batteries

Article
Generalized Characterization Methodology for
Performance Modelling of Lithium-Ion Batteries
Daniel-Ioan Stroe *, Maciej Swierczynski, Ana-Irina Stroe and Søren Knudsen Kær
Department of Energy Technology, Aalborg University, 9220 Aalborg East, Denmark; mas@et.aau.dk (M.S.);
ast@et.aau.dk (A.-I.S.); skk@et.aau.dk (S.K.K.)
* Correspondence: dis@et.aau.dk; Tel.: +45-30-62-25-89

Academic Editor: Sheng S. Zhang

Received: 30 September 2016; Accepted: 22 November 2016; Published: 1 December 2016

Abstract: Lithium-ion (Li-ion) batteries are complex energy storage devices with their performance
behavior highly dependent on the operating conditions (i.e., temperature, load current, and
state-of-charge (SOC)). Thus, in order to evaluate their techno-economic viability for a certain
application, detailed information about Li-ion battery performance behavior becomes necessary.
This paper proposes a comprehensive seven-step methodology for laboratory characterization of
Li-ion batteries, in which the battery’s performance parameters (i.e., capacity, open-circuit voltage
(OCV), and impedance) are determined and their dependence on the operating conditions are
obtained. Furthermore, this paper proposes a novel hybrid procedure for parameterizing the batteries’
equivalent electrical circuit (EEC), which is used to emulate the batteries’ dynamic behavior. Based
on this novel parameterization procedure, the performance model of the studied Li-ion battery is
developed and its accuracy is successfully verified (maximum error lower than 5% and a mean error
below 8.5 mV) for various load profiles (including a real application profile), thus validating the
proposed seven-step characterization methodology.

Keywords: lithium-ion (Li-ion) battery; characterization; methodology; performance modelling;

electrochemical impedance spectroscopy; DC pulses; validation

1. Introduction
Recent developments of lithium-ion (Li-ion) batteries based on new and improved chemistries
have resulted in batteries with high performance, long lifetime and increased safety [1,2]. Thus,
Li-ion batteries have become the key energy storage technology for e-mobility applications [3,4].
Furthermore, energy storage systems based on Li-ion batteries are evaluated in different projects where
they are used in grid ancillary service applications [5–7] or for renewables’ grid integration [6,8,9].
Consequently, Li-ion batteries are expected to become the major player in utility-scale applications as
stated in different surveys [10,11]. Nevertheless, Li-ion batteries are characterized by a high investment
cost in comparison to other energy storage technologies (e.g., lead acid and nickel metal hydride
(NiMH) batteries) and their performance behavior is highly influenced by the operating conditions:
temperature, load current and state-of-charge (SOC). Thus, in order to gain from the aforementioned
advantages, the Li-ion batteries have to be operated in an efficient and cost-effective manner.
This objective can be achieved by relying on accurate performance models, which are able to
estimate accurately the dynamic behavior of Li-ion batteries independently of the operating conditions;
thus, precise energy management strategies can be defined and accurate sizing of the Li-ion batteries
based systems can be realized. Moreover, by relying on accurate performance models, Li-ion batteries
can be tested by simulations considering different scenarios, thereby reducing the laboratory testing
efforts that are usually cost demanding and time consuming.

Batteries 2016, 2, 37; doi:10.3390/batteries2040037 www.mdpi.com/journal/batteries

Batteries 2016, 2, 37 2 of 20

Depending on the modelling approach followed, Li-ion battery performance models are divided
into three main categories: electrochemical models [12,13], mathematical models [14,15], and
electrical models [16–18]; furthermore, combinations between these three modelling approaches
were proposed [19,20]. These approaches are characterized by different accuracy levels and degrees of
complexity. Electrical models represent the most suitable solutions for Li-ion battery modelling when
they need to be integrated in system-level simulations because of their features such as high accuracy
and average computation complexity [21]. Thus, an electrical-based performance modelling approach
is detailed in this paper.
Electrical models use an equivalent electrical circuit (EEC), composed of combinations of resistors,
capacitors, inductors, constant phase elements etc., to describe the performance (dynamic) behavior
of Li-ion batteries. Two methods are traditionally used to parametrize the elements of the EEC.
The first method involves applying a DC pulse to the Li-ion battery and measuring the voltage response
of the battery [16,22–24]. The DC pulse-based electrical models, also referred as Thévenin-based
electrical models, use a series resistance and several resistance-capacitance (RC) parallel networks to
predict the response of the Li-ion battery to transient load events. The accuracy of the Thévenin-based
models depends on the number of RC networks used to estimate the voltage of the Li-ion battery
(i.e., the higher the number of RC parallel networks, the better the accuracy of the model), as shown
in [22,24]. For a thorough parameterization of the EEC, the DC pulses have to be applied at different
SOCs, load currents, and temperatures since the parameters of the EEC are highly dependent on the
operating conditions as illustrated in [16,25,26]. One of the disadvantages of the DC pulse technique is
represented by the fact that high charging and discharging current pulses cannot be applied at high and
low SOC levels, respectively, for low temperatures since the maximum and minimum battery voltage
values will be reached. Consequently, a full parameterization of the EEC is not possible. Furthermore,
the elements of the EEC, which are determined with this method, mostly lack physicochemical
meaning. All these disadvantages are overcome when the second method based on the electrochemical
impedance spectroscopy (EIS) technique is used for parameterization of the EEC [27]. EIS has become
in the last decade an established technique for characterizing Li-ion batteries and for modelling and
parameterizing their performance behavior [17,18,28,29]. The models developed with this technique,
referred to as impedance-based models, have the advantage that their EEC’s elements can be easily
related with physicochemical processes that occur inside the battery, such as charge transfer and
diffusion [28,30,31]. In order to parameterize an impedance-based model, EIS measurements have to
be performed at various battery SOC levels and temperatures. Furthermore, it is possible to perform
EIS measurements with superimposed DC current, in order to obtain the dependence of the EEC
elements on the load current [29]; however, these measurements can be performed only for small
currents (reduced C-rates) because, since the EIS measurements are time demanding, they will result
in changing the Li-ion battery’s SOC. Thus, the impedance-based performance model will not be able
to predict accurately the voltage of the Li-ion batteries for high C-rates, which are characteristic for
Li-ion batteries used in many e-mobility and grid applications.
This paper proposes a new approach to parameterize the EEC, which combines the advantages
of the two traditional methods. Therefore, the proposed hybrid method uses the EIS technique to
parameterize the EEC and express the dependence of its elements on the SOC and temperature and the
DC pulse technique to express their dependence on the load current. Furthermore, a physicochemical
interpretation of the EEC elements and their corresponding values can be realized. Nevertheless,
building and parameterizing a performance model for a Li-ion battery is a complex process and
an extensive laboratory characterization of the targeted battery has to be performed in order to obtain
information about various parameters such as: capacity, open-circuit voltage (OCV) etc. and their
dependence on the operating conditions such as: SOC, temperature, and load current. Thus, besides
proposing a new approach of parameterizing the EEC, this paper presents a comprehensive seven-step
methodology for characterizing and modelling the performance behavior of a Li-ion battery. The
present paper is structured as follows. Section 2 gives brief information about the Li-ion battery used
Batteries 2016, 2, 37 3 of 20

in this work and presents the experimental set-up used for characterizing the Li-ion battery. Section 3
presents the proposed seven-step methodology for characterizing and parameterizing the performance
model of the studied Li-ion battery. Based on the obtained results from the characterization
Batteries 2016, 2, 37 3 of 19
test,
the performance model of the Li-ion battery is developed in Section 4. In Section 5, extensive results
obtainedLi-ion
frombattery. Section 3 presents
the validation the proposed
of the proposed seven-step model
performance methodology for characterizing
are presented, and
while concluding
parameterizing the
remarks are given in Section 6. performance model of the studied Li-ion battery. Based on the obtained results
from the characterization test, the performance model of the Li-ion battery is developed in Section 4.
In Section 5,
2. Experimental extensive results obtained from the validation of the proposed performance model are
Set-Up
presented, while concluding remarks are given in Section 6.
At present, a wide variety of Li-ion battery chemistries are available on the market that are,
depending2. Experimental Set-Up
on their characteristics, suitable for a broad range of applications from electro-mobility
to back-up At power anda wide
present, renewables’
variety ofgrid integration
Li-ion [32]. Nevertheless,
battery chemistries are available onnone of these
the market thatchemistries
are,
depending on their characteristics, suitable for a broad range of applications from
is superior to the others from all perspectives (e.g., energy density, specific energy, cost per electro-mobility to cycle,
back-up power and renewables’ grid integration [32]. Nevertheless, none
etc.) [33]. Therefore, the characterization methodology presented in this work and the proposed of these chemistries is
superior to the others from all perspectives (e.g., energy density, specific energy, cost per cycle, etc.) [33].
performance model can be applied independently of the type of the selected Li-ion battery chemistry.
Therefore, the characterization methodology presented in this work and the proposed performance
For exemplifying
model can be theapplied
proposed characterization
independently methodology,
of the type a commercial
of the selected Li-ion battery2.5 chemistry.
Ah batteryFor based on
the lithium iron phosphate/graphite
exemplifying (further methodology,
the proposed characterization referred to asa commercial
LFP/C), is2.5 used in this based
Ah battery work.on the
lithium
During iron
the phosphate/graphite
whole characterization (further referred
process, thetobattery
as LFP/C),
wasisplaced
used ininside
this work.
a temperature-controlled
chamber, as illustrated in Figure 1, and its temperature was continuouslyinside
During the whole characterization process, the battery was placed a temperature-
monitored using a PT100
controlled chamber, as illustrated in Figure 1, and its temperature was continuously monitored using
sensor. Furthermore, the temperature of the chamber was adjusted in order to obtain the desired
a PT100 sensor. Furthermore, the temperature of the chamber was adjusted in order to obtain the
temperature on the surface of the LFP/C battery; thus, the temperature values mentioned in this work
desired temperature on the surface of the LFP/C battery; thus, the temperature values mentioned in
are the ones
this workthe
of arebattery
the ones and
of thenot the ambient
battery and not thetemperature values of
ambient temperature the chamber.
values of the chamber.

+ EIS
+ Vdc
PT100
+ Idc
- Vdc

- EIS
- Idc

Figure 1. Lithium iron phosphate/graphite (LFP/C) battery during the characterization procedure. EIS:
Figure 1. Lithium iron phosphate/graphite (LFP/C) battery during the characterization procedure.
electrochemical impedance spectroscopy.
EIS: electrochemical impedance spectroscopy.
3. Lithium-Ion Battery Characterization Methodology
3. Lithium-Ion
The Battery Characterization
main objective Methodology
of the proposed characterization methodology was to measure the
Theperformance parameters of the studied battery and to determine their dependence on the operating
main objective of the proposed characterization methodology was to measure the performance
conditions; the operating conditions cover the sum of the conditions given by the load current,
parameters of the studied battery and to determine their dependence on the operating conditions;
temperature, and SOC. Based on the obtained results, which are presented throughout this section,
the operating conditions
the performance cover
model theLFP/C
of the sum battery
of the was
conditions given
developed by the load current,
and parameterized temperature,
(see Section 4). and
SOC. BasedInonorder
the obtained
to determineresults, which arebehavior
the performance presented throughout
of the tested LFP/Cthis section,
battery, the the performance
seven-step
model ofmethodology,
the LFP/Cwhich batteryis was
presented in Figure
developed and 2, parameterized
was proposed. This(seemethodology
Section 4). is composed of
In order to determine the performance behavior of the tested in
various tests, which were carried out following the sequence illustrated Figure 2;
LFP/C the firstthe
battery, six tests
seven-step
are presented throughout the next subsections, while the last step (i.e., verification test) is extensively
methodology, which is presented in Figure 2, was proposed. This methodology is composed of various
discussed in Section 5.
tests, which were carried out following the sequence illustrated in Figure 2; the first six tests are
presented throughout the next subsections, while the last step (i.e., verification test) is extensively
discussed in Section 5.
Batteries 2016, 2, 37 4 of 20
Batteries 2016, 2, 37 4 of 19

Batteries 2016, 2, 37 4 of 19

Figure 2. Test sequence for characterization of the LFP/C battery. OCV: open-circuit voltage; SOC:
Figure 2. Test sequence for characterization of the LFP/C battery. OCV: open-circuit voltage;
state-of-charge; and HPPC: hybrid pulse power characterization.
SOC: Figure
state-of-charge; and HPPC:
2. Test sequence hybrid pulse of
for characterization power characterization.
the LFP/C battery. OCV: open-circuit voltage; SOC:
state-of-charge; Test
3.1. Preconditioning and HPPC: hybrid pulse power characterization.
3.1. Preconditioning Test
The objective of
3.1. Preconditioning the preconditioning test was to remove any possible passivation to which the
Test
The objective
battery of the preconditioning
was subjected to, between the test was to remove
manufacturing time any
and possible
the initialpassivation to which
tests. Moreover, an the
battery was The
additional objective
subjected
goal of to,of between
this the
testpreconditioning
was to thestabilize test
manufacturing was to
the battery remove
time any
and the
capacity since possible
initial passivation
tests.
the solid Moreover,
electrolyte tointerface
which the
an additional
of
battery
the was subjected
batteries is not to, between
shaped after the manufacturing
the manufacturing process
time andand theitsinitial tests. Moreover,
porosity/shape changesan
goal of this test was to stabilize the battery capacity since the solid electrolyte interface of the batteries
additional goalafter
of this
thetest wasfew to stabilize the battery capacity since thetest solid electrolyte interface of
is notmeaningfully
shaped
the
after isthenot
batteries
first
manufacturing
shaped after
cycles.
process
the
Thus,
and the
manufacturing
preconditioning
its porosity/shape
process and
changes
its
was composed
meaningfully
porosity/shape
of five
after the
changes
successive
first few cycles. charge-discharge
Thus, thethe cycles performed
preconditioning waswith
testThus, 1C-rate of
composed (i.e.,
five2.5 A) at 25 °C
successive [21]. The resultscycles
meaningfully
obtained for theafter
case of the first few cycles.
discharging capacity the
of the preconditioning
tested battery aretest wascharge-discharge
presented composed
in Figure 3.of five
performed withcharge-discharge
1C-rate (i.e., 2.5cycles A) at 25 ◦ C [21]. The results obtained for the case of
successive performed with 1C-rate (i.e., 2.5 A) at 25 °C [21]. the Thedischarging
results
capacity of thefor
obtained tested battery
the case of theare 2.6presented
discharging in Figure
capacity 3. tested battery are presented in Figure 3.
of the

2.6
2.58
[Ah] [Ah]

2.58
2.56
Capacity

2.56
2.54
Capacity

2.54
2.52

2.52
2.5
0 1 2 3 4 5 6
Measurement Number
2.5
0 1 2 3 4 5 6
Figure 3. Discharging capacity of the LFP/C battery measured during the preconditioning test (1C-
Measurement Number
rate, T = 25 °C).
Figure 3. Discharging capacity of the LFP/C battery measured during the preconditioning test (1C-
Figure 3. Discharging capacity of the LFP/C battery measured during the preconditioning test (1C-rate,
rate,
The T = 25 °C).
◦ measured capacity of the LFP/C battery was stable, with a tendency of slight monotonic
T = 25 C).
increase of approximately 0.6% during the five performed cycles; nevertheless, the battery was
The measured
considered capacitysince
preconditioned of the
its LFP/C battery
capacity was
did not stable,more
change withthan
a tendency of slight
3% during monotonic
two consecutive
The measured
increase of capacity
approximately
discharges [34]. of the
0.6% LFP/C
during battery
the five was stable,
performed with
cycles; a tendency
nevertheless, of slight
the monotonic
battery was
considered
increase preconditioned
of approximately since
0.6% its capacity
during did not
the five change more
performed thannevertheless,
cycles; 3% during twotheconsecutive
battery was
discharges
considered
3.2. [34].
preconditioned
Relaxation Test since its capacity did not change more than 3% during two consecutive
discharges [34]. the behavior of the Li-ion batteries is altered by the concentration gradients of the ionic
Usually,
3.2. Relaxation Test
charge carriers immediately after the load current is switched off [35]. Consequently, a relaxation
3.2. Relaxation
Usually,Test behavior of the Li-ion batteries is altered by the concentration gradients of the ionic
period, which the will allow the battery to reach thermodynamic stability, has to be applied between the
charge
switch carriers
off of the immediately
load current afterthe
the load current is switched off [35]. Consequently, a relaxation
Usually, the behavior of theand
Li-iondesired
batteriesmeasurement.
is altered by In the
order to determine
concentration the optimal
gradients length
of the ionic
period,
of this which
relaxation will allow
period, the
the battery
behavior to reach
of the thermodynamic
OCV of the LFP/C stability,
battery has
was to be applied
investigated; between
the batterythe
charge carriers
switch off
immediately
of the load current
after
and
the
the
load
desired
current is
measurement.
switched
In
off
order
[35].
to
Consequently,
determine the optimal
a relaxation
length
was fully charged and then discharged at three different SOCs (i.e., 80%, 50% and 20%), where it was
period,
of which
this will allow thethebattery to reach
of thethermodynamic LFP/Cstability, hasinvestigated;
to be applied thebetween
battery the
kept at relaxation
OCV conditionsperiod, behavior
for 24 h, respectively OCV
[21].ofThe
the voltage battery
values wasmeasured, with one-second
switchwasofffully
of the load current
charged and then and the desired
discharged at measurement.
three different SOCs In(i.e.,
order80%,to50%
determine
and 20%),thewhere
optimal it waslength
resolution, during the relaxation period were related to the OCV value measured after 24 h (when the
of this relaxation
kept at OCV period,
conditionsthe behavior
for 24 h, of the
respectivelyOCV of
[21]. the
The LFP/C
voltage battery
values was investigated;
measured,
battery was considered quasi-stabilized); thus, the OCV error was computed according to Equation (1). with the
one-second battery
resolution,
was fully charged during
and the
thenrelaxation
dischargedperiod at were
threerelated to the
different OCV(i.e.,
SOCs value80%,measured
50% andafter20%),
24 h (when
wherethe it was
battery
kept at OCVwas consideredfor
conditions quasi-stabilized);
24 h, respectivelythus, the OCVThe
[21]. error was computed
voltage values according
measured, to Equation (1).
with one-second
resolution, during the relaxation period were related to the OCV value measured after 24 h (when the
battery was considered quasi-stabilized); thus, the OCV error was computed according to Equation (1).
Batteries 2016, 2, 37 5 of 20

|Vi − OCV |
εOCV = · 100 (1)
OCV
where εOCV represents the OCV error and Vi represents the voltage measured at every second during
the 24 hBatteries
relaxation.
2016, 2, 37 5 of 19
Based on the obtained results, which are summarized in Table 1, it was concluded that
the OCV value is only slightly influenced byV a− OCV relaxation time, which varies between 15 min
i
ε OCV hour was ⋅considered
= 100
and 24 h. Thus, a relaxation time of one enough for the LFP/C (1) battery
OCV
to reach quasi-thermodynamic stability and to allow for accurate measurement of the desired
where εOCV represents the OCV error and Vi represents the voltage measured at every second during
performance parameters.
the 24 h relaxation.
Based on the obtained results, which are summarized in Table 1, it was concluded that the OCV
Table 1. isVoltage
value measured
only slightly at 20%,
influenced by a50%,
relaxation time,SOC
and 80% after
which various
varies relaxation
between time
15 min and 24 periods
h. Thus, and
a
corresponding OCVoferrors.
relaxation time one hour was considered enough for the LFP/C battery to reach quasi-
thermodynamic stability and to allow for accurate measurement of the desired performance
parameters.
Relaxation SOC = 20% SOC = 50% SOC = 80%
Time Voltage (V) OCV Error (%) Voltage (V) OCV Error (%) Voltage (V) OCV Error (%)
Table 1. Voltage measured at 20%, 50%, and 80% SOC after various relaxation time periods and
1s 3.157 2.80 3.230 2.03 3.270 2.00
corresponding OCV errors.
15 min 3.236 0.37 3.292 0.15 3.325 0.37
1 h Relaxation 3.241 0.23
SOC = 20% 3.295 SOC = 50% 0.07 3.335
SOC = 80% 0.07
2h Time 3.243 Voltage (V) OCV 0.17Error (%) 3.295 (V) OCV Error
Voltage 0.05(%) Voltage 3.336 0.04
(V) OCV Error (%)
24 h 1s 3.248 3.157 0 2.80 3.297
3.230 2.030 3.2703.337 2.00 0
15 min 3.236 0.37 3.292 0.15 3.325 0.37
1h 3.241 0.23 3.295 0.07 3.335 0.07
3.3. Capacity Test
2h 3.243 0.17 3.295 0.05 3.336 0.04
24 h 3.248 0 3.297 0 3.337 0 applied load
The capacity of Li-ion batteries is dependent on the operating temperature and on the
current [15,17]. Moreover, as presented by various researchers, the capacity of Li-ion batteries degrades
3.3. Capacity Test
over time [36–38]; however, this analysis is out of the scope of the present paper. The objective of the
The capacity of Li-ion batteries is dependent on the operating temperature and on the applied
capacity test was to determine the dependence of the charged and discharged capacity of the LFP/C
load current [15,17]. Moreover, as presented by various researchers, the capacity of Li-ion batteries
battery degrades
on the two overaforementioned parameters.
time [36–38]; however, Thus,
this analysis is the
out capacity
of the scope of of
thethe
battery
presentwas measured
paper. The for
six different C-rates ◦
objective of the(i.e., C/4,
capacity testC/2, 1C,
was to 2C, 3C the
determine anddependence
4C) and at offour different
the charged and temperatures
discharged capacity(i.e., 15 C,
25 ◦ C, 35of◦the
C andLFP/C ◦ C), following
45battery on the twothe procedure which
aforementioned is presented
parameters. Thus, theincapacity
[21]. of the battery was
Themeasured
dependence for sixofdifferent
the LFP/CC-rates (i.e., C/4,discharging
battery’s C/2, 1C, 2C, 3C and 4C) and
capacity on theat four different
C-rate and temperatures
on the temperature
(i.e., 15 °C, 25 °C, 35 °C and 45 °C), following the procedure
is illustrated in Figure 4. The discharged capacity of the LFP/C battery measured which is presented in [21].
at 25 ◦ C decreases as
The dependence of the LFP/C battery’s discharging capacity on the C-rate and on the
the applied current increases; however, the decrease of the capacity for the considered C-rate interval
temperature is illustrated in Figure 4. The discharged capacity of the LFP/C battery measured at 25
(C/4–4C) °Cis lower than
decreases as the1%. Thiscurrent
applied behavior suggests
increases; a Peukert
however, number
the decrease very
of the closefortothe
capacity 1, considered
which is in good
agreement C-ratewith the results
interval (C/4–4C)reported for a1%.
is lower than similar batterysuggests
This behavior chemistry by Omar
a Peukert numberet al. [39].
very Furthermore,
close to 1,
as presented
whichin is Figure
in good 4, for the considered
agreement temperature
with the results interval,
reported for a similara battery
slight increase
chemistryof bythe
OmarLFP/C
et al. battery
capacity[39].
with Furthermore,
increasingastemperature
presented in Figure 4, for the considered
was measured, which mighttemperature interval,
have been a slight
caused byincrease
the increased
of the LFP/C battery capacity with increasing temperature was measured, which might have been
electronic and ionic conductivity in the electrode and electrolyte at high temperatures; similar results
caused by the increased electronic and ionic conductivity in the electrode and electrolyte at high
were reported in [17,39,40].
temperatures; similar results were reported in [17,39,40].

Figure 4. Dependence of the discharged capacity on temperature and load current.

Figure 4. Dependence of the discharged capacity on temperature and load current.
Batteries 2016, 2, 37 6 of 20

Batteries 2016, 2, 37 6 of 19
3.4. Open-Circuit Voltage versus State-of-Charge Test
Batteries 2016, 2, 37 6 of 19
3.4. Open-Circuit
Similar to most Voltage versus State-of-Charge
of the Li-ion Test
battery parameters, the OCV is strongly dependent on the operating
conditions
3.4. Open-Circuit temperature,
Voltage SOC,
Similar to most of the Li-ion battery parameters, thegoal
(i.e., versus etc.). Therefore,
State-of-Charge Test the OCVofisthis
stronglytest was to measure
dependent on the the OCV-SOC
operating
characteristic
conditions
Similar of tothe
(i.e.,
most LFP/C
temperature, battery
SOC,
of the Li-ion at different
etc.).
battery Therefore,temperatures
parameters, the
thegoal
OCVofisthis for
test both
strongly was tocharging
measure
dependent and
on the
the discharging
OCV-SOC
operating
characteristic
conditions; the OCV of the
was LFP/C battery
measured in at
stepsdifferent
of 5% temperatures
SOC following for both
the charging
methodology
conditions (i.e., temperature, SOC, etc.). Therefore, the goal of this test was to measure the OCV-SOC and discharging
described in [21].
conditions;
For instance, the OCV
the was
OCV-SOC measured in steps
characteristic of 5%
of the SOC following
tested battery, the methodology
measured at 25 ◦ C, is presented
described in [21]. in
characteristic of the LFP/C battery at different temperatures for both charging and discharging
Figure For instance,
5. For the
conditions; the
themeasured OCV-SOC characteristic
characteristic,
OCV was measured in steps of the
a hysteresis tested battery,
effect is observed,
of 5% SOC following measured at 25 °C,
which described
the methodology is presented
is more pronounced
in [21].
in Figure 5. For thethe
for the For instance,
10%–40% SOCmeasured
OCV-SOC
interval. characteristic,
characteristic
This behavior a hysteresis
of
ofthe effect
thetested
OCV is observed,
battery,
is which
measured
intrinsic for atis25more
Li-ion °C, pronounced
is presented
batteries based on
forFigure
in the 10%–40%
5. For theSOC interval. This behavior
measured of the OCV isisintrinsic for Li-ionisbatteries based on
active materials, which performcharacteristic,
a two-phaseatransition
hysteresis effect
lithium observed,
insertionwhich more pronounced
and extraction process, as is
active
for the materials,
10%–40% which perform This
SOC interval. a two-phase
behavior transition
ofeffect
the OCVlithium insertion
is intrinsic and
for extraction
Li-ion process,
batteries basedasonis
the case
the
of the
case of
LFP
the
material
LFP
[41].
material
The
[41]. The
hysteresis
hysteresis effect
ismainly
is
mainly influenced
influenced by
by
the
the previous
previous
history
history
active materials, which perform a two-phase transition lithium insertion and extraction process, as is
(e.g., (e.g.,
charging/discharging
charging/discharging C-rate and relaxation time)ofofthe the LFP/C battery and can be modelled
the case of the LFP materialC-rate andhysteresis
[41]. The relaxationeffect
time) is mainlyLFP/C battery
influenced byand
the can be modelled
previous history
following various
following approaches
various approaches reported
reportedininliteratures
literatures [42,43].
[42,43].
(e.g., charging/discharging C-rate and relaxation time) of the LFP/C battery and can be modelled
following various approaches reported
3.6 in literatures [42,43].

3.6
3.4

3.4
[V] [V]

3.2
Voltage

3.2
3
Voltage

3
2.8
Charging
2.8 Discharging
2.6 Charging
0 20 40 60 80 100
State-of-Charge [%] Discharging
2.6
0 20 40 60 80 100
State-of-Charge
Figure 5. The OCV-SOC characteristic [%] battery measured at 25 °C.
of the LFP/C
Figure 5. The OCV-SOC characteristic of the LFP/C battery measured at 25 ◦ C.
Figure 5. The OCV-SOC characteristic of the LFP/C battery measured at 25 °C.
3.5. Hybrid Pulse Power Characterization Test
3.5. Hybrid Pulse Power Characterization Test
3.5. Hybrid Pulse Power
The impedance Characterization
represents Test
a very important performance parameter of the Li-ion batteries since
The impedance
it defines
The impedance represents
the power capability
represents a avery
of
very important
the performance
battery. Moreover,
important performance it is parameter
the parameter
parameter of the
of the which
Li-ionLi-ion batteries
describes
batteries thesince
since
dynamic
it defines behavior of Li-ion batteries. As most of the Li-ion battery parameters,
it defines the power capability of the battery. Moreover, it is the parameter which describes the the
the power capability of the battery. Moreover, it is the parameter the impedance
which shows
describes
a non-linear
dynamic
dynamicbehavior behavior
behaviorof Li-ion depending
of Li-ionbatteries.
batteries. on
AsAsthe operating
most
most of the conditions
theLi-ion
Li-ion battery
battery such as SOC,the
parameters,
parameters, temperature,
the impedance
impedance C-rate
shows shows
[44,45].
a non-linear Furthermore,
a non-linear
behaviorbehavior the
depending battery
depending on theimpedance
on operating is
the operatinghighly dependent
conditions
conditions such suchon the
as SOC,
as SOC, battery’s state-of-health
temperature,
temperature, C-rate
C-rate [44,45].
(SOH) [44];
[44,45]. however, thisbattery
Furthermore, is out of the scopeisofhighly the present work.on In the
order to account for all the
Furthermore, the battery the impedanceimpedance is highly dependent dependent
on the battery’s battery’s state-of-health
state-of-health (SOH) [44];
aforementioned
(SOH) [44]; dependences,
however, this is outtheof impedance
the scope of ofthethe LFP/C
present battery
work. In was
order measured
to account following
for all thea
however, this is out of the scope of the present work. In order to account for all the aforementioned
modified version of the hybrid pulse power characterization (HPPC)
aforementioned dependences, the impedance of the LFP/C battery was measured following a test. The measurement profile
dependences, the
the impedance
applied toversion LFP/C of thepresented
LFP/C battery was measured
consistsfollowing a modified version
and of
modified of thebattery
hybrid is pulse power in Figure
characterization6 and (HPPC) of consecutive
test. The measurement charging
profile
the hybrid pulse
discharging power
DCLFP/C characterization
current pulsesisofpresented (HPPC)
different in test.
C-rates The measurement profile applied to the LFP/C
applied to the battery Figure(i.e.,
6 and0.1C,consists
0.5C, 1C, 2C, 3C and 4C);
of consecutive the pulse
charging and
battery is presented
duration
discharging wasDC in18Figure
setcurrent
to s [34]
pulses6 and
and ofthe consists
relaxation
different oftime
C-ratesconsecutive
between
(i.e., charging
0.1C,each
0.5C,two1C, and3Cdischarging
consecutive
2C, andpulses theDC
4C); was setcurrent
pulseto
pulses
duration was set to 18 s [34] and the relaxation time between each two consecutive pulses was set tos [34]
15 of different
min. C-rates (i.e., 0.1C, 0.5C, 1C, 2C, 3C and 4C); the pulse duration was set to 18
and the relaxation time between each two consecutive pulses was set to 15 min.
15 min. 4C
4C

3C 4C
43CC
Discharge

2C 3C
32CC
Discharge

18 s 1C 2C
21CC
0.5C
0.1C
C-rateC-rate

18 s 1C
1 C0
0.5C
-0.1C 0.1C
-0.5C
-1 0C
-1C
-0.1C 15 min
-0.5C
Charge Charge

-2CC
-1
-1C -2C
15 min
-3CC
-2
-2C -3C

-4CC
-3 Time
-3C -4C

-4 C Time
Figure 6. Pulse profile for measuring the internal resistance of the LFP/C
-4C battery.
Figure 6. Pulse profile for measuring the internal resistance of the LFP/C battery.
Figure 6. Pulse profile for measuring the internal resistance of the LFP/C battery.
Batteries 2016, 2, 37 7 of 20
Batteries 2016, 2, 37 7 of 19
Batteries 2016, 2, 37 7 of 19
Furthermore,ininorder
Furthermore, orderto to consider
consider the dependence
dependenceofofthe theimpedance
impedance ononSOCSOC and andtemperature,
temperature,
thethe test
test profilepresented
Furthermore,
profile presented ininFigure
in order Figure 66 was
to considerwasthe applied to
tothe
dependence
applied the5%–95% SOC
SOCinterval
of the impedance
5%–95% on(considering
interval SOC a 5%
and temperature,
(considering SOC
a 5% SOC
resolution)
the test profileand for
presentedthe same
in temperatures
Figure 6 was (i.e.,
applied 15to °C,
the
◦ 25 °C,
5%–95%
resolution) and for the same temperatures (i.e., 15 C, 25 C, 35 C and 45 C) considered during ◦ 35 °C
SOC ◦ and 45
interval °C) considered
(considering
◦ a during
5% SOC
measurement
resolution) and
measurement ofofthe the other
forother
the sameperformance
temperatures
performance parameters.
(i.e., 15 °C, 25 °C, 35 °C and 45 °C) considered during
parameters.
The typical voltage response of a Li-ion battery
The
measurement typical of voltage
the response
other of
performance a Li-ion batteryto
parameters. toaadischarging
dischargingcurrent current pulse
pulseis is
illustrated
illustrated in in
FigureThe7; atypical
similarvoltagevoltageresponse
behaviorof is aobtained for a charging
Li-ion battery current pulse.
to a discharging currentAs pulse
presented in the ISO
is illustrated in
Figure 7; a similar voltage behavior is obtained for a charging current pulse. As presented in the ISO
12405-1:2011
Figure 7; a similar Standard voltage[34], the charging
behavior is obtainedandfor discharging
a charging impedance
current pulse. of As
thepresented
battery could be
in the ISO
12405-1:2011 Standard [34], the charging and discharging impedance of the battery could be calculated
calculated
12405-1:2011 for Standard
various pulse [34], thelengths (i.e., and
charging 0.1 s,discharging
2 s and 18impedances). Furthermore, the impedances
of the battery could be
for various pulse lengths (i.e., 0.1 s, 2 s and 18 s). Furthermore, the impedances corresponding to
corresponding
calculated for tovarious these pulse
pulselengths
lengthsare caused
(i.e., 0.1 s,by/related
2 s and to18different electrochemical
s). Furthermore, processes
the impedances
these pulse lengths are caused by/related to different electrochemical processes inside the battery [44].
inside the batteryto[44].
corresponding theseNevertheless,
pulse lengths in this
are work,
causedimpedance
by/relatedvalues determined
to different for a current
electrochemical length
processes
Nevertheless,
of 18 s are
in this work, impedance values determined for a current length of 18 s are presented.
inside the presented.
battery [44].Based on the current
Nevertheless, in thisand
work, voltage
impedancemeasurements, the impedance
values determined of the Li-ion
for a current length
Based
of 18on
battery the
s is
are current based
computed
presented. and voltage
Basedonon Ohm’s measurements,
the law, as:and voltage
current the impedance
measurements, of thetheLi-ion batteryofisthe
impedance computed
Li-ion
based on Ohm’s
battery is computed law, as: based on Ohm’s law, as:
∆V
ΔV VV −−V V
Ri R=i = == 11 0 0 (2) (2)
ΔΔIV V1I1−IV
∆I
Ri = = 10
(2)
where RiRrepresents the impedance of the battery, ΔI ∆V represents
I1 the change in the voltage due to the
where i represents the impedance of the battery, ∆V represents the change in the voltage due to the
applied
wherecurrent
applied Rcurrent pulse,
i represents pulse, ∆I
the represents
∆Iimpedance
representsof thethechange
the change
battery, in∆V
in the current,VV
therepresents
current, represents
1 1represents
the change inthe the
the voltage
voltage
voltage measured
measured
due to the
after
after s, s,Vcurrent
1818
applied 0 0represents
V pulse,the
represents ∆I voltage
the measured
voltage measured
represents the change justbefore
just before
in applying
the applying
current, 1the
Vthe current
current
represents pulse,
pulse, andand
the voltage I1measured
represents
I1 represents
thethe
amplitude
18 s, V0 of
amplitude
after ofthetheapplied
represents applied current pulse.
currentmeasured
the voltage pulse. just before applying the current pulse, and I1 represents
the amplitude of the applied current pulse.

Figure
Figure 7.7.Theoretical
Theoreticalvoltage
voltageand
and current
current of
of Li-ion
Li-ion battery
batteryduring
duringimpedance measurement
impedance with
measurement thethe
with
current
Figure pulse
7. technique.
Theoretical voltage and current of Li-ion battery during impedance measurement with the
current pulse technique.
current pulse technique.
The voltage response of the LFP battery to the current profile measured at 50% SOC is illustrated
The voltage
in Figure
The 8. response
voltage ofofthe
response theLFP
LFPbattery
batteryto
to the
the current profile measured
current profile measuredatat50% SOCis is
50%SOC illustrated
illustrated
in Figure 8.
in Figure 8.
3.35
3.35

3.3
3.3
[V] [V]
Voltage

3.25
Voltage

3.25 0.1 C-rate

0.5 C-rate
0.1 C-rate
3.2 1 C-rate
0.5 C-rate
2 C-rate
3.2 1 C-rate
3 C-rate
2 C-rate
4 C-rate
3.15 3 C-rate
0 410
C-rate 20 30 40 50
3.15 Time [s]
0 10 20 30 40 50
Time [s]
Figure 8. Measured voltage response of the LFP battery at discharging current pulses of different C-
rates (SOC
Figure = 50%, T =voltage
8. Measured 25 °C). response of the LFP battery at discharging current pulses of different C-
Figure 8. Measured voltage response of the LFP battery at discharging current pulses of different
rates (SOC = 50%, T = 25 °C).
C-rates (SOC = 50%, T = 25 ◦ C).
Batteries 2016, 2, 37 8 of 20
Batteries 2016, 2, 37 8 of 19
Batteries 2016, 2, 37 8 of 19

Based
Based
Based onon battery’s
on battery’s voltage
battery’s voltage response
voltage response (similar
response (similar to
(similar to those
to those presented
those presented
presented in in Figure
in Figure
Figure 8),8), and
8), and considering
and considering
considering
Equation
Equation (2), the charging and discharging impedance of the tested LFP/C battery was determined
Equation (2),
(2), the
the charging
charging and
and discharging
discharging impedance
impedance of
of the
the tested
tested LFP/C
LFP/C battery
battery was
was determined
determined
for
for all
for all the
all the considered
the considered conditions;
considered conditions; for
conditions; for example,
for example, Figure
Figure 999 presents
example, Figure presents
presents the the measured
the measured impedance
measured impedance
impedance of of the
of the
the
LFP/C
LFP/C during
during pulse
pulse charging
charging and
and discharging
discharging with
with 1C-rate
1C-rate (i.e.,
(i.e., 2.5
2.5 A)
A) at
at TT =
= 25
25 °C.
◦ C.
LFP/C during pulse charging and discharging with 1C-rate (i.e., 2.5 A) at T = 25 °C. Both impedance Both
Both impedance
impedance
characteristics
characteristics
characteristicsareare showing
areshowing
showing aa parabolic
a parabolic dependence
dependence
parabolic on
on the
dependence the
the SOC,
onSOC, SOC, following
following a nearlyaa flat
following nearly flat
region
nearly region
flatbetween
region
between
20% 20% and
20% SOC.
and 90%
between 90% SOC.
and 90% SOC.
0.025
0.025
Charging
Charging
Discharging
0.023 Discharging
0.023
[ Ω] ]

0.021
[Ω

0.021
internal
internal

0.019
0.019
RR

0.017
0.017

0.015
0.0150 20 40 60 80 100
0 20 40 60 80 100
SOC
SOC [%]
[%]
Figure 9.
9. Measured impedance
impedance of the
the LFP/C battery
battery during
during 18
18 sss charging
charging and
and discharging current
current
Figure 9. Measured
Figure Measured impedance of of the LFP/C
LFP/C battery during 18 charging and discharging
discharging current
pulse (1C-rate, T = 25 °C).
◦
pulse (1C-rate, TT == 25
pulse (1C-rate, C).
25 °C).

3.6. Electrochemical
3.6. Impedance Spectroscopy
Electrochemical Impedance Spectroscopy
Spectroscopy TestTest
Test
EIS has
EIS has developed
has developed
developed as as a reliable
as aa reliable method
reliable method
method for for characterization
for characterization
characterization and and modelling
and modelling
modelling the the performance
the performance
performance
behavior of
behavior of the
the Li-ion
the Li-ion batteries.
Li-ion batteries.The
batteries. Thesmall-signal
The small-signalimpedance
small-signal impedance
impedance of
ofofthe Li-ion
thethe Li-ion
Li-ion batteries is
is determined
batteries
batteries is determined
determined by
by
applying
applying
by applyinga small
a small
a small sinusoidal
sinusoidal
sinusoidal current
current (galavantostatic
current(galavantostatic
(galavantostaticmode) mode)
mode)oror voltage
orvoltage (potentiostatic
voltage (potentiostatic mode)
(potentiostatic mode) and and
measuring
measuring the
measuring the amplitude
theamplitude
amplitude and and
and
phasephase
shiftshift
phase of theof
shift the
the output
ofoutput voltagevoltage
output voltage or current,
current, respectively.
or respectively.
or current, respectively. This
This
This procedure
procedure
procedure
is is
repeatedisfor repeated
repeated
a sweepforfor a sweep of
ofafrequencies, frequencies,
sweep of frequencies,
and thus the and thus
andbattery the battery
thus theimpedance impedance
battery impedance spectrum
spectrumspectrum is obtained.
is obtained.
is obtained.
In this
In this work,
work, the the EIS
EIS measurements
measurements were were performed
performed in in galvanostatic
galvanostatic modemode andand for
for the
the frequency
frequency
range
range 10 kHz–10
range 1010kHz–10
kHz–10 mHz. mHz.
mHz. Furthermore,
Furthermore,
Furthermore, all the EIS
all measurements
all the EIS measurements
the EIS measurements
were performed were performed
werewithout
performed without
without
superimposed
superimposed
superimposed
DC current; thus, DC
DCthe current;
current; thus,
thus,
influence the
ofthe influence
the influence of
current onofthe the current
thesmall-signalon
current on thethe small-signal
ACsmall-signal
impedance was AC impedance
AC not
impedance was
was
determined.
not
As determined.
notfor
determined.
the other As As for the other
for the other
considered considered
considered
performance performance
performance
parameters, parameters,
the parameters,
small-signalthe the small-signal
small-signal
impedance impedance
wasimpedance
measured
was
was
at measured
measured
different at
at different
differentand
temperatures temperatures
temperatures
for the entireand for
for the
and SOC the entire
entire SOC
interval SOC interval
interval
(0%–100% SOC(0%–100%
with 5%SOC
(0%–100% SOC
SOC with
with 5%
5% SOC
SOC
resolution).
resolution).
resolution).
The dependenceThe dependence
The dependence of the
of the LFP/Cofbattery’s LFP/C
the LFP/C battery’s impedance
battery’sspectra
impedance impedanceon thespectra
spectra
SOC andon the
on the SOC and temperature
SOC and temperature
temperature is presented
is
is presented
in Figures 10in
presented in Figures
Figures
and 10
10 and
and 11,
11, respectively. 11, respectively.
respectively.
8
8

6
6
ΩΩ
[m
-ImaginaryZZ[m ]]

4
4 5 6 7 8
-Imaginary

5 6 7 8

2
2
SOC=20%
SOC=20%
SOC=50%
0 SOC=50%
0 SOC=80%
SOC=80%
4 6 8 10 12 14
4 6 8 10 12 14
Real Ω
Real Z
Z [m
[mΩ]]
Figure 10. Dependence of the impedance spectra on the SOC; T = 25 ◦°C.
Figure 10.
Figure 10. Dependence
Dependence of
of the
the impedance
impedance spectra
spectra on
on the
theSOC;
SOC; TT == 25
25 °C
C..
Batteries 2016, 2, 37 9 of 20
Batteries 2016, 2, 37 9 of 19

10
Batteries 2016, 2, 37 9 of 19
8

Z [mΩ] Z [mΩ]
10
6
8

-Imaginary
4
6 o
T=15 C
o
2 T=25 C

-Imaginary
4 o
T=35 C
o
0 T=15oC
T=45 C
o
2 T=25 C
4 6 8 10 12 14 16 o 18
T=35 C
Real Z [mΩ] o
0 T=45 C
Figure 11. Dependence of the impedance spectra on the temperature; SOC = 50%.
4 of the
Figure 11. Dependence 6 impedance
8 10 spectra
12 on14the temperature;
16 18 SOC = 50%.
Real Z [mΩ]
4. Lithium-Ion Battery Performance Model
Figure 11.
4. Lithium-Ion Battery Dependence of
Performance the impedance spectra on the temperature; SOC = 50%.
Model
Li-ion battery performance models can be classified into three main categories: electrochemical,
Li-ion battery
electrical, performance
and mathematical
4. Lithium-Ion models
models,
Battery Performance can be classified
depending
Model into threeand
on the development main categories: electrochemical,
parameterization approach.
These modelling approaches are characterized by different degrees of complexity,
electrical, and mathematical models, depending on the development and parameterization different accuracy
approach.
Li-ion battery performance models can be classified into three main categories: electrochemical,
levels and
Theseelectrical,
modelling could be
approaches used
are for different
characterized purposes
by (e.g.,
different battery
degrees design,
of system
complexity, simulation).
different accuracy
and mathematical models, depending on the development and parameterization approach.
levelsNevertheless,
and could bethis work
used forhas focusedpurposes
different on the development
(e.g., anddesign,
battery parameterization
system of the LFP/C
simulation).
These modelling approaches are characterized by different degrees of complexity, different accuracy
battery
Nevertheless,
performance model using an electrical modelling approach.
this work
levelshasandfocused
could onbe the development
used for differentand parameterization
purposes of the
(e.g., battery LFP/C
design, battery
system performance
simulation).
model using an
Nevertheless, electrical
this workmodelling
has focusedapproach.
4.1. Equivalent Electrical Circuit Model on the development and parameterization of the LFP/C battery
performance model using an electrical modelling approach.
Li-ionElectrical
4.1. Equivalent batteries’Circuit
electrical models generally use EECs, composed of simple (e.g., resistors,
Model
capacitors, inductors)
4.1. Equivalent ElectricalorCircuit
more Model
complex elements (e.g., constant phase elements, Warburg elements,
Li-ion
etc.) andbatteries’ electrical
a voltage source models
to express thegenerally use EECs,
dynamic behavior composed
of batteries. of simple
The basic (e.g., of
configuration resistors,
an
capacitors, Li-ion batteries’
inductors) or electrical
more modelselements
complex generally(e.g.,
use constant
EECs, composed
phase of simple Warburg
elements, (e.g., resistors,
elements,
EEC-based performance model is presented in Figure 12. The structure of the EEC can have different
capacitors, inductors) or more complex elements (e.g., constant phase elements, Warburg elements,
etc.) and a voltagedepending
configurations source to onexpress
variousthe dynamic
aspects, such behavior of batteries.
as: model accuracy, modelThe basic configuration
computation time, Li- of
etc.) and a voltage source to express the dynamic behavior of batteries. The basic configuration of an
ion batteryperformance
an EEC-based chemistry, etc.
model is presented in Figure 12. The structure of the EEC can have different
EEC-based performance model is presented in Figure 12. The structure of the EEC can have different
configurations
configurations depending various
depending on aspects,
on various such
aspects, as:as:model
such modelaccuracy,
accuracy, model computationtime,
model computation time,Li-Li-ion
battery
ionchemistry, etc.
battery chemistry, etc.

Figure 12. Basic configuration of an EEC-based performance model for Li-ion batteries.

According to the configuration shown in Figure 12, the voltage of the battery is computed
Figure 12. Basic configuration of an EEC-based performance model for Li-ion batteries.
according to
Figure Equation
12. Basic (3).
configuration of an EEC-based performance model for Li-ion batteries.

According to the configuration shown Vbatin= VOC ± VEEC

Figure 12, the voltage of the battery is computed
(3)
According
according toto the configuration
Equation (3). shown in Figure 12, the voltage of the battery is computed
where Vbat represents the voltage of the battery, VOC represents the OCV, and VEEC represents the
according to Equation (3).
voltage drop across the EEC, which is usedVto = VOC ±the
bat model VEECdynamics of the battery. (3)
Vbat = VOC ± VEEC (3)
The voltage drop across EEC’s impedance is obtained as the sum of various over-voltages, which
where Vbat represents the voltage of the battery, VOC represents the OCV, and VEEC represents the
are caused by different voltage
V bat represents
wherevoltage processesofoccurring insideV OCthe represents
battery with different
OCV, time
and constants during the
V EEC represents
drop acrossthe
the EEC, which the battery,
is used to model the dynamics the of the battery.
charging
voltage drop or discharging [46]:
The across
voltagethe EEC,
drop which
across EEC’sisimpedance
used to model the dynamics
is obtained as the sumofof the battery.
various over-voltages, which
The voltage drop across EEC’s impedance
are caused by different processes occurring V = Vis
EEC inside ± V
obtained
ohmic the ± Vas
diff with different time over-voltages,
chtrbattery the sum of various (4)which
constants during
are caused
chargingbyordifferent processes
discharging [46]: occurring inside the battery with different time constants during
where Vohmic represents the ohmic over-voltage (caused by the resistance of the poles, current
charging or discharging [46]: VEEC = Vohmic the
collectors, electrolytes, etc.), Vchtr represents ± Vchtrcharge
± Vdiff transfer over-voltage (caused (4) by
VEEC = Vohmic ± Vchtr ± Vdiff (4)
where Vohmic represents the ohmic over-voltage (caused by the resistance of the poles, current
V ohmic represents
wherecollectors, the ohmic
electrolytes, etc.), over-voltage (caused
Vchtr represents the by the resistance
charge transfer of the poles, current
over-voltage (causedcollectors,
by
electrolytes, etc.), V chtr represents the charge transfer over-voltage (caused by electrochemical reaction
Batteries 2016, 2, 37 10 of 20

at inner
Batteries 2016, 2, 37and V diff represents the diffusion over-voltage (caused by a deficit or surplus
surfaces), 10 of 19 of
Batteries 2016, 2, 37 10 of 19
reactants at the reaction’s location).
electrochemical reaction at inner surfaces), and Vdiff represents the diffusion over-voltage (caused by
Thus, to model reaction
electrochemical the voltage behavior
at inner of the
andLFP/C battery, the contributions of the V OC andbyof the
a deficit or surplus of reactants atsurfaces),
the reaction’s Vlocation).
diff represents the diffusion over-voltage (caused
V EEC acomponents
deficit
Thus,or to have
surplus to
modelofthe
be known.
reactants
voltageat
The contribution
the reaction’s
behavior location).
of the LFP/C
of the OCV to the battery voltage was obtained
battery, the contributions of the VOC and of the
for various
VEEC Thus,conditions
to model
components (i.e.,
have SOCs
thetovoltage and
be known. temperatures)
behavior of the LFP/Cof
The contribution during
battery, the
the OCV thebattery characterization,
contributions
to the battery of the Vwas
voltage as of
OC and presented
obtainedthe
in Section
V
for 3.4. To
components
EECvarious obtain
havethe
conditions to be
(i.e.,contribution
known.
SOCs andThe of the V EEC
contribution
temperatures) oftothe
during the LFP/C
OCV
the to thebattery
battery voltage,was
battery voltage
characterization, the
as EEC had to
obtained
presented
for various3.4.
be parameterized.
in Section conditions
To obtain (i.e.,
theSOCs and temperatures)
contribution of the VEECduring the battery
to the LFP/C characterization,
battery voltage, the EEC as presented
had to be
in Section
parameterized.3.4. To obtain the contribution of the V to the LFP/C
The parameterization of the EEC was realized by curve fitting the impedance spectra of the
EEC battery voltage, the EEC had to be
LFP/C
parameterized.
The parameterization of the EEC was realized by curve fitting
battery, which were measured at different conditions as presented in Section 3.6. The configuration the impedance spectra of the of
LFP/C Thebattery,
parameterization of measured
the EEC was realized by curve fitting the impedance spectra of the
the EEC which was which
used for were curve fitting,atanddifferent conditions
consequently as presented
to express in Section
the dynamics of3.6.
theTheLFP/C
LFP/C battery, which were measured at different conditions as presented
configuration of the EEC which was used for curve fitting, and consequently to express the dynamics in Section 3.6. The
battery, is illustrated in Figure 13. The curve fitting procedure was based on a complex non-linear
configuration of the EEC
of the LFP/C battery, which wasinused
is illustrated for13.
Figure curve
Thefitting,
curve and consequently
fitting procedure was to express
based the
on adynamics
complex
least-square
of the (CNLS)
LFP/C algorithm,
battery, is whereinboth
illustrated real13.
Figure and
The imaginary
curve components
fitting procedure of the
was measured
based on impedance
a complex
non-linear least-square (CNLS) algorithm, where both real and imaginary components of the
were non-linear
fitted simultaneously
least-squarewere
measured impedance
using
(CNLS) a least square
fitted algorithm,
simultaneously
minimization;
where usingbotha least an
real and example
squareimaginaryof the impedance
components
minimization; an exampleofspectrum
the
of
curvemeasured
fitting results
impedanceis shown
were in Figure
fitted 14.
simultaneously
the impedance spectrum curve fitting results is shown in Figure 14. using a least square minimization; an example of
the impedance spectrum curve fitting results is shown in Figure 14.

Figure 13. Equivalent electrical circuit (EEC) configuration based on ZARC elements used for fitting
Figure 13. Equivalent electrical circuit (EEC) configuration based on ZARC elements used for fitting
Figure 13. Equivalent
the measured electrical
impedance spectracircuit
of the(EEC)
LFP/Cconfiguration
battery. CPE:based on ZARC
constant phase elements
element. used for fitting
the measured impedance spectra of the LFP/C battery. CPE: constant phase element.
the measured impedance spectra of the LFP/C battery. CPE: constant phase element.

6
6
] Ω]
Z [mZΩ[m

4
4
-Imaginary
-Imaginary

2
2

0 Measured spectrum
0 Fitted spectrum
Measured spectrum
5 7 9 Fitted
11 spectrum 13
5 7
Real Z9 [mΩ] 11 13
Real Z [mΩ]
Figure 14. Measured and fitted impedance spectrum using the EEC presented in Figure 13.
Figure 14. Measured and fitted impedance spectrum using the EEC presented in Figure 13.
Figure 14. Measured and fitted impedance spectrum using the EEC presented in Figure 13.
As presented in Figure 14, the EEC configuration based on a series inductance Ls, a series
As presented
resistance Rs, and two in Figure
ZARC 14, the EEC
elements (i.e.,configuration
parallel connections based on of aa series
resistorinductance
and a constant Ls, a phase
series
As presented
resistance
element) in
Rs, and
was ableFigure
two 14, theelements
ZARC
to accurately EEC
fit theconfiguration
(i.e., parallel
measured based on spectrum
connections
impedance a series inductance
of a resistor
of the and
testedaLsconstant
,LFP/C
a series resistance
phase
battery.
element)
Rs , and two ZARC
Therefore, wasthisable to accurately
elements
EEC (i.e., fit
configuration the
wasmeasured
parallel to impedance
connections
used fit all of spectrumand
thea resistor
measured ofimpedance
the tested spectra
a constant LFP/Cphase battery.
(atelement)
the
Therefore,
conditions this EEC
discussed configuration
in Section 3.6) was
and used
thus to to
modelfit all
the the measured
performance
was able to accurately fit the measured impedance spectrum of the tested LFP/C battery. Therefore, impedance
behavior of the spectra
LFP/C (at the
battery.
conditions
this EEC discussed
The configuration
impedance of thewasinselected
Section
used to 3.6)
EEC fitand
is thusmeasured
allgiven
the to model the
by: performance
impedance behavior
spectra of the
(at the LFP/C battery.
conditions discussed
The impedance of the selected EEC is given by:
in Section 3.6) and thus to model the performance Rbehavior of theR2LFP/C battery. The impedance of
Z EEC = jω Ls + Rs + 1
+ (5)
R
the selected EEC is given by: Z = jω L + R + 1 + ( jω )1 Q1 R1 + 1 + ( jω )2 Q2 R2 N1 R N2
EEC s s (5)
1 + ( jω ) 1 Q1 R1 1 + ( jω ) 2 Q2 R2
N N

where R1, Q1, and N1 represent the resistance, theRgeneralized 1 capacitanceR2 and the depression factor
where R1, QZARC
of the first
ZNEEC
1, and element,
= jωL
1 represent
+ resistance,
sthe
respectively;
Rs +
the the explanations
same generalized +capacitance
are N2and the depression
ZARC factor
(5)
1 + ( jω) Q1 R1N1
1 + ( jω)for
valid QtheR
2 2
second of the
of the first ZARC element, respectively; the same explanations are valid for the second ZARC of the
EEC.
EEC.
where R1 ,AQmain
1 , andadvantage
N1 represent of the EIS
the techniquethe
resistance, over other parameterization
generalized capacitancetechniques is represented
and the depression factor of
by A feature
its main advantage
to relate of theelements
EEC EIS technique
to the over other parameterization
physicochemical process thattechniques
occur insideis represented
the battery
the first ZARC element, respectively; the same explanations are valid for the second ZARC of the EEC.
by its feature to
[28,30]. in relate EEC elements
work, thetofirst the term
physicochemical
in Equation process thattooccur inside thethe battery
A main Thus,
advantage the present
of the EIS technique over other (5) is used
parameterization describe
techniques inductive
is represented by
[28,30].
behaviorThus, in the
at high present work,
frequencies, the first
the second termterm is in
used Equation (5) is the
to calculate used to describe caused
over-voltage the inductive
by the
its feature
behaviorto relate
at high EEC elementsthe
frequencies, to the physicochemical
second toprocess that
theoccur inside the battery [28,30].
ohmic resistance (Vohmic), the third term isterm usedistoused calculate calculate over-voltage
the over-voltage caused
corresponding by
to the
the
Thus,ohmic
in theresistance
present work, (Vohmic), the third term is used to calculate the over-voltage corresponding to the at
the first term in Equation (5) is used to describe the inductive behavior
high frequencies, the second term is used to calculate the over-voltage caused by the ohmic resistance
 charge-transfer process (Vchtr), while the last term is used to determine the over-voltage
corresponding to the diffusion process (Vdiff). Consequently, based on the aforementioned aspects,
the battery voltage Equation (3) was rewritten as given in Equation (6). Furthermore, considering
Equation (6), the configuration of the EEC, which was used to model the performance behavior of the
Batteries 2016, 2, 37 11 of 20
LFP/C battery, is presented in Figure 15.
R1 R2
V =V + I ( jω L + R + + ) (6)19
(V ohmic ), the third term isbatused
Batteries 2016, 2, 37 OC bat
to calculate s s
the + ( jω ) Q1 R1 1corresponding
N
1 over-voltage
1
+ ( jω ) Q2 R2
N 2 11 of
to the charge-transfer
process (V chtr ), whileprocess
charge-transfer
the last term is used to determine the over-voltage corresponding to the diffusion
(Vchtr), while the last term is used to determine the over-voltage
process (V diff ). Consequently,
corresponding based
to the diffusion on the
process aforementioned
(Vdiff). Consequently,aspects,
based onthe thebattery voltage Equation
aforementioned aspects, (3)
was rewritten
the battery voltage Equation (3) was rewritten as given in Equation (6). Furthermore, considering of
as given in Equation (6). Furthermore, considering Equation (6), the configuration
the EEC, which
Equation (6),was used to model
the configuration theEEC,
of the performance
which was behavior of the
used to model theLFP/C battery,
performance is presented
behavior of the in
FigureLFP/C
15. battery, is presented in Figure 15.
R R2
Vbat = VOC + Ibat ( jωLs + Rs + R 1 + R2 ) (6)
Vbat = VOC + I bat (jω Ls + Rs +1 + ( jω1)N N1 Q +R 1 + ( jω ) )2Q R
N (6)
1 + ( j ω ) Q R 1 + ( jω ) Q R
1 1 1 N 2 2 2
1 1 2 2

Figure 15. Basic electrical configuration of the performance model of the LFP/C battery.

4.2. Dependence on Load Current

Besides being dependent on SOC and temperature—dependences obtained from the EIS
measurements (see Section 3.6)—the parameters of the EEC are also dependent on the load current
as illustrated in Figure 15. Nevertheless, because all the EIS measurements were performed without
superimposed DC current (see Section 3.6), the effect of the load current on the parameters of the EEC
was not determined. Among the parameters of the EEC, the charge transfer resistance R1 shows a
highlyFigureFigure
non-linear 15. Basic
15. Basic electrical
electrical
dependence configuration
onconfiguration
the of the
of
load current. the performance
performance
This model
non-linearmodelofof
thethe
dependenceLFP/C battery.
LFP/C
of the battery.
charge-transfer
resistance, and of the corresponding over-voltage, on the LFP/C battery current is presented in Figure
4.2. Dependence on Load Current
4.2. Dependence
16. on Load Current
Theoretically,
Besides beingthe exponential
dependent on dependence between the Li-ion battery current
SOC and temperature—dependences andfrom
obtained the potential
the EIS
Besides
can
measurements
being(see
be determined
dependent
from
Section
on SOC
the 3.6)—the
Butler-Volmer
and temperature—dependences
equation
parameters given
of the EEC in are
Equation
obtained from the EIS
(7) [47]. on the load current
also dependent
measurements (see Section 3.6)—the parameters
as illustrated in Figure 15. Nevertheless, because all the of the EEC are also dependent
EIS measurements were performed on the load current
without
 αnF (1−α ) nF 
as illustrated in Figure 15. Nevertheless, because ⋅η
all −
theRTEIS ⋅η measurements were performed without
superimposed DC current (see Section 3.6),
i = i0the effect
 e RT − eof the load current on the parameters of the EEC (7)
superimposed DC current
was not determined. (see Section
Among 3.6), the
the parameters of effect
the EEC, of the the  charge
 load current
transferon the parameters
resistance R1 shows of a the
highly non-linear dependence on the load current. This non-linear
EEC was not determined. Among the parameters of the EEC, the charge transfer resistance R1 showsdependence of the charge-transfer
where i is the
resistance, andelectrode
of current density,
the corresponding i0 is the exchange current density, α is is
the symmetry factor,
a highly non-linear dependence on the over-voltage,
load current.onThis the LFP/C
non-linear battery current
dependence presented in Figure
of the charge-transfer
n is
16. the number of electrons, R is the universal gas constant (8.3144 J·K −1·mol−1), F is the Faraday
resistance,
constant
and of the corresponding
(96,485.339
over-voltage, on the the LFP/C battery currenttheis electrode
presented in
theC·mol
−1), η is the overpotential
Theoretically, exponential dependence between(i.e., the Li-iondifference between
battery current and the potential
Figure 16.
potential and equilibrium potential).
can be determined from the Butler-Volmer equation given in Equation (7) [47].

 αnF ⋅η − (1−α ) nF ⋅η 
i = i0  e RT − e RT  (7)
 
 
where i is the electrode current density, i0 is the exchange current density, α is the symmetry factor,
n is the number of electrons, R is the universal gas constant (8.3144 J·K−1·mol−1), F is the Faraday
constant (96,485.339 C·mol−1), η is the overpotential (i.e., the difference between the electrode
potential and equilibrium potential).

Figure 16. Measured non-linear dependence of the charge transfer resistance and of the corresponding
Figure 16. Measured non-linear dependence of the charge transfer resistance and of the corresponding
over-voltage on the current of the LFP/C battery (note: positive current stands for charging).
over-voltage on the current of the LFP/C battery (note: positive current stands for charging).

Theoretically, the exponential dependence between the Li-ion battery current and the potential
can be determined from the Butler-Volmer equation given in Equation (7) [47].
 
αnF (1−α)nF
i = i0 e RT ·η − e− RT ·η (7)
Figure 16. Measured non-linear dependence of the charge transfer resistance and of the corresponding
over-voltage on the current of the LFP/C battery (note: positive current stands for charging).
where i is the electrode current density, i0 is the exchange current density, α is the symmetry factor, n is
the number of electrons, R is the universal gas constant (8.3144 J·K−1 ·mol−1 ), F is the Faraday constant
Batteries 2016, 2, 37 12 of 20

(96,485.339 C·mol−1 ), η is the overpotential (i.e., the difference between the electrode potential and
equilibrium potential).
Batteries 2016, 2, 37 12 of 19
However, information about various parameters given in Equation (7) were not available and
require specialized
Batteries 2016, 2, 37 laboratory equipment in order to determine them. Consequently, in
However, information about various parameters given in Equation (7) were not available and
12 order
of 19 to
determine the dependence of the LFP/C battery voltage on the load current,
require specialized laboratory equipment in order to determine them. Consequently, in order to the following approach
However, information about various parameters given in Equation (7) were not available and
was followed:
determine the a comparison
dependence of wastheperformed
LFP/C batterybetween
voltage the battery
on the voltagethe
load current, response
followingmeasured
approach over
require specialized laboratory equipment in order to determine them. Consequently, in order to
a period
was of 18 s
followed: in athe laboratory
comparison (during
was the
performed HPPC
between tests—see
the battery Section
voltage3.5) and
response
determine the dependence of the LFP/C battery voltage on the load current, the following approach
the simulated
measured overvoltage
a
period
response. of 18 s
The obtained in the laboratory
difference (during
was the HPPC
considered tests—see
to describe Section 3.5)
the voltage and the
impact response simulated
of the current voltage
on the
was followed: a comparison was performed between the battery measured overvoltage
a
response.
of the LFP/C The obtained
battery. difference
The was considered
aforementioned to describe
procedure wastherepeated
impact of for
the current
all on the
SOCs, voltage
temperatures,
period of 18 s in the laboratory (during the HPPC tests—see Section 3.5) and the simulated voltage
of the LFP/C battery. The aforementioned procedure was repeated for all SOCs, temperatures, and
and C-rates,
response.which
The obtainedwere difference
considered wasduring the to
considered battery
describeimpedance
the impact of measurement
the current on theby voltage
HPPC test.
C-rates, which were considered during the battery impedance measurement by HPPC test. The
of the LFP/C
The obtained battery.
voltage The aforementioned
correction procedure
factors, assigned waseffect
to the repeated for all
of the SOCs,on
current temperatures,
the LFP/Cand battery
obtained voltage correction factors, assigned to the effect of the current on the LFP/C battery voltage
C-rates,
voltage were which
stored were
as considered
a 3D during
look-up the battery
table; for impedance
instance, measurement
Figure 17 presentsby HPPC
the test. Thetable
look-up
were stored as a 3D look-up table; for instance, Figure 17 presents the look-up table implementation
obtained voltage
implementation correction factors, assigned to the
for effect
the 25of◦the current on the LFP/C battery voltage
of the voltageofcorrection
the voltage correction
factors factors
for the 25 °C dimension. C dimension.
were stored as a 3D look-up table; for instance, Figure 17 presents the look-up table implementation
of the voltage correction factors for the 25 °C dimension.

Figure 17. Dependence of the voltage correction factor on the SOC and current for T = 25 °C.
Figure 17. Dependence of the voltage correction factor on the SOC and current for T = 25 ◦ C.
Figure 17. Dependence of the voltage correction factor on the SOC and current for T = 25 °C.
4.3. Model Implementation
4.3. Model Implementation
4.3. Model
BasedImplementation
on the voltage Equation (6) and considering the dependences of the performance
Based on theon
parameters voltage Equation conditions,
the operating (6) and considering the dependences
the performance model ofof the
the performance
LFP/C batteryparameters
was
Based on the voltage Equation (6) and considering the dependences of the performance
on the operating conditions,
implemented according tothe
theperformance
block diagrammodel of the
proposed LFP/C18.
in Figure battery was implemented according
parameters on the operating conditions, the performance model of the LFP/C battery was
to theimplemented
block diagram proposed
according inblock
to the Figure 18. proposed in Figure 18.
diagram

Q1, N1 Q2, N2
Rs Ls

Q1, N1 Q2, N2
R1 R2
Rs Ls

R1 R2
1
SOC = SOC i −
Capacity
 I bat ⋅ dt ZEEC⋅ Ibat

1
SOC = SOC i −
Capacity
 I bat ⋅ dt ZEEC⋅ Ibat

Figure 18. Block diagram implementation of the performance model for the LFP/C battery.

FigureFigure 18. Block

18. Block diagram
diagram implementationof
implementation ofthe
the performance
performance model
model forfor
thethe
LFP/C
LFP/C battery.
battery. load
The inputs of the proposed performance model for the LFP/C battery are the applied
current, the initial SOC and the temperature, while the output of the model is the estimated battery
The inputs of the proposed performance model for the LFP/C battery are the applied load
voltage;
The additionally,
inputs the model can returnmodel
information about thebattery
batteryare SOC theatestimated
every moment.
current, theof the proposed
initial SOC and the performance
temperature, whilefor
thethe LFP/C
output of the model is the applied load current,
battery
Furthermore,
the initial SOC and since
theno thermal
temperature, model was
while developed
the output for
of the
the LFP/C
model battery,
is the the temperature
estimated batteryof the
voltage;
voltage; additionally, the model can return information about the battery SOC at every moment.
battery is considered constant during the whole simulation.
additionally, the model
Furthermore, canthermal
since no returnmodel
information about the
was developed forbattery SOC
the LFP/C at every
battery, themoment.
temperature Furthermore,
of the
The parameterization of the proposed performance model was performed based on the results
since battery is considered
no thermal modelconstant during thefor
was developed whole
thesimulation.
LFP/C battery, the temperature of the battery is
obtained from the characterization test, which was presented in detail in Section 3. The parameters
considered The parameterization
constant during of whole
the proposed performance model was performed based on the results
of the LFP/C battery werethe
implementedsimulation.
as 2D look-up tables (i.e., capacity, OCV, EEC’s parameters)
obtained from the characterization test, which was presented in detail in Section 3. The parameters
of the LFP/C battery were implemented as 2D look-up tables (i.e., capacity, OCV, EEC’s parameters)
Batteries 2016, 2, 37 13 of 20

The parameterization of the proposed performance model was performed based on the results
obtained from
Batteries 2016, 2, 37the characterization test, which was presented in detail in Section 3. The parameters 13 of 19
of the LFP/C battery were implemented as 2D look-up tables (i.e., capacity, OCV, EEC’s parameters)
and 3D-look-up
and 3D-look-up table (i.e., Butler-Volmer
Butler-Volmer correction
correction voltage)
voltage) in in order
order to
to consider
consider their
their dependences
dependences
on the
on the operating
operating conditions
conditions (i.e.,
(i.e., SOC,
SOC, temperature,
temperature, and and load
load current).
current). Because
Because forfor similar
similar operating
operating
conditions, the
conditions, the capacity
capacityofofthetheLi-ion
Li-ion batteries is different
batteries is different between
between charging andand
charging discharging
dischargingcase case
(i.e.,
Coulombic
(i.e., Coulombic efficiency different
efficiency thanthan
different 100%), two two
100%), separate look-up
separate tables
look-up were
tables implemented
were implemented for the
for
charging and discharging capacity, respectively. Because of the hysteresis effect
the charging and discharging capacity, respectively. Because of the hysteresis effect (see Figure 5), (see Figure 5), which
is inherent
which to LFP/C
is inherent batterybattery
to LFP/C chemistry, a similar
chemistry, approach
a similar was followed
approach for implementing
was followed for implementing the OCV the
characteristic;
OCV however,
characteristic; for other
however, Li-ionLi-ion
for other battery chemistries,
battery a single
chemistries, look-up
a single table istable
look-up enough to model
is enough to
the OCV
model thecharacteristic for both
OCV characteristic forcharging and discharging
both charging cases. cases.
and discharging
During the
During the fitting
fitting process
process ofof the
the measured
measured impedance
impedance spectra,
spectra, ititwas
was found
foundout outthat
thatonly
onlyRRss,, R R11,
R22, Q11, and Q22 are
R are dependent
dependent on on the
the SOC and temperature at which the EIS measurements had had been
been
performed. Consequently, these parameters were implemented into the performance
performed. performance model as as 2D 2D
look-up tables
look-up tablessince
sincetheir
theirvalues
valuesare aredependent
dependent ononthethe
SOCSOC andand temperature;
temperature; Figure
Figure 19 exemplifies
19 exemplifies the
the look-up
look-up tabletable implementation
implementation of the ofseries
the series resistance
resistance Rs. other
Rs . The The other parameters
parameters of theofEEC,
the EEC,
Ls , N1L,s,and
N1,
and
N N2 were found
to betoconstant,
be constant, independent of the operating conditions:
Ls =Ls1.8
= 1.8·× − 8
× 1010 H,
−8 H,NN11 == 0.5,
2 were found independent of the operating conditions:
and N
and N22 == 0.8.
0.8.

Figure 19.
Figure 19. Two-dimensional
Two-dimensional look-up
look-up table
table implementation
implementationof
ofRRss of
of the
the EEC.
EEC.

Finally, the estimated voltage of the LFP/C battery was obtained by summing up the
Finally, the estimated voltage of the LFP/C battery was obtained by summing up the contributions
contributions from the OCV, the voltage drop across the EEC, and the Butler-Volmer correction
from the OCV, the voltage drop across the EEC, and the Butler-Volmer correction voltage, as illustrated
voltage, as illustrated in Figure 18.
in Figure 18.
5. Validation
5. Validation of
of the
the Performance
PerformanceModel
Model
In order
In order to
to verify
verify the
the accuracy
accuracy of
of the
thedeveloped
developedperformance
performancemodel
modelofofthe
theLFP/C
LFP/Cbattery
batteryand to
and
validate the proposed characterization procedure, different load current profiles were considered.
to validate the proposed characterization procedure, different load current profiles were considered.
The accuracy
The accuracy ofof the
the developed
developed performance
performance model
model was
was evaluated
evaluated by
by computing
computing the
the coefficient
coefficient of
of
determination R 2 which was obtained by comparing the measured and estimated battery’s voltage
2
determination R which was obtained by comparing the measured and estimated battery’s voltage
profiles, according
profiles, accordingto toEquation
Equation(8).
(8).
n

∑(Vmodel ( x ) − Vmeas ( x ))
(V ( x ) − V ( x )) 2
2
n
model meas
R 2 = 1 − x =1 (8)
R2 = 1 − x=1n n
( )
2 (8)
∑
Vmodel ( x ) − Vmeas
2
V
x =1 model
( x ) − V meas
x =1
where Vmodel is the estimated battery voltage, Vmeas represents the measured battery voltage, x
where V model is the estimated battery voltage, V meas represents the measured battery voltage, x
represents the present observation, and n represents the total number of observations.
represents the present observation, and n represents the total number of observations.
Moreover, the mean ( ε ) and the maximum (εmax) deviation between the measured and estimated
battery voltage profiles were computed according to Equations (9) and (10) for the three considered
verification cases.

ε = Vmeas (x) − Vmodel (x) (9)

Batteries 2016, 2, 37 14 of 20

Moreover, the mean (ε) and the maximum (εmax ) deviation between the measured and estimated
battery voltage profiles were computed according to Equations (9) and (10) for the three considered
Batteries 2016,
verification 2, 37
cases. 14 of 19
Batteries 2016, 2, 37 14 of 19
|ε| = |Vmeas ( x ) − Vmodel ( x )| (9)
 Vmeas (x ) − V model (x ) 
ε max = max 100|V
 ⋅ meas
Vmeas(Vxx))−− Vmodel
Vmodel(x ) ( x )|

(10)
meas (x )
max = max100 ⋅ 
εmax =εmax 100 · (10)
Vmeas(x() x )  (10)
 meas 
5.1.5.1.
Pulse
PulseDischarge
Discharge
5.1. Pulse Discharge
TheThefirst
firstvalidation
validationofofthethedeveloped
developed performance
performance modelmodelwas
wascarried
carriedout
outusing
usinga adischarging
discharging
current The
currentpulse first
pulse validation
profile.
profile. From
From ofaathe
fullydeveloped
fully charged performance
chargedstate, thethe
state, LFP/Cmodel waswas
battery
LFP/C carried
battery was out
dischargedusing
witha 1C-rate
discharged discharging
(i.e.,
with 1C-rate
current
2.5 A) in pulse
steps profile.
of 10% From
SOC; a fully
between charged
two state,
consecutivethe LFP/C battery
discharging was
currentdischarged
pulses,
(i.e., 2.5 A) in steps of 10% SOC; between two consecutive discharging current pulses, a relaxation a with 1C-rate
relaxation (i.e.,
period
2.5
of 15A)
minin steps
was of 10%
applied. SOC;
The between
measured two
and consecutive
estimated discharging
voltage current
profiles of the
period of 15 min was applied. The measured and estimated voltage profiles of the LFP/C for this pulses,
LFP/C a
forrelaxation
this period
verification
of 15 are
case minillustrated
verification was applied.
case in The
are illustrated measured
Figure in20. Theand
Figure 20.estimated
developed voltage performance
profiles
performance
The developed of the
model wasLFP/C for
ablewas
model tothis verification
estimate
able very
to estimate
case are
accurately illustrated
the voltage in Figure
of the 20.
tested The developed
LFP/C battery, performance
with a maximummodel was
error ε able to estimate
max = 3.66% and a mean very
very accurately the voltage of the tested LFP/C battery, with a maximum error εmax = 3.66% and
accurately
error ε = 6.8 the voltage of the tested LFP/C battery, with a maximum error εmax = 3.66% and a mean
a mean error ε =mV.
6.8 mV.
error ε = 6.8 mV.
3.6
Measurement
3.6
3.4 Simulation
Measurement
3.4 Simulation
3.2
3.2
[V][V]

3
3 3.3
Voltage

2.8
3.3
Voltage

2.8
2.6 3.25
2.6 3.25
2.4
3.2
2.4 6000 6500 7000 7500
2.2 3.2
2.2 6000 6500 7000 7500
2
0 2000 4000 6000 8000 10000 12000
2
0 2000 4000 Time [s]
6000 8000 10000 12000
Time [s]
Figure
Figure 20.20. Measuredand
Measured andestimated
estimatedvoltage
voltage profiles
profiles during
during pulse
pulse discharging
dischargingofofthe
theLFP/C
LFP/Cbattery
battery
Figure
(I bat = 2.520.
A, Measured
T = 25 ◦°C).and estimated voltage profiles during pulse discharging of the LFP/C battery
(Ibat = 2.5 A, T = 25 C).
(Ibat = 2.5 A, T = 25 °C).
The distribution of the voltage deviations, expressed as relative error values, is presented in
The
Figure
distribution
The
21,distribution
showing that
ofofthe thevoltage
the voltage
model
deviations,
deviations,
has
expressed
expressed
a slight tendency
as relative
as relativeerror
of over-estimating
errorthe values,
values, isispresented
voltage
presented
behavior ofin in
Figure
Figure
the 21, showing
21, showing
tested that the
that thethe
LFP/C battery; model
model has
mainhas a slight tendency
a slight tendency
contribution of over-estimating
of thisoftendency
over-estimating the voltage
the voltage
is represented by the behavior
behavior
voltageof
of estimation
the tested LFP/C
tested error
LFP/Cafter battery;
battery; themain
the
the current main
pulse contribution
contribution
is cut off. Theofofthis
thistendency
dynamic tendency isisrepresented
behavior represented
of bybythe
the performance thevoltage
voltage
model
estimation
might be error
estimation error
improved after
afterthe
(thus thecurrent
current
the pulse
pulse
accuracy ofisiscut
the cutoff.
off. The
model willdynamic
The dynamic behavior
behavior
further increase) of
byof theperformance
the performance
considering model
an EECmodel
with
might be improved (thus the accuracy of the model will further increase) by
more ZARC elements (or R-C parallel networks) [24]; however, this will cause a high increase ofwith
might be improved (thus the accuracy of the model will increase) by considering
considering anan EEC
EEC with
the
moremore
ZARCZARC
computation elements
elements
time of (orthe(orR-CR-Cparallel
parallelnetworks)
performance networks) [24];
model. [24]; however,
however, thisthis will
willcause
causeaahighhighincrease
increaseofofthe
the
computation
computation timetime
of of
the the performancemodel.
performance model.
1000
Battery cell voltage Battery cell voltage
1000
over-estimation
Battery cell voltage under-estimation
Battery cell voltage
800 over-estimation under-estimation
of Occurances

800
of Occurances

600
600
400
Number

400
Number

200
εmax = 3.66%
200
εmax = 3.66%
0
-4 -2 0 2 4
0
-4 -2 Relative Error
0 [%] 2 4
Relative Error [%]
Figure 21. Distribution of the relative error obtained during pulse discharging verification case.
Figure
Figure 21.21. Distributionofofthe
Distribution therelative
relativeerror
errorobtained
obtained during
during pulse
pulse discharging
dischargingverification
verificationcase.
case.
5.2. Dynamic Pulse Charging and Discharging
5.2. Dynamic Pulse Charging and Discharging
In real applications (e.g., e-mobility, grid support etc.), Li-ion batteries are subjected to complex
In real applications
charging-discharging (e.g., e-mobility,
profiles. grid the
Consequently, support etc.), Li-ion
developed batteriesmodel
performance are subjected to complex
was verified for a
charging-discharging profiles. Consequently, the developed performance model was verified for a
Batteries 2016, Batteries
2, 37 2016, 2, 37 15 of 19 15 of 20

current profile, which was composed of charging and discharging pulses of different C-rates, over
the whole SOC interval; the considered current profile is presented in Figure 22. To allow for a non-
5.2. Dynamic Pulse
biased Chargingthe
verification, and Discharging
C-rates, which were used for this dynamic profile, are different than the C-
rates used to compute the Butler-Volmer correction voltage (see Sections 3.5 and 4.2).
In real applications (e.g., e-mobility, grid support etc.), Li-ion batteries are subjected to complex
The measured and estimated voltage profiles obtained by applying the dynamic current profile,
charging-discharging profiles.
presented in Figure 22, are Consequently,
illustrated in Figurethe developed
23. For this case, theperformance
proposed model model
is able to was
predictverified for
a current profile,
the voltagewhich was
of the composed
LFP/C of acharging
battery with mean errorand
of ε discharging pulses oferror
= 6.4 mV and maximum different
of εmax =C-rates,
4.89%. over the
whole SOC Furthermore,
interval; thetheconsidered
distribution current
of the relative
profileerror, presented ininFigure
is presented Figure24,22.
shows in this for
To allow caseaanon-biased
tendency of the developed model to under-estimate the voltage of the battery; as shown in the
verification, the C-rates, which were used for this dynamic profile, are different than the C-rates used
zoomed
Batteries view
2016, 2, 37of Figure 23, this under-estimation of the voltage was mainly caused by the inability of
15 of 19
to computethethemodel
Butler-Volmer correction
to react to sudden and largevoltage
variation(see Sections
of C-rates 3.5 andto4.2).
(i.e., 3.5C-rate 1C-rate).
Batteries 2016, 2, 37 15 of 19
current profile, which was composed of charging and discharging pulses of different C-rates, over
10
the wholeprofile,
current SOC interval;
which wasthe considered
composed of current profile
charging andisdischarging
presented inpulses
Figureof22. To allow
different for a non-
C-rates, over
biased verification,
the whole the C-rates,
SOC interval; which
the considered
5
were usedprofile
current for this dynamic profile,
is presented are22.
in Figure different
To allowthan
forthe C-
a non-
rates used to compute the Butler-Volmer correction voltage (see Sections 3.5 and 4.2).
biased verification, the C-rates, which were used for this dynamic profile, are different than the C-
ratesThe measured andthe
estimated 0 voltage correction
profiles obtained
voltageby applying the
3.5dynamic
and 4.2).current profile,
Current [A]

used to compute Butler-Volmer (see Sections

presented in Figure 22,
The measured andare illustrated
estimated in Figure
voltage 23. For
profiles this case,
obtained the proposed
by applying model is current
the dynamic able to predict
profile,
the voltage in
presented of Figure
the LFP/C battery
22, are -5
with
illustrated a in
meanFigure 23.ofForε5this
error = 6.4 mVthe
case, andproposed
maximum errorisofable
model εmaxto
= 4.89%.
predict
Furthermore, the distribution of the
the voltage of the LFP/C battery withCharging relative
a mean error of ε = 6.4 mV and maximum error of εmax =case
error, presented
0 in Figure 24, shows in this a
4.89%.
to
-10
tendency of thethedeveloped
Furthermore, distributionmodel
of thetonextunder-estimate
SOC
relative the voltage
error, -5presented in of the battery;
Figure as shown
24, shows in this in thea
case
Charging/Discharging

zoomed
tendencyview of Figure
of the 23, thismodel
developed under-estimation
to under-estimateof the voltage was mainly
the voltage caused by the inability of
profile applied at each SOC

-15 2000 2200 of the battery; as shown in the

the model to react to sudden and 0large variation
1000 of
2000 C-rates
3000 (i.e.,
40003.5C-rate
5000 to 1C-rate).
zoomed view of Figure 23, this under-estimation of the voltage was mainly caused by the inability of
Time [s]
the model to react to sudden and
10
large variation of C-rates (i.e., 3.5C-rate to 1C-rate).
Figure 22. Dynamic current profile used for verification of the LFP/C battery performance model.
Figure 22. Dynamic current profile
10 used for verification of the LFP/C battery performance model.
5
3.6
Measurement
Simulation 5
The measured and estimated voltage profiles obtained by applying the dynamic current profile,
[V][A] [A]

0
3.4
Current

presented in Figure 22, are illustrated 0in Figure 23. For this case, the proposed model is able to predict
-5
3.2 5
Current

the voltage of the LFP/C battery with a mean error of 0ε = 6.4 mV and maximum error of εmax = 4.89%.
Voltage

-5 5
Charging to
Furthermore, the distribution of the -10 relative
3 error, presented
next SOC 3.4 -5
0
in Figure 24, shows in this case a tendency
Charging/Discharging
Charging to
of the developed model to under-estimate -10 nextthe
SOC voltage of the battery; as shown in the zoomed view of
profile applied at each SOC

-15 3.3 -52000 2200

2.8
0 1000 2000 3000 4000 5000
Charging/Discharging
profile applied at each SOC
Figure 23, this under-estimation of the -15
voltage wasTime mainly
[s] 2000 caused
3.2 2200 by the inability of the model to react
0 1000 2000 3000 3800
3600 4000 40005000
to sudden and large variation of C-rates
2.6
0
Figure 22. Dynamic current profile (i.e.,
used 3.5C-rate
Time [s]
1000 for verification
2000 to 1C-rate).
3000of the4000 5000
LFP/C battery performance model.
Time [s]
Figure 22. Dynamic current3.6profile used for verification of the LFP/C battery performance model.
Figure 23. Measured and estimated voltage profiles during dynamic pulse charging and discharging
Measurement
3.6 T = 25
of the LFP/C battery (Ibat = 2.5 A, Simulation
°C).
3.4 Measurement
Simulation
1200
3.4
[V] [V]

3.2
Voltage

1000
3.2
Number of Occurances

3 3.4
Voltage

800
3 3.4
3.3
2.8
600
3.3
2.8 3.2
400 3600 3800 4000
2.6
0 1000 2000 3.2 3000 4000 5000
2.6 Time [s]3600 3800 4000
2000 1000 2000 3000 4000 5000
Figure 23. Measured and estimated voltage profiles Time [s] dynamic pulse charging and discharging
during
Figure 23. Measured
ofFigure
the LFP/C and estimated
battery (I
0voltage profiles during dynamic pulse charging and discharging
bat = 2.5 A, T = -4
25 °C). -2 0 2 4
23. Measured and estimated voltage profiles during dynamic pulse charging and discharging
of the LFP/C battery (Ibat = (I2.5
of the LFP/C battery
A, T = 25 ◦ C). Relative Error [%]
bat = 2.5 A, T = 25 °C).
1200
Figure 24. Distribution of the relative error obtained during dynamic pulse charging and discharging
verification case. 1200
1000
of Occurances

1000
800
of Occurances

800
600
Number

600
400
Number

400
200

200
0
-4 -2 0 2 4
0 Relative Error [%]
-4 -2 0 2 4
Relativeduring
Figure 24. Distribution of the relative error obtained Error [%]
dynamic pulse charging and discharging
verification case.
Figure 24. Distribution of the relative error obtained during dynamic pulse charging and discharging
Figure 24. Distribution of the relative error obtained during dynamic pulse charging and discharging
verification case.
verification case.
Batteries 2016, 2, 37 16 of 20
Batteries 2016, 2, 37 16 of 19
Batteries 2016, 2, 37 16 of 19
Batteries 2016, 2, 37 16 of 19
5.3. Field-Measured Profile
5.3.5.3.
Field-Measured
Field-MeasuredProfileProfile
5.3. Field-Measured Profile of the developed performance model was performed using a realistic
The third verification
TheThethird
thirdverification of the
verification developed
of the developed performance
performancemodel was was
model performed
performed usingusing
a realistic current
a realistic
current profile,
The third which was of measured on field on LFP/C batteries providing primary frequency
profile,
currentwhich wasverification
profile, measured
which wasonmeasured
the developed
field on LFP/C
on field
performance
batteries modelprimary
providing
on LFP/C batteries
was performed
frequency
providing
using
primary
a realistic
regulation
frequency (PFR)
regulation
current (PFR) in
profile, the Danish
which was energy market
measured on [5]. The
field on considered
LFP/C current
batteries profile with
providing a length
primary of two
frequency
in the Danish(PFR)
regulation energy in market
the Danish [5]. energy
The considered
market [5]. current profile with
The considered a length
current of two
profile withhours is illustrated
a length of two
hours is illustrated
regulation (PFR) inin Figure
the Danish 25.energy
The comparison
market [5].between the estimated
The considered current and measured
profile with avoltage
length of of two
the
hours is25.
in Figure illustrated
The in Figurebetween
comparison 25. The comparison
the estimated between
and the estimated
measured voltageandofmeasured
the LFP/C voltage of when
battery, the
LFP/C
hours is illustrated in Figure 25. The comparison between the estimated and measured voltage of 26.
battery, when the field-measured current profile was considered, is presented in Figure the
theLFP/C battery, when
field-measured the profile
current field-measured current profile
was considered, was considered,
is presented in Figure 26. is presented in Figuremodel
The performance 26.
The
LFP/C performance
battery, when modeltheoffield-measured
the LFP/C battery estimates
current profilegenerally with good
was considered, accuracy (inε Figure
is presented = 8.4 mV 26.
of the
TheLFP/C batterymodel
performance estimates
of thegenerally with good
LFP/C battery accuracy
estimates (ε = 8.4with
generally mV good
and εmax = 4.54%)
accuracy ( ε =the
8.4voltage
mV
and ε = 4.54%) the voltage of the battery. Nevertheless, the model has the
The performance model of the LFP/C battery estimates generally with good accuracy ( ε = 8.4 mV
max tendency to over-estimate
and
of the εmax = 4.54%)
the voltage
battery. the voltagethe
Nevertheless, of the battery.
model has Nevertheless,
the tendency the model has the tendency
to over-estimate the to over-estimate
response
and εmax = 4.54%) of the LFP/C
the voltage of thebattery,
battery.when sudden and
Nevertheless, large variation
the model of voltage
has the tendencythe load response
current have
to over-estimate
of the
the voltage
LFP/C battery, response
when of the LFP/C
sudden and battery,
large when of
variation sudden
the andcurrent
load large variation
have of the load
occurred (i.e., current
changes have
ininthe
occurred
the voltage (i.e., changesofin
response thethe current
LFP/C fromwhen
battery, belowsudden
1C-rateandto 4C-rate); this behavior
large variation is illustrated
of the load current have
occurred
current from (i.e.,
belowchanges in
1C-rate the current
to 4C-rate); from below
thisrelative 1C-rate to
behavior is illustrated 4C-rate);
in this behavior is illustrated in
Figure
occurred 27,(i.e.,
where the distribution
changes in the current of the
from below error1C-rateobtained byFigure
to 4C-rate); comparing 27, where
this behavior the is the distribution
measured
illustratedand in
Figure
of estimated
the 27, where
relative the distribution
error obtained by of the relative
comparing error obtained
the measured by comparing
and estimated voltage the measured
profiles and
is plotted.
voltage profiles is plotted.
Figure 27, where the distribution of the relative error obtained by comparing the measured and
estimated voltage profiles is plotted.
estimated voltage profiles is plotted.
10
10
10
5
5
[A][A][A]

5
Current

0
Current

0
Current

0
-5
-5
-5
-10
-100 1000 2000 3000 4000 5000 6000 7000
-100 Time4000
1000 2000 3000 [s] 5000 6000 7000
0 Time 4000
1000 2000 3000 [s] 5000 6000 7000
Figure
25.25. Field-measuredcurrent usedTime
current profile used for [s]
the
the verification
verificationof the
theperformance model.
Figure
Figure 25.Field-measured
Field-measured currentprofile
profile used for
for the verification ofofthe performance
performance model.
model.
Figure 25. Field-measured 3.6
current profile used for the verification of the performance model.
3.6 Measurement
3.6 Measurement
Simulation
3.4 Measurement
Simulation
3.4 Simulation
3.4
[V][V][V]

3.2
3.2
Voltage

3.2
Voltage

3
Voltage

3 3.3
3 3.3
2.8 3.2
3.3
2.8 3.2
2.8 3.1
3.2
2.6 3.1 4000 5000 6000
0 1000 2000 30003.1400040005000500060006000
7000
2.6
0 Time 4000
1000 2000 3000 [s]400050005000
6000 6000
7000
2.6
0 Time 4000
1000 2000 3000 [s] 5000 6000 7000
Figure 26. Measured and estimated voltage profilesTime [s]
obtained for the field-measured current profile at
Figure 26. Measured and estimated voltage profiles obtained for the field-measured current profile at
Figure
= 2526.
TFigure Measured
°C. and estimated voltage profiles obtained for the
26. Measured and estimated voltage profiles obtained for thefield-measured
field-measuredcurrent
currentprofile
profileat at
T = 25 °C.
◦ C.
T =T25= 25 °C. 3000
3000
3000
2500
Occurances

2500
Occurances

2500
2000
Occurances

2000
2000
1500
of

1500
of of
Number

1500
1000
Number

1000
Number

1000
500
500
500
0
-4 -2 0 2 4
0
-4 -2Relative Error
0 [%] 2 4
0
-4 -2Relative Error
0 [%] 2 4
Relative Error
Figure 27. Distribution of the relative error obtained [%] the field-measured profile verification
during
Figure 27. Distribution of the relative error obtained during the field-measured profile verification
case.
Figure 27. Distribution of the relative error obtained during the field-measured profile verification
case. 27. Distribution of the relative error obtained during the field-measured profile verification case.
Figure
case.
Batteries 2016, 2, 37 17 of 20

The developed performance model is able to accurately estimate the voltage behavior of the tested
LFP/C battery—regardless of the applied current profile—returning a maximum error lower than 5%
and a mean error below 8.5 mV as summarized in Table 2.

Table 2. Performance model accuracy for different load profiles. PFR: primary frequency regulation.

Profile Mean Error, |ε| Maximum Error, εmax R2

Pulse discharging 6.8 mV 3.66% 0.9917
Dynamic profile 6.4 mV 4.89% 0.9255
PFR profile 8.4 mV 4.54% 0.7724

6. Conclusions
A seven-step characterization methodology for measuring the performance characteristics
of the Li-ion batteries was proposed in the first part of this work. For exemplification of the
procedure, a commercially available 2.5 Ah LFP/C battery was used. The results obtained during
the characterization test were used to develop and parameterize the performance model of the
LFP/C battery.
In order to model the dynamic behavior of the battery, a novel hybrid procedure was proposed.
Thus, the developed performance model combines information obtained from the electrochemical
impedance spectroscopy (EIS) test (i.e., dependence of the impedance on SOC and temperature) and
HPPC test (i.e., dependence on the load current), respectively. The proposed hybrid performance model
was verified using different representative dynamic current profiles. The obtained results, maximum
error lower than 5% and a mean error below 8.5 mV, have suggested that the developed model is
able to estimate with high accuracy the voltage of the LFP/C battery, independent of the considered
test conditions. Thus, it can be concluded that the proposed seven-step generalized characterization
procedure for Li-ion batteries was validated.
Even though the proposed characterization procedure was illustrated for a certain Li-ion battery
chemistry (i.e., LFP/C), it can be applied to any type of Li-ion battery chemistry with the amendment
that the values of the test conditions (i.e., temperature and load current levels) have to be adjusted in
order to match the manufacturer data-sheets.

Author Contributions: Daniel-Ioan Stroe and Maciej Swierczynski conceived, designed, and performed the
experiments. Daniel-Ioan Stroe and Ana-Irina Stroe analyzed the data; Søren Knudsen Kær reviewed the paper;
and Daniel-Ioan Stroe wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Pistoia, G. (Ed.) Lithium-Ion Batteries: Advances and Applications; Elsevier: Amsterdam, The Netherlands, 2014.
2. Scrosati, B.; Garche, J. Lithium batteries: Status, prospects and future. J. Power Sources 2010, 195, 2419–2430.
[CrossRef]
3. Alexander, D.; Gartner, J. Electric Vehicle Batteries—Lithium Ion Batteries for Hybrid, Plug-in-Hybrid, and Battery
Electric Vehicles: Global Market Analysis and Forecasts; Navigant Research: Boulder, CO, USA, 2014.
4. Catenacci, M.; Verdolini, E.; Bosetti, V.; Fiorese, G. Going electric: Expert survey on the future of battery
technologies for electric vehicles. Energy Policy 2013, 61, 403–413. [CrossRef]
5. Swierczynski, M.; Stroe, D.I.; Lærke, R.; Stan, A.I.; Kjær, P.C.; Teodorescu, R.; Kær, S.K. Field experience from
Li-ion BESS delivering primary frequency regulation in the Danish energy market. ECS Trans. 2014, 61, 1–14.
[CrossRef]
6. Gaillac, L.; Castaneda, J.; Edris, A.-A.; Elizondo, D.; Wilkins, C.; Vartanian, C.; Mendelsohn, D. Tehachapi
wind energy storage project: Description of operational uses, system components, and testing plans.
In Proceedings of the 2012 IEEE PES Transmission and Distribution Conference and Exposition (T&D),
Orlando, FL, USA, 7–10 May 2012; pp. 1–6.
Batteries 2016, 2, 37 18 of 20

7. Rastler, D.M. Electric Energy Storage Technology Options: A White Paper Primer on Applications, Costs, and
Benefits; Technical Report 1020676; Electric Power Research Institute (EPRI): Palo Alto, CA, USA, 2010.
8. Li, X.; Hui, D.; Xu, M.; Wang, L.; Guo, G.; Zhang, L. Integration and energy management of large-scale
lithium-ion battery energy storage station. In Proceedings of the 2012 15th International Conference on
Electrical Machines and Systems (ICEMS), Sapporo, Japan, 21–24 October 2012; pp. 1–6.
9. Vartanian, C.; Bentley, N. A123 systems’ advanced battery energy storage for renewable integration. In
Proceedings of the 2011 IEEE/PES Power Systems Conference and Exposition (PSCE), Phoenix, AZ, USA,
20–23 March 2011; pp. 1–6.
10. Jeffe, S.; Adamson, K.-A. Advanced Batteries for Utility-Scale Energy Storage; Navigant Research: Boulder, CO,
USA, 2014.
11. Battery Storage for Renewables: Market Status and Technology Outlook; The International Renewable Energy
Agency (IRENA): Masdar, United Arab Emirates, 2015.
12. Klein, R.; Chaturvedi, N.; Christensen, J.; Ahmed, J.; Findeisen, R.; Kojic, A. Electrochemical model based
observer design for a lithium-ion battery. IEEE Trans. Control Syst. Technol. 2013, 21, 289–301. [CrossRef]
13. Smith, K.; Rahn, C.; Wang, C.-Y. Model-based electrochemical estimation and constraint management for
pulse operation of lithium ion batteries. IEEE Trans. Control Syst. Technol. 2010, 18, 654–663. [CrossRef]
14. Shepherd, C. Design of primary and secondary cells—II. An equation describing battery discharge.
J. Electrochem. Soc. 1965, 112, 657–664. [CrossRef]
15. Gao, L.; Liu, S.; Dougal, R. Dynamic lithium-ion battery model for system simulation. IEEE Trans. Compon.
Packag. Technol. 2002, 25, 495–505.
16. Chen, M.; Rincon-Mora, G. Accurate electrical battery model capable of predicting runtime and IV
performance. IEEE Trans. Energy Convers. 2006, 21, 504–511. [CrossRef]
17. Swierczynski, M.; Stroe, D.; Stan, A.-I.; Teodorescu, R.; Sauer, D. Selection and performance-degradation
modeling of LiMO2 /Li4 Ti5 O12 and LiFePO4 /C battery cells as suitable energy storage systems for grid
integration with wind power plants: An example for the primary frequency regulation service. IEEE Trans.
Sustain. Energy 2014, 5, 90–101. [CrossRef]
18. Andre, D.; Meiler, M.; Steiner, K.; Walz, H.; Soczka-Guth, T.; Sauer, D.U. Characterization of high-power
lithium-ion batteries by electrochemical impedance spectroscopy. II: Modelling. J. Power Sources 2011, 196,
5349–5356. [CrossRef]
19. Rong, P.; Pedram, M. An analytical model for predicting the remaining battery capacity of lithium-ion
batteries. IEEE Trans. Large Scale Integr. Syst. 2006, 14, 441–451. [CrossRef]
20. Kim, T.; Qiao, W. A hybrid battery model capable of capturing dynamic circuit characteristics and nonlinear
capacity effects. IEEE Trans. Energy Convers. 2011, 26, 1172–1180. [CrossRef]
21. Stroe, D.-I. Lifetime Models for Lithium-Ion Batteries Used in Virtual Power Plants. Ph.D. Thesis, Aalborg
University, Aalborg, Denmark, 2014.
22. Jackey, R.; Saginaw, M.; Sanghvi, P.; Gazzarri, J.; Huria, T. Battery Model Parameter Estimation Using a Layered
Technique: An Example Using a Lithium Iron Phosphate Cell; Society of Automotive Engineers: Warrendale, PA,
USA, 2013.
23. Hentunen, A.; Lehmuspelto, T.; Suomela, J. Time-domain parameter extraction method for
Thévenin-equivalent circuit battery models. IEEE Trans. Energy Convers. 2014, 29, 558–566. [CrossRef]
24. Zhang, H.; Chow, M.-Y. Comprehensive dynamic battery modeling for PHEV applications. In Proceedings of
the 2010 IEEE Power and Energy Society General Meeting, Minneapolis, MN, USA, 25–29 July 2010; pp. 1–6.
25. Rahimi-Eichi, H.; Balagopal, B.; Chow, M.-Y.; Yeo, T.-J. Sensitivity analysis of lithium-ion battery model to
battery parameters. In Proceedings of the 39th Annual Conference of the IEEE Industrial Electronics Society
(IECON 2013), Vienna, Austria, 10–13 November 2013; pp. 6794–6799.
26. Hsieh, Y.-C.; Lin, T.-D.; Chen, R.-J.; Lin, H.-Y. Electric circuit modelling for lithium-ion batteries by
intermittent discharging. IET Power Electron. 2014, 7, 2672–2677. [CrossRef]
27. Buller, S.; Thele, M.; Karden, E.; Doncker, R.W.D. Impedance-based non-linear dynamic battery modeling for
automotive applications. J. Power Sources 2003, 113, 422–430. [CrossRef]
Batteries 2016, 2, 37 19 of 20

28. Karden, E.; Buller, S.; Doncker, R.W.D. A frequency-domain approach to dynamical modeling of
electrochemical power sources. Electrochim. Acta 2002, 47, 2347–2356. [CrossRef]
29. Buller, S.; Thele, M.; de Doncker, R.; Karden, E. Impedance-based simulation models of supercapacitors and
Li-ion batteries for power electronic applications. IEEE Trans. Ind. Appl. 2005, 41, 742–747. [CrossRef]
30. Andre, D.; Meiler, M.; Steiner, K.; Wimmer, C.; Soczka-Guth, T.; Sauer, D.U. Characterization of
high-power lithium-ion batteries by electrochemical impedance spectroscopy. I. Experimental investigation.
J. Power Sources 2011, 196, 5334–5341. [CrossRef]
31. Barsoukov, E.; Macdonald, J.R. (Eds.) Impedance Spectroscopy: Theory, Experiment, and Applications, 2nd ed.;
Wiley: Hoboken, NJ, USA, 2005.
32. Stan, A.-I.; Swierczynski, M.; Stroe, D.-I.; Teodorescu, R.; Andreasen, S. Lithium ion battery chemistries from
renewable energy storage to automotive and back-up power applications—An overview. In Proceedings of
the 2014 International Conference on Optimization of Electrical and Electronic Equipment, Brasov, Romania,
22–24 May 2014; pp. 713–720.
33. Batteries for Electric Cars—Challenges, Opportunities, and the Outlook to 2020; The Boston Consulting Group:
Boston, MA, USA, 2010.
34. Electrical Propelled Road Vehicles—Test Specification for Lithium-Ion Traction Battery Packs and Systems—Part
1: High-Power Applications; ISO 12405-1:2011 Standard; The International Organization for Standardization
(ISO): Geneva, Switzerland, 2011.
35. Bohlen, O. Impedance Based Battery Monitoring. Ph.D. Thesis, RWTH Aachen University, Aachen,
Germany, 2008.
36. Stroe, D.-I.; Swierczynski, M.; Stan, A.-I.; Teodorescu, R.; Andreasen, S. Accelerated lifetime testing
methodology for lifetime estimation of lithium-ion batteries used in augmented wind power plants.
IEEE Trans. Ind. Appl. 2014, 50, 4006–4017. [CrossRef]
37. Ecker, M.; Gerschler, J.B.; Vogel, J.; Käbitz, S.; Hust, F.E. Development of a lifetime prediction model for
lithium-ion batteries based on extended accelerated aging test data. J. Power Sources 2012, 215, 248–257.
[CrossRef]
38. Grolleau, S.; Delaille, A.; Gualous, H.; Gyan, P.; Revel, R.; Bernard, J.; Redondo-Iglesias, E.; Peter, J.
Calendar aging of commercial graphite/LiFePO4 cell—Predicting capacity fade under time dependent
storage conditions. J. Power Sources 2014, 255, 450–458. [CrossRef]
39. Omar, N.; Verbrugge, B.; Mulder, G.; van den Bossche, P.; van Mierlo, J.; Daowd, M.; Dhaens, M.; Pauwels, S.
Evaluation of performance characteristics of various lithium-ion batteries for use in BEV application.
In Proceedings of the 2010 IEEE Vehicle Power and Propulsion Conference (VPPC), Lille, France,
1–3 September 2010; pp. 1–6.
40. Omar, N.; van den Bossche, P.; Mulder, G.; Daowd, M.; Timmermans, J.M.; van Mierlo, J.; Pauwels, S.
Assessment of performance of lithium iron phosphate oxide, nickel manganese cobalt oxide and nickel
cobalt aluminum oxide based cells for using in plug-in battery electric vehicle applications. In Proceedings
of the 2011 IEEE Vehicle Power and Propulsion Conference (VPPC), Chicago, IL, USA, 6–9 September 2011;
pp. 1–7.
41. Roscher, M.; Bohlen, O.; Vetter, J. OCV hysteresis in Li-ion batteries including two-phase transition materials.
Int. J. Electrochem. 2011, 2011, 984320. [CrossRef]
42. Roscher, M.; Bohlen, O.; Sauer, D. Reliable state estimation of multicell lithium-ion battery systems.
IEEE Trans. Energy Convers. 2011, 26, 737–743. [CrossRef]
43. Tang, X.; Mao, X.; Lin, J.; Koch, B. Li-ion battery parameter estimation for state of charge. In Proceedings of
the American Control Conference (ACC), San Francisco, CA, USA, 29 June–1 July 2011; pp. 941–946.
44. Waag, W.; Käbitz, S.; Sauer, D.U. Experimental investigation of the lithium-ion battery impedance
characteristic at various conditions and aging states and its influence on the application. Appl. Energy
2013, 102, 885–897. [CrossRef]
45. Stroe, D.; Swierczynski, M.; Stan, A.-I.; Teodorescu, R.; Andreasen, S. Experimental investigation on the
internal resistance of lithium iron phosphate battery cells during calendar ageing. In Proceedings of
the 39th Annual Conference of the IEEE Industrial Electronics Society (IECON 2013), Vienna, Austria,
10–13 November 2013; pp. 6734–6739.
Batteries 2016, 2, 37 20 of 20

46. Hafsaoui, J.; Scordia, J.; Sellier, F.; Aubret, P. Development of an electrochemical battery model and its
parameters identification tool. Int. J. Automot. Eng. 2012, 3, 27–33.
47. Bard, A.; Inzelt, G.; Scholz, F. (Eds.) Electrochemical Dictionary, 2nd ed.; Springer: Berlin, Germany, 2012.

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).