A Piece-Wise Linear Model-Based Algorithm for the Identification of Nonlinear Models in Real-World Applications

Round 1

Reviewer 1 Report

 

This paper deals with an interesting topic "A piecewise linear model based algorithm for the identification of nonlinear models in real-world applications" Generally, the topic of this research is interesting, however, major revision is required. In addition, my specific comments are shown below:

 

1-               In page 2 from the sentence" In this work, the k-means 49 clustering [17] and optimization algorithm have been combined to identifying....", which optimization algorithm is used in the paper? the name of the optimization algorithm must be mentioned.

2-               please, number the equations of lines 123, 125, 134, 136, and 137

3-               What is the paper novelty?

4-               It could be interesting to summarize the commented literature works in a table to have a clear comparison between all. This could also help precisely formulating the contribution of the paper with respect to previous works.

5-               I recommend the authors improve the presentation of the results "Figure 1. Normalized Mean Absolute Error for ARX model ", "Figure 2. Correlation Coefficient for ARX... ", "Figure 3. Hit Ratio for ARX model... ", .....

6-               No comparison with other previous works !!, the results should be compared with the most recent methods in the literature

7-               Why are the results images not colored? I recommend the authors to colorize the images of the results such as " Figure 11. Boxplot of coordinate...", "Figure 12. Boxplot of coordinate... ", " Figure 13...", etc.

8-               Please improve the conclusion section to support the results and add suggestions for future works.

9-               I recommend the authors rewrite the references according to the journal style " [18] Nocedal, J.;Wright, S.J. Numerical...., [19] INEMAR - Arpa Lombardia...

10-           The authors may enrich their references with the latest and related work further

 

·       Jian ZhangJian ZhangKwai-Sang ChinKwai-Sang ChinMaciej ŁawryńczukMaciej Ławryńczuk, "Nonlinear model predictive control based on piecewise linear Hammerstein models ", Springer, May 2018, Nonlinear Dynamics 92(3), DOI: 10.1007/s11071-018-4105-5

 

·       Zain-Aldeen S. A. Rahman, Basil H. Jasim, Yasir I. A. Al-Yasir, Raed A. Abd-Alhameed and Bilal Naji Alhasnawi,  "A New No Equilibrium Fractional Order Chaotic System, Dynamical Investigation, Synchronization, and Its Digital Implementation ", Inventions 2021, 6(3), 49; https://doi.org/10.3390/inventions6030049

 

 

·       Zeineb Lassoued, Kamel Abderrahim, " Identification and control of nonlinear systems using PieceWise Auto-Regressive eXogenous models", Transactions of the Institute of Measurement and Control 41(2), DOI: 10.1177/0142331219846239

 

 

·       Zain-Aldeen SA Rahman, Basil H Jasim, Yasir IA Al-Yasir, Yim-Fun Hu, Raed A Abd-Alhameed, Bilal Naji Alhasnawi, "A New Fractional-Order Chaotic System with Its Analysis, Synchronization, and Circuit Realization for Secure Communication Applications", Mathematics 2021, 9(20), 2593; https://doi.org/10.3390/math9202593

Author Response

The authors kindly thank the reviewer for the comments, allowing to greatly improve the original manuscript.

Please see the attachment for the answers.

Best regards

Author Response File: Author Response.pdf

Reviewer 2 Report

Author has proposed a data-driven approach for the identification of piecewise linear model for nitrogen oxides daily concentration simulation. It’s quite interesting to link with the environmental problem. It’s a good trial, even though the model seems simple to me. I have several suggestions for authors.

1. I think Section 2 needs to be extended, including the idea to apply such model.

2. The contribution of this work should be highlighted at the end of Section 1.

3. Related work seems too short for me. I have several suggestions for authors to explore the link between nature and human:

(1) https://doi.org/10.1007/s11554-020-01062-y

(2) https://doi.org/10.1186/s13638-020-01689-2

(3) https://doi.org/10.3390/computation10050076

4. All related figures could be merged into one figure for representation.

5. Experimental analysis should be added.

6. Some sentences could be written with better version.

 

Author Response

The authors kindly thank the reviewer for the comments, allowing to greatly improve the original manuscript.

Please see the attachment for the answers.

Best regards

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have made many efforts to improve the quality of the paper. Please address the two following concerns before possible publication:

1- supporte the  abstract by the results

2- supporte the  conclusions  by the results

 

Author Response

Dear,

Thank you very much for your work. 

Your comments have been considered in the new version of the paper (in green in the last version).

Best regards

The authors

Reviewer 2 Report

I think right now this file is good in status. Congratulations to authors!

Author Response

Dear,

Thank you so much for your time.

Best regards,

Claudio Carnevale