Application of Deep Reinforcement Learning to Predict Shaft Deformation Considering Hull Deformation of Medium-Sized Oil/Chemical Tanker
Round 1
Reviewer 1 Report
The work is very interesting - it extends the existing research in this field. The tested vessel is equipped with a relatively low-power engine. In the future, tests should be repeated on a drive system with an engine power of 40,000 kW.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This paper deals with important topic of the shaft line alignment. It is generally nicely written paper at high technical level. The main challenge that authors resolve is the influence of the stiffness of the aft part of the ship on the alignment results. Authors use deep reinforcement learning as a tool to find out bearing reactions or deformations.
I have following recommendations to improve the quality of the paper and clarity of presentation:
- It is not completely clear what are the input variables to the problem and what are the output. I could not understand even after repeated reading of the paper if authors measure deformations and then find out reaction forces or vice versa. It would be good to clarify that in the introduction.
- In table 5, results of reaction forces are presented for different ship drafts. If these reactions are measured, I don't see the purpose of their calculations again with the DRL. This should also be clarified.
- The purpose of the inverse FEM is not completely clear. Please again write exactly what is the input and what is expected from such inverse calculations.
- Table 11 - It is not clear in the paper what is the purpose of the fluid density. Is it used for calculation of bearing stiffness or ...?
- Please clarify the meaning of the cases in Fig16-21. The reader will be confused if it is not described.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
