Review Reports
- Alexander Shestakov,
- Dmitry Galyshev and
- Olga Ibryaeva*
- et al.
Reviewer 1: Anonymous Reviewer 2: Anonymous Reviewer 3: Hsin-Jang Shieh
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe comments are attached.
Comments for author File: 
Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for Authors- what is the fundamental reason for the hybrid CNN-MLP model is that it performs better.
- What are the limitations of the MLP, CNN, and hybrid CNN-MLP model?
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for Authors- It is unclear about what the normal and defective data are. How to define the normal and defective? This manuscript appears to built a network, to train it by a set of the known data, and then show the results. It is not about an advanced skill development in detecting the defects of an induction motor drive.
- It seems that only the stator current spectra is concerned. Did the developed method use three-phase stator currents or one phase? If using one phase current solely, how can we know the defects from one of the three-phase currents?
- What are GASF (Line #266) and GADF (Line #269)?
- Why needs the DFT?
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
Comments and Suggestions for AuthorsThe authors did not completely address the comments. How to know the normal and defective conditions from the measured currents? Perhaps the measured currents have a spike during a short time, a large noise impact on the current, or abnormal due to jitter or vibration.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 3
Reviewer 3 Report
Comments and Suggestions for AuthorsThanks for the authors' responses.