Adaptive Neural Network for a Stabilizing Shunt Active Power Filter in Distorted Weak Grids
, Milad Mansouri 1, Sajjad Chaharmahali 1, Mohammad H. Moradi 1 and Mohammad Sajjad Tahriri 3
Round 1
Reviewer 1 Report
This research proposes a stable shunt active power filter (SAPF). The adaptive linear neuron (ADALINE) is used to extract the harmonics and obtain the current reference where a robust synchronizer is employed for synchronization operation. The ADALINE-based shunt active power filter has been proposed before by many researchers. The reviewer think that section 2.5 titled, "Robust Synchronizer Quasi-Static Filter (RSQSF) PLL Model For WEAK-GRID Conditions", should be more elaborate as the main idea of this research. The margin stability analysis should be added as well as the controller parameter design section. The results with MATLAB Simulink for this topology is considered as not sufficient. The experimental results are needed to validate the proposed idea, especially with the dynamic response such as the change in electrical load. HIL experiment is also suggested to verify the stable operation performance of the proposed controller.
Author Response
Thank you for your effective comment. The authors have prepared the necessary answers in the attached file.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Weaknesses in the study have been addressed in references 19 through 24; however, it is not clear how each flaw was substantiated and why it is necessary to address these difficulties in these studies.
Section 2.3 of your framework should include a discussion of the ADALENE module, in which you should explain why it is more practicable.
It's necessary to have a 75:25 ratio of recent to old references because 18 of your 24 references are more than a decade old.
Author Response
Thank you for your effective comment. The authors have prepared the necessary answers in the attached file.
Author Response File: Author Response.pdf
Reviewer 3 Report
This paper proposes the development of SAPF. This paper uses the ANN and one of its sub-methods,ADALINE, to extract the harmonics and obtain the current reference. So the research and methods are modern and actual. The novelty is on medium but acceptable level. Research is based on simulation done in Matlab. The research topic is in scope of Applied Sciences.
My recommendations to improve paper:
-> In the abstract, the results should be described in a quantitative way. Please revise.
-> The literature review should be extended with 6 new (2021-2022 year) positions to reach 30 in total. Also some present papers from Applied Sciences should be included to confirm that paper is in scope of journal.
-> Between 2. System Configuration and 2.1. THE OVERALL STRUCTURE OF THE SAPF please introduce in general way section 2.
-> for equations presented in section 2. System Configuration please provide adequate references.
-> please also discuss in section 2 why ANN, ADALINE are suitable to solve indicated problem for SAPF.
-> please add details about software in section 3. SIMULATION RESULTS - version etc.
-> results presented in section 3 should be discussed in context of secondary literature.
-> conclusions should be extended with limitations and future research dirrections.
Author Response
Thank you for your effective comment. The authors have prepared the necessary answers in the attached file.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
After revision paper is suitable to publication.
best regards,
Michał Jasiński
