Fault Location and Restoration of Microgrids via Particle Swarm Optimization
Round 1
Reviewer 1 Report
This paper presents an integrated fault location and restoration approach for microgrids using particle swarm optimization (PSO). Numerical simulations have been performed. The paper is well organized. The following are my comments.
- The introduction should be revised and the paragraph should be written in a separate paragraph as it is long.
- Edit equation (4) with imaginary numbers in the reactive power and why use current to check for convergence? Voltage is used to check the convergence in backward /forward sweep power flow.
- Explain in detail the fault location optimization methods, what are the objective function, the decision variables, and the constraints?
- Discuss the time response and how to use the proposed methods in real-time.
- Remove the unnecessary words on page 15.
- Add more recent references and work on microgrid fault location and restoration, such as:
[a] Early Identification and Location of Short-circuit Fault in Grid-connected AC Microgrid. IEEE Transactions on Smart Grid. 2021 Mar 17.
DOI: https://doi.org/10.1109/TSG.2021.3066803
[b] Fault detection and location in a microgrid using mathematical morphology and recursive least square methods. International Journal of Electrical Power & Energy Systems. 2018 Nov 1;102:324-31. DOI: https://doi.org/10.1016/j.ijepes.2018.04.009
[c] Fault Detection, Classification, and Location by Static Switch in Microgrids Using Wavelet Transform and Taguchi-Based Artificial Neural Network. IEEE Systems Journal 2019, 1–11. DOI: https://doi.org/10.1109/JSYST.2019.2925594
[d] Intelligent Fault Classification and Location Identification Method for Microgrids Using Discrete Orthonormal Stockwell Transform-Based Optimized Multi-Kernel Extreme Learning Machine. Energies. 2019; 12(23):4504.
DOI: https://doi.org/10.3390/en12234504
[e] Advanced fault location strategy for modern power distribution systems based on phase and sequence components and the minimum fault reactance concept. Electric Power Systems Research 2016, 140, 933–941. DOI: https://doi.org/10.1016/j.epsr.2016.04.008
Author Response
Dear Reviewer:
Thanks for your valuable comments on our paper, entitled "Fault Location and Restoration of Microgrids via Particle Swam Optimization". This paper has been revised as possible as we could according to your comments, as follows. Additionally, we have carefully checked the spell and grammar of our manuscript.
We have rewritten some sentences to explain and correct some errors of this paper; the changes are highlighted in the manuscript by the “Track Changes” function in Microsoft Word, as shown as follows.
Comment 1: The introduction should be revised and the paragraph should be written in a separate paragraph as it is long.
Answer: we have divided the introduction into five paragraphs, from line 26~94.
Comment 2: Edit equation (4) with imaginary numbers in the reactive power and why use current to check for convergence? Voltage is used to check the convergence in backward /forward sweep power flow.
Answer: We have revised equation (4) in page 4. The bus voltage is usually used to check the convergence in power flow solution method, but in this paper, the power demand or generation at each bus is calculated as the equivalent bus injection current during the iteration procedure by equation (4), and then the voltage derivation of each bus is computed by equation (5) and to update the bus voltage; therefore, it is direct to check for convergence by injection current instead of bus voltage.
Comment 3: Explain in detail the fault location optimization methods, what are the objective function, the decision variables, and the constraints?
Answer: we have explained in line 235~236, in page 7. The explanation is show as the following: “In this study, PSO [28, 29] is used to search for the faulted bus, and the assumed faulted bus is modeled as a particle whose objective function is shown in Equation (16). The fault will be detected according to the minimal value of the voltage deviation in the VEM. The proposed fault location approach of the MG is shown in Figure 3.”
Comment 4: Discuss the time response and how to use the proposed methods in real-time.
Answer: the proposed method in real-time application is described in line 206~213, in page 6, as follows:” In real-time application, the voltage derivation ( ) at bus j is caused by the fault current contributed by source bus i, which represents the upstream utility grid or DG. The value of the measured voltage ( ) can be obtained by the synchronized voltage and current measurement device with millisecond level response time installed in each fault current contributed source from the energy management system of the MG, and m in Equation (14) represents the number of fault current sources. For an n-bus MG, the VEM can be established, as shown in Equation (15). The smallest value of the element in the VEM denotes the short-circuit faulted bus.”
Besides, the time response of the proposed method is explained in line 415~418, in page 17, as follows: “The proposed approach is coded using Matlab and executed on a Windows 10 Intel® Core™ i7-10700 CPU @4 GHz personal computer, and the average computing time of each scenario of the IEEE 37-bus is less than 1 second.”
Comment 5: Remove the unnecessary words on page 15.
Answer: we have removed that.
Comment 6: Add more recent references and work on microgrid fault location and restoration, such as:
[a] Early Identification and Location of Short-circuit Fault in Grid-connected AC Microgrid. IEEE Transactions on Smart Grid. 2021 Mar 17. DOI: https://doi.org/10.1109/TSG.2021.3066803
[b] Fault detection and location in a microgrid using mathematical morphology and recursive least square methods. International Journal of Electrical Power & Energy Systems. 2018 Nov 1;102:324-31. DOI: https://doi.org/10.1016/j.ijepes.2018.04.009
[c] Fault Detection, Classification, and Location by Static Switch in Microgrids Using Wavelet Transform and Taguchi-Based Artificial Neural Network. IEEE Systems Journal 2019, 1–11. DOI: https://doi.org/10.1109/JSYST.2019.2925594
[d] Intelligent Fault Classification and Location Identification Method for Microgrids Using Discrete Orthonormal Stockwell Transform-Based Optimized Multi-Kernel Extreme Learning Machine. Energies. 2019; 12(23):4504. DOI: https://doi.org/10.3390/en12234504
[e] Advanced fault location strategy for modern power distribution systems based on phase and sequence components and the minimum fault reactance concept. Electric Power Systems Research 2016, 140, 933–941. DOI: https://doi.org/10.1016/j.epsr.2016.04.008
Answer: We have added more recent reference about fault location and restoration in line 49~93, in page 1~2.
Yours sincerely,
Wei-Tzer Huang
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper presents an interesting application of Optimization Algorithm for fault location identification and restoration. The writing style of the paper should be improved and some other major concerns should be addressed:
- The Introduction does not provide any background on Evolutionary Algorithm. It is important providing a general overview on the most important algorithm (GA https://doi.org/10.1016/j.jestch.2017.11.006, BBO https://doi.org/10.1016/j.eswa.2017.12.039, COA 10.1109/ACCESS.2020.2978398, and SNO https://doi.org/10.1002/int.22515)
- The entire Introduction is composed by only one paragraph. Dividing the text in paragraphs helps the reader
- The novelty content of the paper should be described at the end of the Introduction
- Improve the quality of equations (brackets, spaces,…)
- No details are provided regarding the optimization process (convergence curves, population size, number of iterations). See as example 10.1109/TIE.2017.2756599
Author Response
Dear Reviewer:
Thanks for your valuable comments on our paper, entitled "Fault Location and Restoration of Microgrids via Particle Swam Optimization". This paper has been revised as possible as we could according to your comments, as follows. Additionally, we have carefully checked the spell and grammar of our manuscript.
We have rewritten some sentences to explain and correct some errors of this paper; the changes are highlighted in the manuscript by the “Track Changes” function in Microsoft Word, as shown as follows.
Comment 1: The Introduction does not provide any background on Evolutionary Algorithm. It is important providing a general overview on the most important algorithm (GA https://doi.org/10.1016/j.jestch.2017.11.006, BBO https://doi.org/10.1016/j.eswa.2017.12.039, COA 10.1109/ACCESS.2020.2978398, and SNO https://doi.org/10.1002/int.22515)
Answer: We have added the general overview about the algorithms in the 4th paragraph in line 84~93, in page 2. As follows:” Swarm intelligence (SI) is the collective behavior of decentralized, self-organized systems which usually inspired by nature such as biological systems. SI has been used for wide range of engineering applications including power system for its simple rules and being able to solve any type of problems. Ahmed et al. [22] discussed the advantage and disadvantage of genetic algorithm (GA), artificial bee colony (ABC), etc. for fault location estimation. Li et al. [22] proposed a combined biogeography-based optimization with population competition algorithm method to plan the substation location. Abdelwanis et al. [23] used coyote optimization algorithm to estimate the parameters of transformer. Niccolai et al. [24] used the evolutionary algorithms to optimize the electric vehicles charging station deployment.”
Comment 2: The entire Introduction is composed by only one paragraph. Dividing the text in paragraphs helps the reader
Answer: We have divided the introduction into five paragraphs, from line 26~106.
Comment 3: The novelty content of the paper should be described at the end of the Introduction.
Answer: We have written some sentences to describe the novelty at the end of the introduction, in line 95~101, in page 2. As follows:” In this paper, the network connection matrices are used to form the system topology rapidly before and after the occurrence of fault, and the voltage error matrix is computed to search for the fault section by using PSO instead of brute force. After the allocation of the fault section, the multi-objective function that is also implemented by PSO for optimal restoration with its constraints. Consequently, an integrated fault location and restoration approach for MGs is proposed to effectively solve the FDIR problem in MGs.”
Comment 4: Improve the quality of equations (brackets, spaces, …)
Answer: We have improved that.
Comment 5: No details are provided regarding the optimization process (convergence curves, population size, number of iterations). See as example 10.1109/TIE.2017.2756599
Answer: we have added the convergence characteristics and illustration in figure 7 and in line 326~328; in addition to, the related parameters of PSO are set to n = 500, wmax = 0.9, wmin = 0.2, c1 = 2, and c2 = 2; the maximum iteration number is 200(in line 270~276, in page 9).
Yours sincerely,
Wei-Tzer Huang
Author Response File: Author Response.pdf
Reviewer 3 Report
The paper describes an integrated fault location and restoration algorithm for microgrids. The paper is organized in two sections. In the first, the fault location algorithm is described, in the second section, the restoration algorithm is proposed.
It is necessary to clarify which are the original parts of the paper with respect to what is cited in the bibliography.
About the first section, it is necessary to define each symbol used in the equations of the model.
About equation (4), it is necessary to declare that at the iteration 0, Vk is set to 1 in per unit.
The identification of the fault location depends on the equation (14) where the errors about the voltage magnitude are computed comparing the measured values with the voltages given by the equation. Does the accuracy of the method depend on the number of nodes where the voltage is measured and their location?
About the restoration algorithm, can the authors discuss how the performance of the algorithm depend on the parameter of the PSO algorithm?
Author Response
Dear Reviewer:
Thanks for your valuable comments on our paper, entitled "Fault Location and Restoration of Microgrids via Particle Swam Optimization". This paper has been revised as possible as we could according to your comments, as follows. Additionally, we have carefully checked the spell and grammar of our manuscript.
We have rewritten some sentences to explain and correct some errors of this paper; the changes are highlighted in the manuscript by the “Track Changes” function in Microsoft Word, as shown as follows.
Comment 1: It is necessary to clarify which are the original parts of the paper with respect to what is cited in the bibliography.
Answer: We have marked the original text of the bibliography with double quotes and italic style, in line 26~27, in page 1.
Comment 2: About the first section, it is necessary to define each symbol used in the equations of the model.
Answer: We have added the definitions of some symbols in the content.
Comment 3: About equation (4), it is necessary to declare that at the iteration 0, Vk is set to 1 in per unit.
Answer: We have declared that in line 164~165, in page 5.
Comment 4: The identification of the fault location depends on the equation (14) where the errors about the voltage magnitude are computed comparing the measured values with the voltages given by the equation. Does the accuracy of the method depend on the number of nodes where the voltage is measured and their location?
Answer: Yes, but the number of nodes in a microgrid is relatively small, so it is enough accurate according the simulation results. We have explained in line 218~221, in page 6. As follows: “Additionally, the computing time dependents on the bus number of the system; fortunately, it is not the problem of MGs, which the system is relatively smaller then a distribution feeder, such as a partial feeder MG or a community MG.”
Comment 5: About the restoration algorithm, can the authors discuss how the performance of the algorithm depend on the parameter of the PSO algorithm?
Answer: we have added the convergence characteristics and illustration in figure 7 and in line 326~328; in addition to, the related parameters of PSO are set to n = 500, wmax = 0.9, wmin = 0.2, c1 = 2, and c2 = 2; the maximum iteration number is 200(in line 270~276, in page 9).
Yours sincerely,
Wei-Tzer Huang
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Thank you for addressing my concerns
