Next Article in Journal
Biometric-Based Key Generation and User Authentication Using Voice Password Images and Neural Fuzzy Extractor
Previous Article in Journal
Securing Cyber Physical Systems: Lightweight Industrial Internet of Things Authentication (LI2A) for Critical Infrastructure and Manufacturing
 
 
Article
Peer-Review Record

A Multivariate Machine Learning Approach for the Prediction of Wind Turbine Blade Structural Dynamics

Appl. Syst. Innov. 2025, 8(1), 12; https://doi.org/10.3390/asi8010012
by Amr Ismaiel
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Appl. Syst. Innov. 2025, 8(1), 12; https://doi.org/10.3390/asi8010012
Submission received: 25 November 2024 / Revised: 7 January 2025 / Accepted: 15 January 2025 / Published: 16 January 2025

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Having read this paper, I believe that it is suitable for publication after minor revision of the presentation.

1. Section 2.2. What formula was used for the Pearson correlation? Many readers will know, but this paper appears to be aimed at a broad audience and a few words of explanation would improve the legibility of the text.

2. Caption of Figure 4. The caption should contain a brief explanation of what the audience is looking at. The figures themselves are nearly incomprehensible.

3. In table 2,  the training time is given as zero for many cases. This is likely due to the limitations of a 4-digit format. Please adjust this to reflect the actual training time.

4. Also in table 2, it would be good to have an extra column that shows the accuracy in percentage.

Finally, as a more general question: The authors assume a mean wind speed of 12 m/s. Have they made any attempts to change this and, if so, does it fundamentally change the result?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This work focuses on the prediction of wind turbine blade structural dynamics. In the paper, an approach using ML models is presented to predict the structural dynamics of a wind turbine blade based on the flow conditions and the turbine’s control actions. Three datasets are generated for the NREL 5 MW wind turbine using OpenFAST tool for aeroelastic analysis under different turbulence classes.

The work of the paper is meaningful and helps to provide useful references for the prediction of wind turbine blade structural dynamics. The paper work still needs some improvement as follows:

1. The paper is mainly based on the historical data of the selected wind turbines, but in the model construction part, the difficulties of predicting the blade tip reflections and root shear forces in the flare and edgewise directions should be analyzed, but cannot just focus on historical data.

2. In the paper, ten different ML models including linear, nonlinear, and ensemble models are trained to predict the blade tip deflections and root shear forces in the flapwise and edgewise directions. Applying ten models for analysis and validation is a significant workload, which is highly commendable. However, the analysis of the characteristics, advantages, and disadvantages of the ten models in the article is not yet in-depth enough.

3. There are many curves in Figure 4 that cannot be identified. It is recommended to provide clearer graphics.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The paper has been carefully revised and responded to the review comments, and the original issues have been resolved.

Back to TopTop