Experimental and Computational Vibration Analysis for Diagnosing the Defects in High Performance Composite Structures Using Machine Learning Approach
, Saravanakumar Devaraj 2, Senthilkumar Marikkannan 1, Sakthivel Gnanasekaran 2, Sivakumar Ramasamy 1, Jegadeeshwaran Rakkiyannan 2,* and Yigeng Xu 3
Round 1
Reviewer 1 Report
fine paper, good work, but the work must be improved before publishing, especially the claim IS WRONG.
i) to much formatting and grammar errors, the authors must go through their paper again and homogenize wording and formatting. furthermore it is strongly recommended to integrate equations into sentences.
ii) 75 % of the modeling part is well known, there is no need to repeat existing knowledge. the authors should concentrate on the frame and also clearly state how delamination is modeled (what they already did).
iii) the same (as ii) is with table 1 and with chapter 2.4 no need for that (well known), a suitable reference is fine.
iv) it should be clearly noted where the sensor (only 53th node is not suitable, geometric information are required) is located.
the authors should also give information about the the environmental temperature and state (if possible) that temperature is equal between the different experimental measurements.
the claim: identify/classify delaminiation can not be justified in this way. the authors consider two cases (with and without change). the claim can only be: a change can be detected. from the research it can not be concluded that this change results from a delamination.
the delamination claim can only be justified if other failures are also existent for training and test and then beside others the delamination case can be clearly identified as what is is.
sorry for that ... but this must be corrected.
Author Response
Please see the attachment
Author Response File: 
Author Response.docx
Reviewer 2 Report
1. An experiment on the actual composite structure may be carried out further to verify the proposed method.
2. Does the measuring distance to the delamination influence on the diagnosed result?
Author Response
Please see the attachment
Author Response File: Author Response.docx
