Comparing Different Resonant Control Approaches for Torque Ripple Minimisation in Electric Machines

Round 1

Reviewer 1 Report

In my opinion proposed topic is potentially interesting.

However proposed resonant filter is substantially a pole cancellation.

Is possible to verify the robustness of proposed solution respect to a shift of the filter frequency respect to frequency that must be cancelled.

This is not a trivial question and it's  a quite common trouble in many other sectors such as design of active or semi-active suspensions  for vehicles or for other mechatronic equipments.

In this sense i suggest to authors to look at some works concerning the physical analogy between mechanical and electrical system such as this one (that you can eventually cite).

Also looking at classical approaches followed in mechanics and mechatronics a strict pole cancellation is not very robust so i suggest to look at techniques of simulation of mechanical impedance which don't use the knowledge of a precise couple of poles.

Allotta, B., Pugi, L., Bartolini, F. An active suspension system for railway pantographs: The T2006 prototype (2009) Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit, 223 (1), pp. 15-29.  DOI: 10.1243/09544097JRRT174

Author Response

Based on the comments and on an additional internal review, we have made significant edits to the revised paper that should help to make it much more accessible and more explicit about the choices and the contributions made.

Our responses to the reviewers’ comments are given below. Changes in the revised manuscript are set in blue text.

Authors’ Response to Reviewer 1

• In my opinion proposed topic is potentially interesting.

Thank you for your interest in this topic. We appreciate you spending your time reviewing this manuscript and your comments to help us improve it.

• However proposed resonant filter is substantially a pole cancellation.
• Is possible to verify the robustness of proposed solution respect to a shift of the filter frequency respect to frequency that must be cancelled.

Thank you for the suggestion. This is a similar problem, but not quite the same. We have also highlighted that there is a difference between the control of resonant systems (where the resonance is in the plant) and the control of ripple, where the resonance is in the disturbance and therefore in the controller.

• This is not a trivial question and it's a quite common trouble in many other sectors such as design of active or semi-active suspensions  for vehicles or for other mechatronic equipments.
• In this sense i suggest to authors to look at some works concerning the physical analogy between mechanical and electrical system such as this one (that you can eventually cite).
• Pugi, L., Reatti, A., Corti, F. Application of modal analysis methods to the design of wireless power transfer systems (2019) Meccanica, 54 (1-2), pp. 321-331.  DOI: 1007/s11012-018-00940-x
• Also looking at classical approaches followed in mechanics and mechatronics a strict pole cancellation is not very robust so I suggest to look at techniques of simulation of mechanical impedance which don't use the knowledge of a precise couple of poles.
• Allotta, B., Pugi, L., Bartolini, F. An active suspension system for railway pantographs: The T2006 prototype (2009) Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit, 223 (1), pp. 15-29.  DOI: 1243/09544097JRRT174

Thank you for the references, we have included one as [9] on page 2 in the revised manuscript. Both lead to a resonance in the closed loop, but the robustness issues is quite different. Because the speed of a synchronous motor is known exactly, there is no error on the ripple frequency. It is mentioned in the paper that some of the approaches can be based on the angle measurement, which makes them fully synchronous.  

Reviewer 2 Report

In the submitted manuscript, the authors compare resonant control methods for torque ripple minimisation in permanent magnet synchronous machines.

There are two main issues that makes me to reconsider your work after revision.

Please explicitly state the novelty of your research. If you study and compare already established control methods (PI, PIR variations, PIRA, ILC), then please clearly distinguish distinctive features of your research making it article paper, but not the review.

Please also explain in detail the limitations on which you examin the control approaches only with simulations. Solid journal papers must rely on measurement data.

Author Response

Authors’ Response to Reviewers

Based on the comments and on an additional internal review, we have made significant edits to the revised paper that should help to make it much more accessible and more explicit about the choices and the contributions made.

Our responses to the reviewers’ comments are given below. Changes in the revised manuscript are set in blue text.

Authors’ Response to Reviewer 2

• In the submitted manuscript, the authors compare resonant control methods for torque ripple minimisation in permanent magnet synchronous machines.

Thank you for taking the time to read our manuscript in detail. We are grateful for your constructive comments to help improve our manuscript.

• There are two main issues that makes me to reconsider your work after revision.
• Please explicitly state the novelty of your research. If you study and compare already established control methods (PI, PIR variations, PIRA, ILC), then please clearly distinguish distinctive features of your research making it article paper, but not the review.

We apologise that the novelty was not clear in the original manuscript. The novelty here comes from comparing the algebraic definitions of the different control laws, and then proving that they are mathematically equivalent. This is new work, building on previous authors’ works and providing new knowledge and insight, thus we believe it should be an article rather than a review. We have added a section in the Introduction on page 2 of the revised manuscript. In addition, we have edited the manuscript to make the novelty of the concept clearer.

• Please also explain in detail the limitations on which you examine the control approaches only with simulations. Solid journal papers must rely on measurement data.

Thank you for your comments. Unfortunately, in this case we did not have the resources to obtain experimental results. Instead, we have shown that the different controllers are algebraically similar and used Simulink simulations to further reinforce this. We believe that this is proof enough of the underlying premise of the paper and that it provides fertile ground for our future work in this area. In light of your comment, we have added a clarification to the paper to explain the choice of methodology on page 9 of the revised manuscript.

Round 2

Reviewer 2 Report

The authors have correctly addressed the comments of the previous review stage. The revised manuscript is closer to the version for publication. I can only recommend minor revision aimed at formatting and proofreading. I wish the authors success in the subsequent experimental studyes.