Next Article in Journal
Experimental Study on Cooling Performance of a Hybrid Microchannel and Jet Impingement Heat Sink
Previous Article in Journal
Influence of Biosynthesized Nanoparticles Addition and Fibre Content on the Mechanical and Moisture Absorption Behaviour of Natural Fibre Composite
 
 
Article
Peer-Review Record

Wind Speed Forecasting with a Clustering-Based Deep Learning Model

Appl. Sci. 2022, 12(24), 13031; https://doi.org/10.3390/app122413031
by Fuat Kosanoglu
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Reviewer 3:
Appl. Sci. 2022, 12(24), 13031; https://doi.org/10.3390/app122413031
Submission received: 14 October 2022 / Revised: 10 November 2022 / Accepted: 13 December 2022 / Published: 19 December 2022

Round 1

Reviewer 1 Report

Check that the style of writing is in the third person throughout. Don’t use ‘we’.

The abstract should provide an accurate synopsis of the paper, thus revising it thoroughly. Describe what a novel is

In figure 1, what’s the unit of speed? Label the designation of X axis as time (hours). Same for figure 2 and 7.

Is figure 3 original?

Figure 4 is not attractive; make it more illustrative which assists readers to understand the overall framework. You refer to following article titled “A machine learning approach for vibration-based multipoint tool insert health prediction on vertical machining centre (VMC)”.

The authors have provided results considering the 80:20 ratio of training:testing. Additionally, results must be provided considering different holdout % and holdout validation approaches. Refer to this article to understand the holdout validation approach. The title is “A Bayesian Optimized Discriminant Analysis Model for Condition Monitoring of Face Milling Cutter Using Vibration Datasets”.

Justify why deep learning algorithm is preferred over machine learning when data is limited. Refer to the following articles Multi-Point Face Milling Tool Condition Monitoring Through Vibration Spectrogram and LSTM-Autoencoder, Cutting Tool Condition Monitoring using a Deep Learning-based Artificial Neural Network

Was the data normalized/ standardized?

What’s the limitation of this method?

What’s the scope of transfer learning? Mention in future scope.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear authors

The paper presents wind speed forecasting with a clustering-based deep learning model. However, there are some comments for the authors. The comments are:

1.     The units are not indicated in Figure 2.

2.     The references require reordering through the manuscript.

3.     Also, there are many references, in the MDPI journals, related to the subject of wind energy and may be helpful in this research such as:

·       https://www.mdpi.com/2071-1050/14/8/4775.

·       https://www.mdpi.com/2227-9717/7/2/85. Or any other recent references.

4.     Please indicate the difference between MAE, MAPE, and RMSE. Why do you use all these measures simultaneously?

5.     The resolution of Figure 6 needs improvement.

6.     The units are not indicated in Figure 7. Also, its caption requires rewording.

7.     Too many language errors are presented such as: (Please review all the manuscript carefully)

·       Line 16 “demand on” replace by “demand for

·       Line 23 “to decrease” replace by “to a decrease”

·       Line 23 “and increase in reliability replace by “and an increase in the reliability”

·       Line 33 " and difficult" replace by " and are difficult"

·       Line 38 " due to the ability of learning " replace by " due to their ability to learn "

 

 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

Dear authors,

The paper entitled "Wind Speed Forecasting with a Clustering Based Deep Learning Model” deals with a machine learning methods to forecast wind speed. The topic of the paper is of interest for the journal “Applied Sciences” of MDPI editor. The paper is well written and could be accepted after some minor revisions as follows:

 

-          The introduction is well organized but i would suggest to enrich the bibliography about other studies of machine learning approaches in wind forecasting such as:

1.       Brusca et al. (2019). Wind farm power forecasting: New algorithms with simplified mathematical structure, E3S Web of Conferences, Volume 2191, Article number 020028.

2.       Shabbir et al. (2022). Short-Term Wind Energy Forecasting Using Deep Learning-Based Predictive Analytics, Computers, Materials and Continua, Volume 72(1), pages 1017-1033.

3.       Praveena et al. (2018). Wind power forecasting in short-term using fuzzy K-Means clustering and neural network, Proceedings of IEEE International Conference on Intelligent Computing and Communication for Smart World, I2C2SW 2018, pages 336-339.

-          The equations are not in the same format and should be better referred;

-          The units of measure should me checked and be in the international system and each figure should have them on the axis.

-          The english form is generally good but a deeper proofreading is necessare since many missprints are still present

-          Figures are too little, especially figure 6 and 7. I would suggest to enlarge them and consider a better quality.

-          Results are properly motivated but I would suggest to enrich this paragraph since it is the most important of the paper.

Kind Regards

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have addressed my comments, and the paper can be accepted.

Reviewer 2 Report

The manuscript has been improved. Thanks

Back to TopTop