Content uploaded by Clara Serrano
Author content
All content in this area was uploaded by Clara Serrano on Jul 07, 2020
Content may be subject to copyright.
Impact Analysis of V2G Services on EV Battery
Degradation - A Review
Jingli Guo, Jin Yang, Zhengyu Lin, Clara Serrano, Ana Maria Cortes
School of Engineering and Applied Science
Aston University
Birmingham, UK
j.guo16@aston.ac.uk
Abstract—Due to their promising feature in reducing
greenhouse gas emissions, electric vehicles (EVs) have received
overwhelming support in recent decades. One of the compelling
ideas is that EVs serve as distributed energy storage and provide
ancillary services to the electricity grid. From the concept to
deployment, thorough research is required from the respective
of technology, economics and policy. This paper gives a review
of research works in terms of vehicle-to-grid (V2G) impacts on
battery degradation. Battery degradation mechanisms and main
stress factors are briefly summarized. Commonly used
degradation models, classified as theoretical models, empirical
models and semi-empirical models are reviewed in terms of
mathematical expressions, advantages and disadvantages.
Suitable applications of different models are highlighted. Based
on the review of studies on V2G impacts on battery degradation,
conclusions from current research and remarks for future work
are given in this paper.
Index Terms—Ageing model, battery degradation, electric
vehicle, lithium-ion battery, vehicle-to-grid
I. INTRODUCTION
The technology of electric vehicle (EV), regarded as a
promising solution to environmental pollution problem and oil
dependency, has witnessed rapid development. The potential
function of EVs as distributed energy storage has attracted
great attention from both industry and academia [1], [2].
Under the concept of vehicle-to-grid (V2G), EVs can serve as
energy storage and provide support to electricity grids such as
emergency demand response and frequency regulation. The
deployment of V2G can also provide load profile shaping
service for renewable integration, which enables a reduction in
investments in new electricity infrastructures caused by
increasing penetration of renewable energy [1].
Despite of the aforementioned advantages, one of the
major barriers to the deployment of V2G is the concern that
V2G operation could accelerate EV battery life degradation
[2]. Charging and discharging actions result in the reduction of
battery capability to store energy and to provide a certain
amount of power over the battery lifetime. In V2G services,
more charging/discharging cycles occur; hence, battery
degradation might be more severe than no-V2G service cases.
Although the production cost of batteries continuously
decreases, it still contributes to more than 40% of the total cost
of an EV [3]. EV users prefer to prolong the battery service
life in order to reduce the battery replacement expenses, and
EV manufacturers are reluctant to warrant EVs for such a
service that might reduce battery life. Therefore, how V2G
service affects EV battery life needs to be examined before
getting support from EV users and manufacturers.
This paper reviews current research status in areas related
to the impact analysis of V2G services on battery degradation.
V2G technology and potential V2G services are introduced
first. Battery degradation mechanism analysis and model
development are the basis of quantifying V2G impacts. Key
stress factors and commonly used battery degradation models
are summarized in this paper. A review of studies on the V2G
impact analysis on battery degradation is presented in Section
IV, followed by remarks for future work given in Section V.
II. V2G SYSTEM AND SERVICES
V2G technology enables EVs to provide bi-directional
flows of energy when connected to electric vehicle supply
equipment (EVSE) [2]. A V2G system is implemented based
on power connection and logic connection between an EV and
a grid [4]. The power connection enables the power
transmission route from the vehicle to the grid or vice versa.
The logic connection provides communications between the
vehicle and the grid to control when and in which direction to
send the power.
EVs are conventionally considered as loads from the grid
side point of view. The idea of V2G evokes another role of
EVs as distributed storage devices that can deliver power to
the grid when needed. Grid efficiency, reliability and stability
are enhanced with EVs offering V2G services such as active
power regulation, load balancing, peak shaving, frequency
regulation and providing support to incorporate renewable
energy [1], [2], [4]. The challenges that V2G technology faces
include changes in infrastructures of distribution networks,
communication issues between an EV and a grid, and battery
degradation issues.
This work is supported by Innovate UK (UK Research and Innovation)
through the VIGIL (Vehicle-to-Grid Intelligent Control) project (Reference
number 104222).
,(((
III. DEGRADATION OF EV BATTERIES
Battery is the critical component in an EV in terms of cost
and reliability. Electrode materials and packaging design have
been the research focus to improve battery performance. The
impact analysis of V2G services on battery degradation relies
on a good understanding of battery degradation mechanism
and modelling. Compared to other battery technologies, such
as lead acid and nickel metal hydride batteries, lithium-ion
batteries show advantages in energy density, power density,
environmental friendliness and charging properties [5]. With
an energy density over 220 Wh/kg [6], lithium-ion battery has
been dominating the current EV market. Therefore, a review
of lithium-ion battery degradation analysis in terms of
mechanisms, measurements and modelling is presented in this
section.
A. Battery Degradation Mechanisms
From the perspective of degradation origins, battery
degradation is categorized as either calendar ageing or cycle
ageing. Calendar ageing is the irreversible proportion of lost
capacity during storage, and cycle ageing happens during
charging and discharging of batteries [7]. Solid electrolyte
interphase (SEI) growth, chemical decomposition, and lithium
plating are microscopic phenomena of battery degradation.
Among them, SEI formation is generally accepted as the main
process responsible for battery degradation [7], [8]. The
electrochemical reactions result in a reduction of battery
capacity or available power output, shown in macroscopic
phenomena as capacity fade and power fade respectively. An
illustration of battery degradation from main causes to
consequences is shown in Fig. 1 [7], [8].
Time
High temperature
Low SOC
High SOC
SEI growth
Loss of lithium
inventory
Loss of active
materia l
Capacity fade
Powe r fade
Large cycle number
Large DOD
Low temper ature when
cycling
Large cycling current
Lith ium plat ing
Structural
changes during
cycling
Chemical
decomposition/
dissolution
reaction
Accelera tion
Factors
Degradation
Mechan isms
Degradation
Modes Consequences
Figure 1. Causes and consequences of battery degradation
Commonly considered stress factors that influence battery
degradation include battery temperature, state of charge
(SOC), current rate (C-rate), depth of discharge (DOD), and
number of cycles [7]-[9]. Battery temperature and SOC are
two principal factors considered in the calendar ageing
analysis. Cycle ageing is prone to be influenced by DOD,
cycle numbers, energy throughput and C-rate. The
aforementioned tress factors are all related to driving patterns
and charging strategies.
B. Battery Degradation Measurement
Generally, battery life is characterized by calendar life (in
chronological time) or cycle life (in cycles). Based on the
lifetime concept, remaining useful life (RUL) is introduced to
quantify battery degradation [10]. The calculation of RUL is
affected by the end of life (EOL) criterion, which is defined as
a 30% increase in degradation at a reference temperature [11].
Research shows that EV batteries meet driver needs well even
down to 30% remaining power capacity [12]. Although the
definition of RUL is straightforward, it fails to reflect driver’s
needs in the estimation of battery health condition.
Another general indicator for the health level of a battery
is state of health (SOH), which is usually defined as the ratio
of the actual capacity to the nominal capacity [8], [13], [14].
This measurement is easy to be integrated into battery
management systems. However, SOH partially indicates the
battery degradation because it excludes power output
performance.
Engineering metrics to measure the macroscopic
phenomena of battery degradation are capacity fade and power
fade, which are denoted as the capacity loss and the resistance
increase compared to initial conditions respectively [15]-[17].
Capacity fade and power fade are able to capture battery
degradation characteristics, but an overall indicator is
preferred to combine two features for the convenience of the
integration to battery management systems. A possible
solution is to convert both capacity fade and power fade to the
degradation cost [18].
C. Battery Degradation Models
Battery degradation models are categorized as theoretical
models, empirical models and semi-empirical models.
1) Theoretical Models
Theoretical models are grounded on clear principles, and
they give a profound understanding of battery degradation.
Based on the analysis of physical and chemical degradation
mechanisms, battery degradation is theoretically modelled in
terms of capacity loss [19]-[21] and power fade [19], [20].
Battery design parameters such as electrode thickness, particle
radius and porosity are considered in the theoretical
electrochemical models, and their effects on battery
degradation can be examined [19], [20]. Theoretical models
have high accuracy, and allow extrapolation to different
battery type or design. However, due to different causes and
inter-dependencies, degradation mechanisms are difficult to
model. Most theoretical models focus on the dominant
phenomena, such as the formation and growth of SEI [19]-
[21]. Moreover, theoretical models are normally complex, and
their accuracy is dependent on the availability and the
accuracy of battery design parameters.
2) Empirical Models
Degradation models are generally developed by curve
fitting to a large amount of datasets. Most empirical models
address either calendar ageing or cycle ageing; some work
combine both ageing effects on battery degradation. In terms
,(((0LODQ3RZHU7HFK
of indicators, most research works focus on the capacity fade;
but a few evaluate the power fade as well. Commonly used
empirical models include Arrhenius-based models, cycle
counting models, Ah/Wh-throughput models, other regression
models and artificial neural network (ANN) based models.
a) Arrhenius-based Models
The Arrhenius-based model is the most widely used
battery degradation model. It captures the dependence of the
battery degradation rate on stress factors due to calendar
ageing [11], [15], [22], [23] or cycle ageing [15], [23]-[26].
The model is with a basic formulation representing the
dependency of ageing rate k on temperature T as:
a
E
RT
kAe
−
=
(1)
where A, E
a
, R are the pre-exponential factor, the activation
energy and the Boltzmann constant respectively.
Most works expand the basic model to consider more
stress factors in the capacity fade analysis. Besides the
temperature, calendar time and battery SOC are included in
the capacity fade model due to calendar ageing [15], [22],
[23]. An example expression is as follows [15]:
()
()
a
cal
E
z
RT
cal cal
CASOCet
−
Δ=
(2)
where ǻC
cal
is the capacity variance due to the calendar
ageing; A
cal
(SOC) is the pre-exponential factor depending on
SOC; t is the calendar time; z
cal
is a fitting parameter.
Similarly, in the modelling of capacity fade due to cycle
ageing, the Arrhenius-based model is expanded to include the
temperature, current rate, DOD, number of cycles, and Ah-
throughput [15], [23]-[26]. An example expression in [24],
[25] is:
()
()
()
cyc
EC
z
RT
cyc cyc
CACe Ah
−
Δ=
(3)
where ǻC
cyc
is the capacity variance due to the cycle ageing;
A
cyc
(C) is the pre-exponential factor depending on the current
rate C; E(C) is the activation energy depending on the current
rate C; Ah is the battery Ah-throughput relating to DOD; z
cyc
is
a fitting parameter.
In [22], Arrhenius law is also applied to represent the
dependence of battery power fade due to calendar ageing on
the temperature, time and voltage:
()
()
a
cal
E
y
RT
cal cal
R
BVe t
−
Δ=
(4)
where ǻR
cal
is the resistance variance due to the calendar
ageing; B
cal
(V) is the pre-exponential factor depending on the
battery cell voltage; t is the calendar time; y
cal
is a fitting
parameter.
Note that Arrhenius-based models can also be categorized
as semi-empirical approaches, as they are basically physical
equations combined with parameter estimation [16].
b) Cycle Counting Models
Cycle counting models assess battery degradation in terms
of the number of cycles a battery can withstand until the end
of its lifetime [27], [28]. The basis of these models is the high
dependency of battery lifetime on DOD. Higher DOD value
results in more severe battery degradation, and thus the battery
RUL is shorter.
DOD is the main factor considered in the cycle counting
models. To apply this approach, a data sheet of number of
cycles to failure at some DOD levels for the considered
battery type, and battery charging/discharging profiles are
required. First, based on the battery charging/discharging
profile, number of cycles that the battery has experienced for
each DOD level is calculated using cycle counting methods,
such as the rain-flow counting method. Secondly, an equation
to express the maximum number of cycles that the battery can
endure until its EOL versus DOD is obtained by curve fitting
to the data sheet value. Finally, battery lifetime loss L
loss
is
estimated by summing-up the incremental loss of lifetime
caused by the different cycles to which the battery is subject:
loss
1
m
ii
i
L
NCF
=
=
¦
(5)
where m is the number of DOD levels; N
i
is the number of
cycles the battery has experienced at DOD level i; CF
i
is the
number of cycles to failure at DOD level i. The battery
reaches the end of its lifetime when the lifetime loss equals to
one.
c) Ah/Wh-throughput Models
Ah/Wh-throughput models link the battery capacity fade to
the severity of charging/discharging transfer events, as results
show a linear relationship between the battery capacity fade
and the Ah/Wh-throughput [29]. In these models, the Ah/Wh
that goes through the battery is counted first. To estimate
battery RUL, the total Ah/Wh-throughput is compared to the
predefined Ah/Wh-throughput value that a battery can
withstand until the end of its lifetime [28], [30]. The lifetime
loss, defined as the relative energy capacity fade, is calculated
by:
loss drv drv cha cha norm
()
LEEC
αα
=+
(6)
where E
drv
and E
cha
are daily energy processed during driving
and charging respectively; Į
drv
and Į
cha
are the relative driving
and charging energy capacity fade coefficients, and are
obtained by curve fitting to the experimental data; C
norm
is the
battery nominal capacity at the beginning of its lifetime. The
battery reaches the end of its lifetime when the lifetime loss
equals to one. Ah/Wh-throughput during discharging due to
V2G services can be included in (6) to consider V2G impacts
on the battery lifetime loss.
d) Regression Models
Some degradation models apply traditional regression
methods to capture the relationship between capacity/power
fade and stress factors [18], [22], [31]. Stress factors and
degradation metrics to be considered in the model are first
chosen. The formulation of degradation model, such as double
quadratic expressions [31], is chosen according to available
,(((0LODQ3RZHU7HFK
data. Corresponding parameters in the model that are the best
fit of the experimental data are then determined using
regression methods.
e) Artificial Neural Network (ANN)-based Models
Instead of developing a mathematic expression, ANN,
such as self-organizing maps [32] and dynamically driven
recurrent networks [33], analyzes the relationship between
input variables (stress factors) and outputs (degradation
metrics) through supervised or unsupervised training. With the
aid of ANN, robust degradation models are developed, and
on-board estimation of battery SOH can be achieved [32],
[33]. The accuracy of SOH estimation depends on training
data, thus a large amount of data are needed.
These empirical models are suitable for the integration
with system planning and operation studies, and are flexible to
be expanded to include more stress factors. However, in most
models, parameters are obtained by curve fitting to the
experimental data of a particular battery type, which restricts
the extrapolation to different battery chemistries or designs. In
addition, the aforementioned empirical models are limited in
accuracy due to assumptions considered in these models.
3) Semi-empirical Model
Semi-empirical models are developed based on both
experimental results and ageing mechanism analysis, and can
be considered as a combination of theoretical analysis and
experimental observations [34]-[37]. [34] considers battery
degradation due to charging/discharging cycles, and a semi-
empirical battery wear model is analytically derived from the
cycle life data combined with battery design parameters.
Cycling DOD, battery size and cycle efficiency are included in
the model. Based on the assumption that battery degradation is
dominated by SEI growth, a degradation model is developed
in [35] by combining an electro-thermal model with an
empirical mathematical expression. The model enables
conducting battery degradation analysis under different
electrical and thermal conditions. Reference [36] incorporates
linear degradation models with SEI film formation models,
and the proposed model is applied to the off-line lifetime
assessment [36] and the system-level optimization with other
criterions [37].
Compared to empirical models, semi-empirical models
compromise on computing time for accuracy, and has an
advantage in terms of the extrapolation to different battery
chemicals and designs. However, due to the model
complexity, semi-empirical models are usually implemented
in off-line analysis.
To sum up, a comparison of models in terms of indicators,
the considered degradation origins, model accuracy and
complexity, data dependency and suitable applications is
presented in Table I [15], [19]-[37]. In the analysis of battery
degradation mechanisms, the accuracy of a model is the
principal criterion to be considered, thus theoretical models
are preferred. For the system planning and operation analysis,
on the contrary, less complex models are preferred in order to
incorporate the battery degradation analysis with system-level
planning and operation studies. Both empirical models and
semi-empirical models can be applied to off-line battery
lifetime assessment. In the applications of on-board SOH
estimation where computing time is the main criterion,
empirical models are more suitable than other models.
TABLE I. S
UMMARY OF
B
ATTERY
D
EGRADATION
M
ODELS
Models Indicators
Degradation
Origins
Captured
Accuracy
Complexity of
Model
Implementation
Data
Dependency
Suitable
Applications
Theoretical models Capacity loss;
Resistance increase
Calendar ageing;
Cycle ageing High High Low Mechanism analysis
Empirical
models
Arrhenius-based
models
Capacity loss;
Resistance increase
Calendar ageing;
Cycle ageing Low Low Medium
System planning and
operation analysis;
On-board estimation
Cycle counting
models RUL Cycle ageing Low Low Medium
Ah/Wh-throughput
models RUL Cycle ageing Low Low Medium
Other regression
models
Capacity loss;
Resistance increase
Calendar ageing;
Cycle ageing Low Low Medium
ANN-based models SOH Calendar ageing;
Cycle ageing Medium Low High
Semi-empirical models Capacity loss;
Resistance increase
Calendar ageing;
Cycle ageing Medium Medium Medium System planning and
operation analysis
IV. V2G IMPACT ANALYSIS
Additional discharging phases arise due to V2G services
besides normal battery operation, and EV users suspect that
V2G services accelerate battery degradation. To quantify the
impacts of offering V2G services on EV battery degradation,
sufficient inputs, ‘fit for purpose’ models and comprehensive
measurements are needed. The general process to evaluate
V2G impacts on battery degradation is summarized in Fig. 2.
Battery parameters, driving patterns, charging regimes and
V2G scenarios are transferred to variables such as energy
throughput in battery degradation models. Depends on the
model expressions, battery degradation is quantified by SOH,
capacity loss, resistance increase or degradation cost.
,(((0LODQ3RZHU7HFK
Inputs:
• Batte r y par am et ers;
• Driving patterns;
• Charging regimes;
• V2G scenarios.
Degradation modelling:
• Empi rical mode l;
• Or semi-empirical
model.
Outputs:
• Battery SOH;
• Capacity/power fade;
• Degradation cost;
• Other metrics.
Figure 2. Evaluation process of V2G impacts on battery degradation
Research works have been conducted on the impact
analysis of V2G services on battery degradation, besides other
impact factors such as driving patterns and charging strategies.
Due to the differences in battery test data and degradation
models, different conclusions have been drawn in terms of
how V2G services affect battery degradation, from acceptable
impacts [15], [24], [38] to detrimental impacts [29].
Nevertheless, these studies share some common findings as
follows.
(1) Different battery chemistries exhibit different
behaviors when providing V2G services. Reference [15]
evaluates the impact of V2G services on capacity loss of two
different battery technologies, with temperature, SOC and C-
rate as stress factors. It is found that providing V2G services
once per day slightly increases the capacity fade of NCA/C
lithium-ion battery; but causes lower ageing for LFP/C
lithium-ion battery compared to no-V2G scenarios.
(2) Battery degradation is sensitive to charging/discharging
regimes. Reference [24] examines the influence of offering
bulk energy services and ancillary services on battery
degradation. According to simulation results, more battery
replacements over vehicle lifetime arise due to V2G services,
but associated degradation is minimized by changing
charging/discharging regimes, such as restricting the service
time and extent.
(3) Different V2G services as well as service frequency
result in different impacts on battery degradation [25], [29],
[38]. Compared to demand response and frequency regulation,
providing net load shaping for the same service frequency has
much greater impact on battery degradation [25], [38]. Service
frequency is found to have significant effects on battery
degradation. Capacity loss is doubled when the V2G service
frequency increased from once per day to twice per day [29].
Unlike cycle ageing, battery degradation in the idle state is
always neglected by EV users. If an EV is mostly in the idle
state, calendar ageing overtakes cycle ageing and dominates
battery degradation. Studies show that a higher degradation
rate reveals at high SOC level when the battery is in storage
[9]. In reality, EV users prefer to maintain a high SOC level.
One of the interesting findings in the ‘My Electric Avenue’
trails is that most users charge their EVs after work and end
charging events with a high final SOC [39]. If V2G services
are applied appropriately, a balance between reducing the
storage related degradation and increasing the cycling related
degradation could be reached. In this case, battery degradation
might be minimized. Reference [18] proposes an algorithm to
minimize battery degradation by optimizing V2G cycling
when EV is in idle state. The algorithm limits battery
degradation by modifying SOC to a value when storage
related degradation cost is minimized and determining the
cycling region where V2G cycling associated degradation cost
is minimized. The EV will provide V2G services only when
the degradation caused by V2G cycling is less than storage
degradation. Thus, under this management algorithm, the
worst case is that battery degrades as if there was no V2G
[40]. Compared to the reference case where no-V2G is
implemented and EVs are recharged to 100% at night, the
proposed management method reduces battery capacity fade
and power fade by up to 9.1% and 12.1% respectively [18]. It
is worthy to note that intelligent V2G systems with smart
meters and two-way controllers are the prerequisite for this
battery cycling management method.
V. CONCLUSION
A review of research works on V2G impacts on battery
degradation is presented in this paper, together with a
summary of battery degradation studies in terms of
mechanisms, measurements and modeling. Apart from
infrastructure and technology development, accurately
quantifying how V2G affects battery degradation is an
important issue for the fulfillment of V2G deployment. To
achieve this, both battery degradation modeling and active
battery management need improvements.
Convincing study of V2G impacts on battery degradation
relies on accurate battery degradation models. A better
understanding of battery degradation causes and mechanisms
is required for the development of more accurate models.
Moreover, comprehensive measurements of battery health
status are needed. Possible improvements include proposing
new metrics to measure battery degradation status
comprehensively, and modifying the SOH metric to
incorporate the analysis of driving patterns and the prediction
of user’s expectations into battery degradation estimation.
Many factors, including battery technologies, driving
patterns and regulations influence the economic revenue of
V2G services, and can be included in the overall analysis of
V2G impact. In addition, the effectiveness of improving
battery lifetime by actively cycling the battery with V2G
services needs to be validated on various battery types and
battery usage scenarios. Active management methods that
mitigate the impact of V2G services on battery degradation
have to be adjusted accordingly.
One of the ultimate purposes of modeling battery
degradation and analyzing V2G impacts is to achieve better
system-level energy management. A communication and
control platform, which aggregates information at different
substations and EVs in the system and controls the
bidirectional flow of power from EV batteries, is needed. Part
of the platform is an optimized charging/discharging regime
that minimizes the overall cost of a system (e.g. a building)
with the consideration of battery characteristics, users’ driving
patterns, V2G requirements and network constraints. Another
part of the platform is a control algorithm for bi-directional
power flow to achieve active control of the start time, the end
time, the period, the power, and the frequency of batteries
charging/discharging.
REFERENCES
,(((0LODQ3RZHU7HFK
[1] N. Karali, A. Gopal, D. Steward, E. Connelly, and C. Hodge, “Vehicle-
grid integration – A global overview of opportunities and issues,”
Lawrence Berkeley National Lab. and National Renewable
Energy Lab., Tech. Rep., June 2017.
[2] M. MacLeod and C. Cox, “V2G market study – Answering the
preliminary questions for V2G: What, where and how much?” CENEX,
Loughborough, UK, Tech. Rep., July 2018.
[3] C. Curry, “Lithium-ion battery costs and market,” Bloomberg New
Energy Finance, Tech. Rep., July 2017.
[4] S. Habib, M. Kamran, and U. Rashid, “Impact analysis of vehicle-to-
grid technology and charging strategies of electric vehicles on
distribution networks – A review,” J. Power Sources, vol. 277, pp.
205–214, 2015.
[5] L. Lu, X. Han, J. Li, J. Hua, and M. Ouyang, “A review on the key
issues for lithium-ion battery management in electric vehicles,” J.
Power Sources, vol. 226, pp. 272–288, 2013.
[6] Cell, module, and pack for EV applications. Automotive Energy Supply
Corp., Zama, Japan. [Online]. Available: http://www.eco-aesc-
lb.com/en/product/liion_ev/
[7] A. Barré, B. Deguilhem, S. Grolleau, M. Gérard, F. Suard, and D. Riu,
“A review on lithium-ion battery ageing mechanisms and estimations
for automotive applications,” J. Power Sources, vol. 241, pp. 680–689,
2013.
[8] C. Birkl, M. Roberts, E. McTurk, P. Bruce, and D. Howey,
“Degradation diagnostics for lithium ion cells,” J. Power Sources, vol.
341, pp. 373–386, 2017.
[9] S. Pelletier, O. Jabali, G. Laporte, and M. Veneroni, “Battery
degradation and behaviour for electric vehicles: Review and numerical
analyses of several models,” Transp. Res. Part B Methodol., vol. 103,
pp. 158–187, 2017.
[10] X. Si, “An adaptive prognostic approach via nonlinear degradation
modeling: Application to battery data,” IEEE Trans. Ind. Electron., vol.
62, no. 8, pp. 5082–5096, 2015.
[11] J. Christophersen, I. Bloom, E. Thomas, and V. Battaglia, “Battery
calendar life estimator manual,” Idaho National Lab., Idaho Falls,
Idaho, Tech. Rep. INL-EXT-08-15136, Oct. 2012.
[12] S. Saxena, C. Floch, J. Macdonald, and S. Moura, “Quantifying EV
battery end-of-life through analysis of travel needs with vehicle
powertrain models,” J. Power Sources, vol. 282, pp. 265–276, 2015.
[13] M. Berecibar, I. Gandiaga, I. Villarreal, N. Omar, J. Mierlo, and P.
Bossche, “Critical review of state of health estimation methods of Li-
ion batteries for real applications,” Renew. Sustain. Energy Rev., vol.
56, pp. 572–587, 2016.
[14] M. Coleman, W. Hurley, and C. Lee, “An improved battery
characterization method using a two-pulse load test,” IEEE Trans.
Energy Convers., vol. 23, no. 2, pp. 708–713, 2008.
[15] M. Petit, E. Prada, and V. Sauvant-Moynot, “Development of an
empirical aging model for Li-ion batteries and application to assess the
impact of vehicle-to-grid strategies on battery lifetime,” Appl. Energy,
vol. 172, pp. 398–407, 2016.
[16] J. Jaguemont, L. Boulon, and Y. Dubé, “A comprehensive review of
lithium-ion batteries used in hybrid and electric vehicles at cold
temperatures,” Appl. Energy, vol. 164, pp. 99–114, 2016.
[17] M. Swierczynski, D. Stroe, A. Stan, R. Teodorescu, and S. Kær,
“Lifetime estimation of the nanophosphate LiFePO4/C battery
chemistry used in fully electric vehicles,” IEEE Trans. Ind. Appl., vol.
51, no. 4, pp. 3453–3461, 2015.
[18] K. Uddin, T. Jackson, W. Widanage, G. Chouchelamane, P. Jennings,
and J. Marco, “On the possibility of extending the lifetime of lithium-
ion batteries through optimal V2G facilitated by an integrated vehicle
and smart-grid system,” Energy, vol. 133, pp. 710–722, 2017.
[19] E. Prada, D. Domenico, Y. Creff, J. Bernard, V. Sauvant-Moynot, and
F. Huet, “A simplified electrochemical and thermal aging model of
LiFePO4-Graphite Li-ion batteries: Power and capacity fade
simulations,” J. Electrochem. Soc., vol. 160, no. 4, pp. A616–A628,
2013.
[20] R. Ahmed, M. Sayed, I. Arasaratnam, J. Tjong, and S. Habibi,
“Reduced-order electrochemical model parameters identification and
state of charge estimation for healthy and aged Li-ion batteries—Part
II: Aged battery model and state of charge estimation,” IEEE J. Emerg.
Sel. Top. Power Electron., vol. 2, no. 3, pp. 678–690, 2014.
[21] J. Purewal, J. Wang, J. Graetz, S. Soukiazian, H. Tataria, and M.
Verbrugge, “Degradation of lithium ion batteries employing graphite
negatives and nickel-cobalt-manganese oxide + spinel manganese oxide
positives: Part 2, chemical-mechanical degradation model,” J. Power
Sources, vol. 272, pp. 1154–1161, 2014.
[22] A. Marongiu, M. Roscher, and D. Sauer, “Influence of the vehicle-to-
grid strategy on the aging behavior of lithium battery electric vehicles,”
Appl. Energy, vol. 137, pp. 899–912, 2015.
[23] E. Sarasketa-Zabala, E. Martinez-Laserna, M. Berecibar, I. Gandiaga,
L. Rodriguez-Martinez, and I. Villarreal, “Realistic lifetime prediction
approach for Li-ion batteries,” Appl. Energy, vol. 162, pp. 839–852,
2016.
[24] J. Bishop, C. Axon, D. Bonilla, M. Tran, D. Banister, and M.
McCulloch, “Evaluating the impact of V2G services on the degradation
of batteries in PHEV and EV,” Appl. Energy, vol. 111, pp. 206–218,
2013.
[25] M. Jafari, A. Gauchia, S. Zhao, K. Zhang, and L. Gauchia, “Electric
vehicle battery cycle aging evaluation in real-world daily driving and
vehicle-to-grid services,” IEEE Trans. Transp. Electrif., vol. 4, no. 1,
pp. 122–134, 2017.
[26] X. Han, M. Ouyang, L. Lu, and J. Li, “A comparative study of
commercial lithium ion battery cycle life in electric vehicle: Capacity
loss estimation,” J. Power Sources, vol. 268, pp. 658–669, 2014.
[27] I. Moghaddam, B. Chowdhury, and S. Mohajeryami, “Predictive
operation and optimal sizing of battery energy storage with high wind
energy penetration,” IEEE Trans. Ind. Electron., vol. 65, no. 8, pp.
6686–6695, 2018.
[28] J. Barreras, C. Pinto, R. Castro, E. Schaltz, S. Andreasen, P.
Rasmussen, and R. Araújo, “Evaluation of a novel BEV concept based
on fixed and swappable Li-ion battery packs,” IEEE Trans. Ind. Appl.,
vol. 52, no. 6, pp. 5073–5085, 2016.
[29] S. Peterson, J. Apt, and J. Whitacre, “Lithium-ion battery cell
degradation resulting from realistic vehicle and vehicle-to-grid
utilization,” J. Power Sources, vol. 195, no. 8, pp. 2385–2392, 2010.
[30] C. Shiau, N. Kaushal, C. Hendrickson, S. Peterson, J. Whitacre, and J.
Michalek, “Optimal plug-in hybrid vehicle design and allocation for
minimum life cycle cost, petroleum consumption and greenhouse gas
emissions,” J. Mech. Des., vol. 132, pp. 183–195, 2010.
[31] M. Dubarry, A. Devie, and K. McKenzie, “Durability and reliability of
electric vehicle batteries under electric utility grid operations:
Bidirectional charging impact analysis,” J. Power Sources, vol. 358,
pp. 39–49, 2017.
[32] I. Fernández, C. Calvillo, A. Sánchez-Miralles, and J. Boal, “Capacity
fade and aging models for electric batteries and optimal charging
strategy for electric vehicles,” Energy, vol. 60, pp. 35–43, 2013.
[33] H. Chaoui and C. Ibe-ekeocha, “State of charge and state of health
estimation for lithium batteries using recurrent neural networks,” IEEE
Trans. Veh. Technol., vol. 66, no. 10, pp. 8773–8783, 2017.
[34] S. Han, S. Han, and H. Aki, “A practical battery wear model for electric
vehicle charging applications,” Appl. Energy, vol. 113, pp. 1100–1108,
2014.
[35] M. Ecker, J. Gerschler, J. Vogel, S. Käbitz, F. Hust, P. Dechent, and D.
Sauer, “Development of a lifetime prediction model for lithium-ion
batteries based on extended accelerated aging test data,” J. Power
Sources, vol. 215, pp. 248–257, 2012.
[36] B. Xu, A. Ouda lov, A. Ulbig, G. Andersson, and D. Kirsch en,
“Modeling of lithium-ion battery degradation for cell life assessment,”
IEEE Trans. Smart Grid, vol. 9, no. 2, pp. 1131–1140, 2018.
[37] B. Foggo and N. Yu, “Improved battery storage valuation through
degradation reduction,” IEEE Trans. Smart Grid, vol. 9, no. 6, pp.
5721–5732, 2018.
[38] D. Wang, J. Coignard, T. Zeng, C. Zhang, and S. Saxena, “Quantifying
electric vehicle battery degradation from driving vs. vehicle-to-grid
services,” J. Power Sources, vol. 332, pp. 193–203, 2016.
[39] J. Quirós-Tortós, L. Ochoa, and T. Butler, “How electric vehicles and
the grid work together: Lessons learned from one of the largest electric
vehicle trials in the world,” IEEE Power Energy Mag., vol. 16, no. 6,
pp. 64–76, 2018.
[40] K. Uddin, M. Dubarry, and M. Glick, “The viability of vehicle-to-grid
operations from a battery technology and policy perspective,” Energy
Policy, vol. 113, pp. 342–347, 2018.
,(((0LODQ3RZHU7HFK