Common-Mode Voltage Reduction Algorithm for Photovoltaic Grid-Connected Inverters with Virtual-Vector Model Predictive Control

Round 1

Reviewer 1 Report

It is advised not to use so many abbreviations in the Abstract section, as the section has to have meaning by its own.
There are some spelling errors corrected in the attached file, but there is a need to review the comma usage throughout the entire paper.

 

 

Author Response

Reviewer#1, Concern # 1:

1. It is advised not to use so many abbreviations in the Abstract section, as the section has to have meaning by its own.

Author response: Thank you for pointing this out.

Author action: We have revised our manuscript and minimize the use of abbreviations in Abstract. Please see the attachment.

Reviewer#1, Concern # 2:

2. There are some spelling errors corrected in the attached file, but there is a need to review the comma usage throughout the entire paper.

Author response: Thank you for pointing this out.

Author action: Apologies for the careless mistakes. We have to the best of our ability proofread the entirely manuscript, including improving some sentences (including abstract) and correcting all grammatical errors. All changes have been highlighted in yellow. Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

No experiments and analysis is found towards the proposed title

The predictive control method analysis is missing

The complete manuscript focus only on SVPWM technique where it fails to give details about the proposed predictive control algorithm

The research progress was not done completely with reference to the proposed title

Author Response

Reviewer#2, Concern # 1:

1. 1. No experiments and analysis is found towards the proposed title

Author response: Thank you for pointing this out. We have now elaborated both CMRSVPWM and MPC, and included more results. Please see the attachment.

Author action: We have modified the Introduction (lines 35 to 98), added the MPC analysis (lines 123 to 147) and more simulation results in paper 9 (lines 222 to 242) to explain how the MPC embedded with CMRSVPWM can improve the CMV as compared to the state of the arts. Please see the attachment.

Reviewer#2, Concern # 2:

2. 2. The predictive control method analysis is missing

Author response: We have included further explanation of MPC. We have also highlighted that the conventional MPC will suffer from higher harmonics when the control sampling frequency is insufficiently high. Details of MPC, which uses virtual vectors, is described further. Virtual voltage vectors with 30 degree angle interval (around the hexagonal voltage vector space) and ten sub-voltage vectors (for each phase angle) are used. Please see the attachment.

Author action: We have modified Section 2.2 (lines 123 to 140).

Reviewer#2, Concern # 3:

3. 3. The complete manuscript focus only on SVPWM technique where it fails to give details about the proposed predictive control algorithm

Author response: Thank you for pointing this out. We have now included more details of MPC. 

Author action: We have added the simulation in the newly added Section 4.1 (lines 228 to 242 ). Please see the attachment.

Reviewer#2, Concern # 4:

4. 4. The research progress was not done completely with reference to the proposed title

Author response: Apologies for not including sufficient details in the first version. We have reorganized the manuscript to reflect the content better. More simulation is added to highlight that CMRSVPWM with virtual vector MPC can improve the CMV amplitude and toggling frequency.

Author action: We have modified Introduction (lines 33 to 98), detailed more MPC part (lines 99 to 140) and added experiments (lines 223 to 242). Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

In the paper, the authors proposed an improved CMRSVPWM modulation method to address the issues of high common-mode voltage and high harmonic distortion rate in the AC output side of a three-phase two-level PV inverter. The approach is interesting and the obtained results are encouraging. However, there are some observations which the author should consider to improve the structure and quality of the paper:

1. A list with acronyms and notations should be introduced.
2. In the Abstract the authors assert “The findings indicate that the proposed approach is capable of suppressing CMV and reducing the frequency jump in a single cycle ….". It would be useful for readers to be indicated the values (they could be given in percent) which to strengthen these statements.
3. In the first part of the paper, the authors present various approaches from the literature. Maybe, a synthesis of the solutions proposed in the literature depending on the type of analysis, which to highlight more clearly the advantages and disadvantages, is useful for readers. This synthesis can be given as a table.
4. The references should be indicated for the general equations presented in Section 2, which did not obtain by the authors through demonstration.
5. The Simulink model should be introduced so that the verification of the results to be easily made.
6. The obtained results should be better explained. Additional information should be introduced so that all details to be understood.
7. The authors should present the limits of the proposed approach and their future work.

Author Response

Reviewer#3, Concern # 1:

1. 1. A list with acronyms and notations should be introduced.

Author response: Thank you for raising this. This is indeed a useful addition to the work.

Author action: We have added the Appendix with S1 (line 294) and S2 (line 295) to list with acronyms and notations. Please see the attachment.

Reviewer#3, Reviewer#2:

2. 2. In the Abstract the authors assert “The findings indicate that the proposed approach is capable of suppressing CMV and reducing the frequency jump in a single cycle ….". It would be useful for readers to be indicated the values (they could be given in percent) which to strengthen these statements

Author response: We have added more intuitive data to make it clearer to the reader. 

Author action: We have revised the Abstract (lines 13 to 28). Please see the attachment.

Reviewer#3, Concern # 3:

3. 3. In the first part of the paper, the authors present various approaches from the literature. Maybe, a synthesis of the solutions proposed in the literature depending on the type of analysis, which to highlight more clearly the advantages and disadvantages, is useful for readers. This synthesis can be given as a table.

Author response: We list a table to compare SVPWM, AZSPWM, NSPWM and RSPWM with advantages and disadvantages in the from the reference.

Author action: We have added the Table 1 (line 98) to highlight the advantages and disadvantages. Please see the attachment.

Reviewer#3, Concern # 4:

4. The references should be indicated for the general equations presented in Section 2, which did not obtain by the authors through demonstration.

Author response: Thank you for raising this.

Author action: We have indicated the references for the general equations and highlight them (lines 143, 157 and 195). Please see the attachment.

Reviewer#3, Concern # 5:

5. 5. The Simulink model should be introduced so that the verification of the results to be easily made

Author response: We added a control block diagram as Figure 8. We’re of the opinion that “screen capture” of Matlab-Simulink is less common in any journal paper; however, if the reviewer insists, please kindly let us know and we shall include that in the next version.

Author action: We have supplemented Section 4 (lines 223 to 227).

Reviewer#3, Concern # 6:

6. 6. The obtained results should be better explained. Additional information should be introduced so that all details to be understood.

Author response: We have redrawn the experimental results to include more measurable parameters. And a more detailed description is added for better analysis. Please see the attachment.

Author action: We have adjusted Figure 10 (line 252), Figure 11 (line 257) and Figure 12 (line 258)

Reviewer#3, Concern # 7:

7. 7. The authors should present the limits of the proposed approach and their future work.

Author response: Thank you for pointing this out.

Author action: We have added the Section 6 (lines 281 to 285) to present deficiencies and prospects. Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

The investigation presented in the introduction is critical / comparative /, highlighting advantages / disadvantages of the mentioned methods / techniques, but other Common-Mode Reduction SVPWM (CMRSVPWM) methods are not discussed (see for example 10.3390/en13153884 etc.).

1) define CMRSVPWM; include a nomenclature.

2) discuss other CMRSVPWM methods in the introduction.

The research objective is defined.

The novelty is mentioned at the end of introduction in points-wise manner.

3) Mention the new added value compared to 10.3390/en13153884.

The model, parameters and operation conditions are presented.

The model is validated and the main results are presented and discussed.

4)  Enlarge the figures for better visibility (using left indentation = 0).

5) include summary tables of the results presented; includes tables comparing the results obtained in other CMRSVPWM methods.

The main findings of the research study are presented in conclusion section.

6) Mention the next work based on the results presented in this paper

7) include more reference published in the last 5 years to highlight the topicality of this research study (more than 50% should be).

Author Response

Reviewer#4, Concern # 0:

1. 0. The investigation presented in the introduction is critical / comparative /, highlighting advantages / disadvantages of the mentioned methods / techniques, but other Common-Mode Reduction SVPWM (CMRSVPWM) methods are not discussed (see for example 10.3390/en13153884 etc.).

Author response: We list a table to compare SVPWM, AZSPWM, NSPWM and RSPWM with advantages and disadvantages in the from the reference.

Author action:We have added the Table 1 (line 98) to highlight the advantages and disadvantages. Please see the attachment.

Reviewer#4, Concern # 1:

1. 1. Define CMRSVPWM; include a nomenclature

Author response: Thank you for pointing this out.

Author action: We have defined the CMRSVPWM when we first mentioned it in Abstract (lines 18 to 20). Please see the attachment.

Reviewer#4, Concern #2:

1. Discuss other CMRSVPWM methods in the introduction. The research objective is defined. The novelty is mentioned at the end of introduction in points-wise manner.

Author response: We list a table to compare SVPWM, AZSPWM, NSPWM and RSPWM with advantages and disadvantages in the from the reference.  And Novelty has mentioned in points-wise manner.

Author action: 1) We have added the Table 1 (line 98) to highlight the advantages and disadvantages in and introduced other methods in Section 2.4 (lines 162 to 168). Please see the attachment.

• Novelty has mentioned in Introduction (lines 80 to 86). Please see the attachment.

Reviewer#4, Concern # 3:

3. 3. Mention the new added value compared to 10.3390/en13153884.

The model, parameters and operation conditions are presented.

The model is validated, and the main results are presented and discussed.

Author response: The proposed CMRSVPWM has higher DC-bus utilization then the application scope is wider than 10.3390/en13153884. Meanwhile , we have compared other modulation methods in Table 6.

Author action: We have modified the Section 3.3 (lines 202 to 221) to explain the new added value compared to 10.3390/en13153884 and added the Table 6 (line 267) to compare the CMR methods include 10.3390/en13153884. Please see the attachment.

Reviewer#4, Concern # 4:

4. 4. Enlarge the figures for better visibility (using left indentation = 0).

Author response: We have redrawn the experimental results to include more measurable parameters. And a more detailed description is added for better analysis.

Author action: We have adjusted Figure 10 (line 252), Figure 11 (line 257) and Figure 12 (line 258). Please see the attachment.

Reviewer#4, Concern # 5:

4. Include summary tables of the results presented; includes tables comparing the results obtained in other CMRSVPWM methods.

  The main findings of the research study are presented in conclusion section.

Author response: We have compared CMRSVPWM and other mentioned modulation methods in peak CMV, CMV frequency, CMV frequency at changing sectors, DC-bus utilization and Phase-A current THD.

Author action: We have added the Table 6 (line 267) and supplemented conclusions (lines 278 to 280). Please see the attachment.

Reviewer#4, Concern # 6:

6. Mention the next work based on the results presented in this paper

Author response: Thank you for pointing this out.

Author action: We have added the Section 6 (lines 281 to 285) to present deficiencies and prospects. Please see the attachment.

Reviewer#4, Concern # 7:

7. Include more reference published in the last 5 years to highlight the topicality of this research study (more than 50% should be).

Author response: We have re-read paper published in the last 5 years and sorted out the Introduction and Reference, and quoted more articles in the recent five years.

Author action: We have modified the Introduction section(lines 33 to 98),and some reference have changed (lines 298 to 303, 310 to 312, 316 to 318, 327 to 329, 337 to 339, 346 to 350, 383 to 385, 388 to 390 and 393 to 400). Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The author focused only on SVPWM concept. The proposed title is not only the concept of SVPWM. He should also focus on Common-mode Voltage Reduction Algorithm and Multi-Step Model Predictive Control technique. Few lines of explaining what is MPC concept will not complete the research. 

Model predictive control is an advanced method of process control that control the process by satisfying a set of constraints. Here there is no set of constraints defined and no algorithm is framed for the same.

MPS is an algorithm of optimal control. For the proposed title where the optimization technique is focused. No clarity with respect to sampling time, Time control horizon in Photovoltaic Grid-connected Inverters.

Author Response

Reviewer#2, Concern # 1:

1. The author focused only on SVPWM concept. The proposed title is not only the concept of SVPWM. He should also focus on Common-mode Voltage Reduction Algorithm and Multi-Step Model Predictive Control technique. Few lines of explaining what is MPC concept will not complete the research.

Model predictive control is an advanced method of process control that control the process by satisfying a set of constraints. Here there is no set of constraints defined and no algorithm is framed for the same.

MPC is an algorithm of optimal control. For the proposed title where the optimization technique is focused.

 

Author response: Once again, thank you for your comments. There are several points to address. All responses with changes to the manuscript are described under “Author action” while the rest is explained directly below here.

On your comments on MPC being an advanced process/optimal control. We very much agree to that. MPC has long history of successful implementation in slow process control such as oil refinery. The history of MPC in conjunction with power electronic application is relatively shorter – it started in around 2008-2010, and it is still being actively studied in many power electronic application. In majority of the application, constraints have often not considered. This is due to several technical reasons:

• Exceptionally short control sampling period, of about sub 100μs. This short implementation time demands for a simpler algorithm.
• Discreate nature of power electronic converters with finite control input set. The finite amount of discrete control inputs (e.g. 8 in standard MPC, 120 in virtual-vector in this work) warrants a simpler, enumeration-based optimal searching algorithm as compared to the continuous input counterpart (or discrete-continuous control input counterpart), which are often implemented without constraints.
• The fast electrical dynamic often only require very short prediction and control horizons. This is an established MPC feature in most (if not all) power converter application.

We are fully aware of the nature of MPC in the process control – indeed, it is where the inspiration is drawn from for its application in power electronic application. We hope that you can accept the clarification on the algorithmic differences in these two distinctive applications.

Author action:

On your first comment above, we think that the concern could be due to the use of “multi-step” wording in the title, which could mislead the reviewer and readers. Hence, we have changed the wording from “multi-step model predictive control” to “virtual-vector model predictive control”, which can reflect more accurately on the type of MPC used in the work. In addition, we have added Figure 10 (line 243) to highlight the feature of 120 virtual vectors used by the MPC.

Reviewer#2, Concern # 2:

2. No clarity with respect to sampling time, Time control horizon in Photovoltaic Grid-connected Inverters.

Author response: Thank you for pointing this out. We have missed that. Apologies.

Author action: We have included the prediction and control horizon values in Table 5 (line 248).

Author Response File: Author Response.pdf

Reviewer 3 Report

The authors have tried and succeed to respond satisfactorily to each issue raised by the reviewer. Thus, they performed changes in the initial manuscript, and new explanations and elaborations of details have been brought. 

Author Response

Once again, thank you for your comments. 

Author Response File: Author Response.pdf

Reviewer 4 Report

The authors revised.

Author Response

Once again, thank you for your comments. 

Author Response File: Author Response.pdf