Estimation of the Hot Swap Circulation Current of a Multiple Parallel Lithium Battery System with an Artificial Neural Network Model
Round 1
Reviewer 1 Report
The paper deals with an artificial neural network (ANN) to face the reliability issue of a battery pack.
1)Starting to reading the paper I start thinking the author may refer to what is usually named as "Battery Management system" (BMS) even if this keyword is totally not considered along with the paper.
On the contrary, the authors refer to "energy storage systems" (ESSs). May the authors explain the difference (maybe also in the paper as BMS is much more popular)?
I believe that from the reliability/safety-critical features, the topic is similar if not the same.
I suggest mentioning and differentiate ESSs and BMS and/or highlight the same/similar reliability issues to make your paper more appealing to those who work in the same field.
Please look at the several papers from MDPI on the same topic such as Technology (i.e. https://doi.org/10.3390/technologies9020028), Energy (https://doi.org/10.3390/en4111840), and Applied Science ( https://doi.org/10.3390/app8040534). Same for
Electronics ( https://doi.org/10.3390/electronics10030293 , https://doi.org/10.3390/electronics9030510, https://doi.org/10.3390/electronics10060705
https://doi.org/10.3390/electronics10030293).
Also in an Electronic special issue (https://www.mdpi.com/journal/electronics/special_issues/Battery_Management_System) or also other IEEE paper (i.e. 10.1109/EMCEurope.2019.8872061 , 10.1109/ISEMC.2015.7256257)
the importance of the precise charge in any condition is shown.
2)Moreover, please refer to those paper shown similar investigation of your paper. The comparison with the state of the art should be considered more in detail.
Notice that no limit of pages in MDPI publication, so a more consistent reference makes your paper more valuable.
3) I believe that the presence of a Hot-Swap Circulation Current is critical itself and to be completely avoided for safety- reasons in any case instead of being estimated. Please comment on this.
4) I believe that the importance of proper charge equalization in any operating condition prioritizes deterministic and real-time monitoring.
instead of a prevision method. The key point I would highlight is still safety. As far as I read, a precise charge in terms of less than tens of mV should be guaranteed in any condition otherwise the battery pack could also explode. Please comment on this also along with the paper.
5) Please in the Legenda in Fig.11b reports "current" while the y-axis is voltage. I notice you used Exp and Sim as abbreviations of experimental and simulated. I suggest removing "Exp" which is misleading. It's better to use extensively the word "Measured" and "Simulated" or "Meas. results" and "Sim. results).
Please wisely and carefully look at all your figure and respective legenda and caption.
6)Please list in proper acronyms list the several acronyms used along with the paper at the end of the paper (according to the MDPI format).
Apart from the above-mentioned points to be addressed, the paper looks solid and I appreciate the match between model estimation and measurement results.
Author Response
Thank you very much for your kind comments on my paper.
Please check the attached file for answers to the reviewer's questions and corrections to the paper.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This paper presents the estimation of hot swap circulation current of parallel LIBs. This topic is interesting. The paper is generally well written with sufficient analysis. It can be considered for acceptance if the authors can address the following issues.
- The authors can reconsider the title. Is it more proper to use “estimation” or “modeling” of the current?
- The methods depend largely on the equivalent circuit model in use. So it is advised to add some brief review about the ECMs and their associated battery management. Some useful works, e.g., DOI: 10.1109/TIE.2021.3063968; DOI: 10.1109/TII.2020.3047687; Journal of Power Sources 489 (2021): 229462.
- The load current estimation can be also achieved by ECM-based state observer, e.g., DOI: 10.1109/TPEL.2021.3068725. This very relevant work can be discussed briefly to improve the explanation.
- The calibration of ECM and the modeling accuracy should be explained more clearly. The parameter tunning of ANN is critical to the estimation performance, so it should be also explained.
- The conclusion can be improved. Typically, primary results with statistical data should be given.
Author Response
Thank you very much for your kind comments on my paper.
Please check the attached file for answers to the reviewer's questions and corrections to the paper.
Author Response File: Author Response.pdf
Reviewer 3 Report
Journal: Electronics (ISSN 2079-9292) Manuscript ID: electronics-1243503 Title: Estimation of the Hot Swap Circulation Current Estimation of a Multiple Parallel Lithium Battery System with Applying ANN (an Artificial Neural Network) Model In this paper, the authors studied the parallel battery configuration, temperature, voltage deviation, and load current, and conducted experiments to validate the simulation model. The design of the hot-swap algorithm could reflect the results of the case study analysis in a look-up table model, but this can only be used to turn simple hot-swap operations on and off. However, estimation of the circulating current generated during the hot-swap operation is useful for power management algorithms before, during, and after the hot-swap operation. Therefore, the circulating current generated during the hot-swap operation was estimated by applying the fitting network (Fitnet) model, an artificial neural network (ANN), to algorithms that estimate the SOC and state of health (SOH) of batteries. The ANN model was trained with data generated in the simulation-based hot-swap analysis, and the ensuing circulating current was expressed as the resulting value. The performance accuracy of the hot-swap circulating current estimation for the 1S4P lithium battery pack using the ANN model was confirmed to be 94% based on the experimental results. The research is of great interest. The following manuscript has some weakness and here are the most important topics/questions to be dealt with: 1. Please rewrite the abstract by identifying the purpose, the problem, the methodology and the important results (not all) and conclusions of your work. 2. Introduction The Introduction should consist of five paragraphs answering the following five questions: What is the problem? Why is it interesting and important? Why is it hard? Why hasn't it been solved before? (Or, what's wrong with previously proposed solutions?) What are the key components of my approach and results? As the above-mentioned questions should be replied to, it will be better that write more relation to the benefits & disadvantages of the blending techniques and more investigate about limitation of previous studies. Also, this part needs more explanations to state clearly the objectives & hypothesis of this study at the end of the Introduction part. It should be mentioned to the factors that be shed light by this study. 3. Please improve the resolution of Figure 3. 4. Please specify input data to modelling in this study. 5. Due to the fact that the work concerns experimental measurements, the reviewer did not find the reply to the question: what error formula was used to estimate the tests? 6. Conclusion section is extremely short. The conclusions are very weak and ít requires a deeper analysis of the results. 7. What is the next step in the research activity? Please provide some information about further work in the Conclusions section. The reviewer suggests carefully read the whole manuscript again before resubmitting it to the journal Energies. Authors should consider the above-mentioned remarks in order to revise the manuscript. The reviewer thinks that a publication of the draft manuscript may be possible after a minor revision.Author Response
Thank you very much for your kind comments on my paper.
Please check the attached file for answers to the reviewer's questions and corrections to the paper.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The work has been improved as requested
Reviewer 2 Report
This paper can be accepted.
