mostly, they are 2015-2017.
Line 210 - mistake in jth.
The whole experiment as well as numerical approuch is given for one speed and one load? There should be given additional tests in which you will show comparison of exerimental and numerical results.
In 4.3 is given table with results. It should be better explaned with procentual difference.
This results (4.3) are crucial for the paper. Some sentence of it should be found in Conclusion.
Author Response
Response to Reviewer 3 Comments
Point 1: Dear authors, for good paper with appropriate scientific approuch you should compare 3 cases, not just one. You gave results with 1600 min-1, but there should be with for example 750 min-1 and 2500 min-1. Or you should give some other difference like in load.
Response 1: Thank you very much for your advice! We redid the experimental and obtained the vibration displacement of the output shaft under 50%, 100% and 150% rated torque, and compared them with the theoretical results. In all the three types of load, the theoretical results are similar with the tested results. And the experiment part in this paper is modified.
Point 2: Figure 1, 3, 4, 5, 14 and 15 are not clear. Please increase quality.
Response 2: The quality of Figures 1,3,4,5,14, and 15 has been improved.
Point 3: In Introduction, last part or as separate, please give what is novelty of this paper. Is it just a case or what?
Response 3: We have add the novelty of this paper in the last part of Introduction. Accordingly, yaw and pitch drives are actuated by definite motor loads, and they consist a multistage planeraty gear system in engineering. Thus, it is necessary to investigate the load sharing performance of the multistage planetaty gear system of wind power systems with definite loads.
Point 4: "However, all the above studies are related to the load sharing behavior of the single-stage planetary gear system, few studies have focused on the multistage planetary gear transmission. " - I am not sure about this. There are studies that are dealing with multistage panetary gear transmission.
This paper https://www.mdpi.com/2076-3417/9/24/5501 is dealing with this problematic.
Response 4: The mentioned paper about multistage planetary gear transmission is cited after revision. However, the difference between their proposed model and those discussed in the literature is the excitation, their proposed model focused interaction of random input speed and stochastic wind load. However, the model considers definite input speed, internal excitation and external torque.
Point 5: References are not the newest. Please update the references. Mostly, they are 2015-2017.
Response 5: We update the references and the state-of-the-art mainly covers papers from the period 2017-2023.
Point 6: Line 210 - mistake in jth.
Response 6: We have corrected our mistake
Point 7: The whole experiment as well as numerical approuch is given for one speed and one load? There should be given additional tests in which you will show comparison of exerimental and numerical results.
Response 7: We finished the experiment once more to test the displacement under 50%,100% and 150% rated torque. The comparison between exerimental and numerical results in three different load cases are listed in this manuscript.
Point 8: In 4.3 is given table with results. It should be better explaned with procentual difference.
This results (4.3) are crucial for the paper. Some sentence of it should be found in Conclusion.
Response 8: The difference between theoretical and experimental calculations may be because of the installation of the sensor，it is magnetic attracted on the output shaft housing. We have explaned, and the results are listed in the conclusion in the revision.
Sincerely thanks for your advice!
Wish you everything goes well.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Reviewer: Furthermore, the idea behind this paper is not clear. The authors claim to have modeled a multistage planetary gearbox and that all the stages are connected with each other. However, do we really need this model? Is the ith stage influencing the jth one? And, if we really need this model, why “All planetary gears are assumed to be identical with the same mass, inertia, gear profile error and gear mesh stiffness.” when studying (and focusing) on a multistage planetary gearbox?
Authors: we really need this model of a multistage planetary gearbox because the uneven load distribution of planet gear pair in the first-stage is more intense than other stages, and the load sharing coefficient of the internal meshing decreases with the stage. About this sentence “All planetary gears are assumed to be identical with the same mass, inertia, gear profile error and gear mesh stiffness”, I am very sorry for not writing clearly, they are assumed to be identical in the same stage, but the mass, inertia, gear profile error and gear mesh stiffness of planetary gears in different stages are not the same. We have declared this point clearly in this manuscript.
Reviewer: The fact that we see, different load sharings at each load stage does not imply that we need a model in which the stages are coupled. We need the author model if it is proved that "problems" in the first stage are affecting also the other stages as well. This aspect is not clearly depicted in the article.
Author Response
Point 1:
Reviewer: Furthermore, the idea behind this paper is not clear. The authors claim to have modeled a multistage planetary gearbox and that all the stages are connected with each other. However, do we really need this model? Is the ith stage influencing the jth one? And, if we really need this model, why “All planetary gears are assumed to be identical with the same mass, inertia, gear profile error and gear mesh stiffness.” when studying (and focusing) on a multistage planetary gearbox?
Authors: we really need this model of a multistage planetary gearbox because the uneven load distribution of planet gear pair in the first-stage is more intense than other stages, and the load sharing coefficient of the internal meshing decreases with the stage. About this sentence “All planetary gears are assumed to be identical with the same mass, inertia, gear profile error and gear mesh stiffness”, I am very sorry for not writing clearly, they are assumed to be identical in the same stage, but the mass, inertia, gear profile error and gear mesh stiffness of planetary gears in different stages are not the same. We have declared this point clearly in this manuscript.
Reviewer: The fact that we see, different load sharings at each load stage does not imply that we need a model in which the stages are coupled. We need the author model if it is proved that "problems" in the first stage are affecting also the other stages as well. This aspect is not clearly depicted in the article.
Response 1: Thank you very much for your insist on this problem. We decouple the load sharing model of a multistage planetary gearbox to be a model of a single-stage by delete the coupling stiffness between stages in the stiffness matrix, and discover that there is very small difference between the coupled model and decoupled model. We sincerely sorry for last reply for this problem. Therefore, we modify the title and content of this manuscript. The revised title is “ Dynamic Load Sharing Behavior for The Pitch Drive in MW Wind Turbines”, and we have changed the “multistage planetary gearbox” to be “pitch drive”. Besides, the introduction has also been rewritten. Because we submit the manuscript to the special issue “Renewable Energy in Environmental Engineering ”of the Processes, thus, the revision still satisfies the theme of the journal. Thanks a million for your support!!!
Thanks from the bottom of my heart for your suggests!
Wish you everything goes well.
Author Response File: Author Response.docx
Reviewer 3 Report
The authors made significant improvement of the paper according to the comments.
Author Response
Point 1: The authors made significant improvement of the paper according to the comments.
Response 1: The introduction and cited references has been revised based on the theme. Thank you very much for your approval!
Thanks from the bottom of my heart for your suggests!
Wish you everything goes well.
Author Response File: Author Response.docx
Round 3
Reviewer 2 Report
Accept in present form
