The Effect of Voltage Dataset Selection on the Accuracy of Entropy-Based Capacity Estimation Methods for Lithium-Ion Batteries

Round 1

Reviewer 1 Report

General Remarks

·        In which way were measurement errors being taken into account?

·        Why was LFP chosen? – should mention, that this system is difficult to measure otherwise, because of its flat charge/discharge profile

Positive

·        Well structured paper, good explanations given for most parts

·        Interesting field of study

·        Suggested to be accepted with minor corrections

Suggested Changes

·        Line 26: include EVs (currently one of the largest markets, even when grid is potentially larger)

·        Line28-31: SOH helps prolong service life, but only if a BMS is involved; capacity loss measurement is an issue, but the question here is, if time resolution is really such a big issue, especially for grid applications

·        Line34-35: the “long time needed” for capacity measurement is not a problem – can happen during service life (and often does, depending on the BMS); capacity estimation via current accumulation during charge/discharge is basically the method of choice for nearly every “practical application”

·        Line 38-39: data driven estimation also needs current and voltage measurement – so would include all the challenges of the conventional estimation

·        Line 48: definition of “SVM”

·        Line 60-61: according to this, one of the most basic parameters (voltage) has not yet been included in any model before? Slightly hard to believe – or better, expand on what “adequately” means in this context

·        Line 75-76: parameters and actual profile of the “primary frequency regulation mission profile” should be described (perhaps in supplementary info)

·        Table1: why is the lower cut off voltage 2V? seems too low, even for LFP and in following diagrams, it doesn’t look like the cell has been discharged to that level; can graphite support 28C? (70A)

·        Figure 2: y axis is not in % -> rename

·        Figure 3: the overall process of how many cycles have been done – difficult to understand, perhaps define better within the text

·        Figure 4: are charging and discharging correctly labelled here? Why is charge a negative current?

·        Line 173-174: which datasets are lost? Is that an error to be considered?

·        Line 200: where do the values 3, 0.04, 30 come from?

·        Line 238: does this mean, that the cell needs to be fully charged or discharged anyway, in order for this method to work? What are the main advantages of this method then?

·        Line 281: author contribution: what has Jinhao Meng done within this work? Seems to be missing

Comments for author File: Comments.pdf

Author Response

Thank you very much for the review of our manuscript. Please see the attachment for our responses to the reviewer’s comments. 

Author Response File: Author Response.docx

Reviewer 2 Report

Very nice manuscript. The results are important. The presentation is clear. 

Author Response

Thank you very much for your encouragement and for taking the time and effort necessary to review the manuscript.