Energy Management Strategy to Coordinate Batteries

and Ultracapacitors of a Hybrid Energy Storage

System in a Residential Prosumer Installation

M. Ruiz-Cortés, E. Romero-Cadaval, C. Roncero-Clemente, F. Barrero-González, E. González-Romera

Power Electrical and Electronic Systems (PE&ES)
University of Extremadura
Badajoz, Spain

Abstract—In recent years, power systems have experimented addition, prosumers which are equipped with control and
a great development as a consequence of the apparition of the communication systems can provide services to the power
distributed generation. Renewable energy sources, active users system, such as the increase of the penetration of intermittent
and electric vehicles play a fundamental role in power systems, renewable energy sources (like photovoltaic and wind sources)
which has supposed a new topic research. In this scenario, the into the grid, the load/generation shifting, ancillary services
figure of prosumer has gained relevance due to the multiple and grid system applications [3], [4]. Thus, it is possible to
advantages that they can contribute to improve the behavior of decrease both the required peak power generation and the
power systems and to achieve energy and cost savings. In charge on the electric grid, and therefore a reduction of the
particular, special attention is focus on photovoltaic (PV) arrays
usage of fossil fuels [5]; moreover, it allows that end users
integrated in residential buildings with energy storage systems
achieve lower electricity bills and to optimize the utilization of
(ESSs), such as batteries or ultracapacitors (UCs) or even a
combination of both in a hybrid energy storage system (HESS).
renewable energy sources [4].
This paper presents an energy management strategy to exchange The renewable energy source which has experimented de
energy between the batteries and the UCs that form a HESS greatest development in the last years has been the PV
installed with a PV array in a prosumer dwell in order to generation [6], especially in residential and commercial
maintain its UC bank available to support the energy installations connected to low voltage distribution networks
requirements during the most time possible. [2], [7]. The growth of small PV plants has been considerably
Keywords—prosumer; battery; ultracapacitor; hybrid energy
high in urban areas, where they use to be integrated to the roofs
storage system; energy management strategy of buildings [8], instead of occupy large land areas, and allow
users can benefit from electricity cost savings [2].
I. INTRODUCTION In spite of the multiple advantages that PV systems offer,
they also present some issues. In residential applications, the
In last years, the proliferation of distributed generation has
peak power consumption is usually produced in the evening;
caused power systems have to face multiple challenges [1] in
however, the peak power generation is usually produced at
order to integrate renewable energy sources, electric and hybrid
noon; hence, a mismatch between peak PV power generation
vehicles and active users into the grid. Distributed generation
and peak power consumption times occurs [5]. At this
needs to solve several technical issues associated with it, such
disadvantage, we have to add the intermittent and
as electromagnetic interference, harmonic distortion, problems
unpredictable character of PV generation as a result of its
in voltage and frequency regulation or islanding [2]. As a
dependence of the solar irradiance. These problems can be
possible solution to mitigate the problems exposed below, a
mitigated incorporating an ESS associated to the PV
new concept about the role of end users has appeared recently:
installations, whose main purpose is to store the excess of
end users have been called to become prosumers, instead of
power produced by the PV system in order to deliver the
being considered only as consumers.
necessary energy to satisfy the user demand when PV power is
A prosumer can be defined as a combination of a producer insufficient [7]. ESS can be designed using different
and a consumer. It means that a prosumer is able both to extract technologies and even combining up to one of them to
power from the grid (like a conventional consumer) and to compose a hybrid energy storage system (HESS) [9].
deliver power into the grid (like a producer) [3]. At
This paper focuses on simulating by Simulink/Matlab
installations of prosumers, loads and energy sources coexist, in
software the operation of a residential prosumer installation
a way that the energy sources provide energy to feed their own
with a PV array and a HESS based on batteries and
loads or to inject power to the electric grid according to
ultracapacitors, as shown in Fig. 1. A brief explanation about
economic decisions. For this reason, a large number of
ESS technologies and HESS configurations that can be
prosumers can contribute to smooth the energy demand curve,
integrated in PV systems is offered in Section II. In Section III,
and, therefore, to reduce the required power generation. In

978·1·5090-4639·31171$31.00 Q2017 IEEE 20171ntemationai Young Engineers Forum (YEF··ECE)

Caparica, Portugal, May 5, 2017
30
the analyzed case study is described, showing the typical power integrate a PHS based ESS with a residential PV installation.
generation and consumption curves of the prosumer. In The most common technologies which have been proposed to
addition, in Section IV an energy management system (EMS) be combined with PV systems are batteries, UCs or fuel
is proposed in order to maintain UCs voltage in a desired level cells [12].
within the HESS by means of exchanging energy between the
In the design of a ESS associated to a PV system it is
batteries and the UCs. Simulation results obtained for the case
important to take account that PV power is composed by a low-
study are discussed in Section V in order to evaluate if the
frequency component and high-frequency oscillations
proposed control strategy is valid. Finally, in Section VI, the
associated to the intermittency of the solar irradiance. Thus, in
main conclusions which have been obtained in this work
order to smooth the low-frequency oscillations, a ESS with
are exposed.
high energy density is required, while ESS with high power
density is also required to provide or absorb the high-power
transients [13].
Batteries are the most employed ESS with PV systems, and
they are able to store acceptable amounts of energy, although
they can not deliver large amount of energy in a very short time
[14]. On the other hand, UCs can supply high peaks of energy
and can be charged and discharged without missing of
efficiency for thousands of cycles, but their storage capacity is
low [14]. Both technologies can be combined to add the high
energy density of batteries to the high power density of UCs
[9], [11], [13], [14]. This combination provides another
advantage, because the inclusion of UCs can contribute to
improve the batteries lifetime due to they are responsible for
provide high power demands [9]. This combination of ESS
Fig. 1. Case study scenario.
technologies makes up a hybrid energy storage system
(HESS) [9]. Hence, a HESS based on batteries and UCs is
II. ESSS INTEGRATED IN PV SYSTEMS characterized by a fast response and a large energy
capacity [15].
ESSs have become in an interesting topic to move forward
in research in recent years. In the case of residential PV Different topologies of HESSs exist. In [16] and [17], a
systems, ESSs are destined to store the surplus of power from review of the most widely used HESS topologies is given,
the PV system during noon time and hence reducing the energy which are classified in passive and active configurations with
loss associated with the excess power from the solar PV array different possibilities concerning the number of power
[7] and decreasing the injection in the grid from the PV system converters and the position of the used storage devices.
[1], due to an uncontrolled power supply to the grid could However, one of the most used HESS topologies is the parallel
cause issues. This stored energy could be used later to feed the configuration of the battery/ultracapacitor HESS with
load when PV supply is not sufficient, serving as support to the bidirectional converters (multiple converter configuration). The
grid and reducing the electric consumption of the user. use of bidirectional converters allows managing each ESS
Therefore, ESSs allow the displacement of the energy independently and even exchanging the energy between the
consumption of customers, in a way that they only inject when ESSs [9], for example, the batteries can charge and discharge
the energy is needed most and charge their ESSs when there is the UCs, which is very useful to guarantee that UCs are
an excess of generation or the price of the energy are cheaper available to support batteries at all times. In addition, this
than in other periods [1]. Hence, ESSs contribute to smooth the topology allows to maintain both the battery and the UC
load consumption curve and to improve the energy voltage lower than the dc-link voltage, which avoids balancing
management [1]. problems; also, the UC voltage can vary in a wide range,
therefore, the capacitor is fully used, but two converters are
Multiple technologies can be used in a ESS. Some of them required [16].
are the next [10]: pumped hydroelectric storage (PHS),
compressed air energy storage (CAES), batteries, fuel cells,
flow batteries, solar fuels, superconducting magnetic energy III. CASE STUDY
storage (SMES), flywheels, UCs or thermal energy storage The proposed case study has been represented in Fig. 1. It
(TES). Each one of them presents different characteristics consists on a residential prosumer installation connected to a
regarding technical maturity, power and energy density, power low distribution network. It has a small PV array whose peak
rating and discharge time, storage duration, capital cost, cycle PV power is equal to 3 kWp. This prosumer is able to store
efficiency, life time and life cycle, influence on environment energy in a HESS based on the combination of lead-acid
and size. Because of this, it is essential to decide what batteries and UCs. The HESS has been dimensioned in order to
technology is the most suitable for the application in question batteries have three days of autonomy [18]. The characteristics
[11]. For example, PHS is a mature technology with large of the PV array and the HESS are collected in Table I. The
volume, long storage period, high efficiency and relatively low typical diary PV generation and load consumption profiles of
capital cost per unit of energy [10], but it is unfeasible to the analyzed prosumer are shown in Fig. 2. These curves have

976-1-5090-4639-31171$31,00 C 2017 IEEE 20171nlemalional Young Engineers Forum (YEF~ECE )

Caparica, POfIugal, May 5, 2017
31
been obtained through the simulation of provided data using Pref , sto = Pload − PPV (1)
Simulink/Matlab software.

TABLE I. SIMULATION PARAMETERS (

Pbat = LPF Pref , sto ) (2)
Parameter Value
Peak PV power 3 kWp PUC = Pref , sto − Pbat (3)
Battery bank capacity 2000 Ah

Ultracapacitor bank capacity 600 F

HESS devices voltage 48 V

Fig. 3. Simulink diagram used to calculate Pbat and PUC.

state to the discharged state [20]. Thus, the SOC of a battery

Fig. 2. PV generation and load consumption curves of the prosumer. (SOCbat) means the ratio between the remaining capacity (Q)
and the full charge or nominal charge (Qnom), as it is shown in
(4). Hence, a SOC equal to 100% implies that the battery is
IV. PROPOSED ENERGY MANAGEMENT STRATEGY fully charged, whereas a SOC equal to 0% indicates that the
In order to the devices of a HESS could interact properly battery is fully discharged [20]. However, the SOC of a battery
with the PV system and even between themselves, an EMS is is usually calculated following (5), where SOCbat,0 and i are the
required. As we mentioned in Section II, batteries are suitable initial SOC and the output current of the battery respectively,
to provide the bulk of energy during large intervals of time, but and it is expressed in Ampere-hour (Ah).
its response is slow, and, because of this, UCs are responsible
for supplying the power peak demands and absorbing the PV
peak power excess, due to their fast response. For this reason, it Q
SOCbat = (4)
is needed to separate the low-frequency component of the Qnom
power assumed by the HESS, which is destined to the battery,
and the high-frequency component, which is took on the UC.
idt
This objective can be reached using a conventional low- SOCbat = SOCbat ,0 − (5)
pass filter [13], that distribute the power among the storage Q
devices depending on their response times: the low frequency
power part is assigned to the long-term storage device Moreover, the lifespan of a battery can be damaged if it
(batteries in this case), and the rest is absorbed/supplied by the works out of a certain working range [21]. Because of this, a
short-term storage device (UCs in this case) [17]. This power limitation of the SOC of the batteries has been considered in
sharing is represented by (1), (2) and (3), where the power that this paper. In this way, batteries are able to work only within a
have to be provided by the HESS (Pref,sto) is calculated as the range whose upper limit has been established in a value of
difference between the demanded power (Pload) and the PV 100% and the lower limit is equal to 20%, according to lead-
generated power (PPV), whereas Pbat and PUC are the power acid batteries characteristics [18]. Thus, it is possible to avoid
assigned to the batteries bank and the UCs bank respectively. the over-charging and the over-discharging of the batteries.
Fig. 3 shows the block diagram used in Simulink in order to
model the previous equations, and it has been designed using On the other hand, the stored energy in an UC (EUC) can be
the guidelines of [19]. calculated using (6), where V is the voltage of the UC in a
certain instant and C is the UC capacity [22]. As we can see in
As it is mentioned in [17], the explained power distribution (6), an UC stores energy proportionally to the square of its
method is not enough to coordinate the HESS, but it can be voltage [23]. UCs are usually interfaced with the rest of the
used as a first power distribution procedure. Also, an EMS installation by a power converter, which needs a minimum
must have into account the state of charge (SOC) of the storage operating voltage and, therefore, a minimum energy [24]. Thus,
devices that form the ESS or the HESS. the SOC of an UC (SOCUC) can be expressed as it is shown in
(7), where E is the stored energy in the UC at a certain
In the case of batteries, their capacity is defined as the
moment, whereas Emax and Emin are the maximum and the
quantity of electrical charge in the cell from the fully charged

976,'·5090-4639·31171$31,00 C 2017 IEEE 20171nlemalional Young Engineers Forum (YEF~ECE )

Caparica, Portugal, May 5, 2017
32
minimum energy that the UC is able to store. In this way, the result of combining (6), (7) and (8). Thus, in the analyzed
SOCUC is the ratio between the available energy in a case study, Vref must be around 37.95 V. Fig. 4 depicts the
determined instant and the total energy that the UC is able relation between SOCUC and V in a general case.
to provide.
2
1 V
1 SOCUC = ⋅ 4⋅ −1 (9)
EUC = ⋅ C ⋅V 2 (6) 3 Vnom
2

100
E − Emin
SOCUC = (7)
Emax − Emin

SOC (%)
50
Using (7), when the UC voltage reaches the 50% of its
nominal value, SOCUC is equal to 0%. It means that an UC is
considered as fully discharged when its voltage is the half of its 0
nominal value (Vnom), in order to maintain a minimum energy 0 Vnom/2 Vref Vnom
equal to 25% of the total energy that the UC could store (8) and Voltage (V)
ensure the minimum operating voltage required by the
power converter. Fig. 4. Evolution of SOCUC with V.

In order to achieve this purpose, the HESS batteries can

1 V
2
1 1 cooperate with the UC bank thanks to a coordination realized
⋅ ⋅ C ⋅ (Vnom ) = 0.25Emax
2
Emin = ⋅ C ⋅ nom = (8) by the EMS [9]. In this way, when the UC voltage falls below
2 2 4 2 Vref, the EMS generates a current reference through a PI
controller to charge the UC from the battery until the UC
With the aim of guaranteeing that the UC bank is always reaches Vref again and vice versa. With this control action, the
available to provide or absorb the required power, it is UC remains at every moment able to provide energy (positive
necessary to maintain SOCUC in a certain level [13]. In this or negative), avoiding that it suffers high charge fluctuations.
paper, we have established the desired SOCUC equal to 50%.
As a consequence, we need to calculate the corresponding Fig. 5 shows the model built in Simulink in order to
reference voltage (Vref) to a SOCUC equal to 50%. This can be simulate the HESS and its energy management to exchange
achieved using (9), which relates SOCUC with V and Vnom and is energy between batteries and UCs. As we can see in this figure,
we have used the battery and UC models that Simulink

Fig. 5. Simulink model used to simulate the battery and UC banks work with the proposed control.

976·'·5090-4639·31171$31,00 C 2017 IEEE 2017 Intemational Young Engineers Forum (YEF~ECE )

Caparica, POfIugal, May 5, 2017
33
 provides in SimPowerSystems Blockset. Both blocks receive energy to the UC when it is necessary in order to SOCUC
their corresponding reference current signals through remains in the desired value. However, the difference between
controlled current sources. The PI controller has been tuned both SOC curves is small.
using the PID Tuner tool, using KP=10 and KI=200.
In the other hand, Fig. 7 shows the current that the battery
and the UC supply respectively during the period contained
V. SIMULATION RESULTS between 7:00 a.m. and 4:00 p.m. As it can be observed, the
The proposed energy management strategy has been tested batteries system delivers the low-frequency component of the
by simulation in the HESS of the studied prosumer to required current, whereas the UCs system provides the required
coordinate the energy flow between the battery and the UC high-frequency component of the current to operate properly
systems that compose the HESS, with SOCbat=65% and V=40 with the PV system. In order to understand the figure, it is
V as initial values. As we can see in Fig. 6, the control strategy important to take into account that we have consider positive
accomplishes the requirement of avoiding UCs suffer currents imply discharge in the ESSs and negative currents
saturation when their SOC reaches 100% and a fully discharge. mean charge.
Instead of this, the voltage of the UC bank is maintained in the
calculated value (Fig. 6(a)), and, as a consequence, SOCUC is
approximately equal to 50% along the simulation time
(Fig. 6(b)). This has been achieved through the exchange of
energy between the battery and the UC banks. Thus, the SOC
of the battery has been affected in the way shown in Fig. 6(c),
which compares the SOC of the battery when the proposed
EMS is used with the case in that it is not applied. In this
figure, it can be observed how the battery provides and absorbs

50 Voltage
Voltage* Fig. 7. Comparison of the currents provided by the battery and the UC
40
systems since 7:00 a.m. to 4:00 p.m.
Voltage (V)

30

20
VI. CONCLUSIONS
10
This paper exposes the relevance that prosumers have
0
reached as a result of the new scenario that the development of
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 distributed generation has caused in recent years. Besides, a
Time (h)
brief discussion about ESSs is given. Namely, the case of a
(a) prosumer dwell with a small PV array combined with a HESS
100 SOC
based on lead-acid batteries and UCs has been simulated.
80
SOC* Moreover, a comprehensive control strategy has been proposed
in order to optimize the energy management and it has been
SOC (%)

60 validated by simulation.
40
The implemented EMS allows maintaining the SOC of the
20 UC bank in the half of their maximum value, which is useful to
0
UCs remain available at all times and even to take advantage of
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 the surplus of the energy stored in UCs to charge batteries.
Time (h)
As further works, it would be interesting to continue
(b)
analyzing the proposed EMS with different PV conditions and
68
SOC load characteristics in order to evaluate its operation in other
66 SOC* possible cases. In addition, this work could be complemented
64 with an economical study which considers the variable energy
SOC (%)

62
prices and allows to determine the economic impact that a
HESS and a EMS can cause in a residential installation.
60

58
ACKNOWLEDGMENT
56
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 This research work has been partially supported by the
Time (h) project (TEC2016-77632-C3-1-R) from “Ministerio de
(c) Economía y Competitividad (Gobierno de España)” and by
Fig. 6. Comparison of the evolution of the voltage (a) and the SOC (b) of the “Junta de Extremadura” with funds for research groups
UC system and the SOC of the batteries bank (c) along a day with the (GR15177).
proposed control action (marked with *) and without it.

976~ '·5090-4639·31171$31,00 C 2017 IEEE 20171nlemalional Young Engineers Forum (YEF~ECE )

Caparica, Portugal, May 5, 2017
34
 REFERENCES and Ultracapacitor With Dynamic Energy Management in Microgrid
Applications,” in IEEE Transactions on Power Electronics, vol. 26, no.
[1] S. M. Souza, M. Gil, J. Sumaili, A. G. Madureira and J. A. P. Lopes, 3, pp. 923-930, March 2011.
“Operation scheduling of prosumer with renewable energy sources and
[14] H. F. Gharibeh, A. S. Yazdankhah and M. R. Azizian, “Improved energy
storage devices,” 2016 13th International Conference on the European
Energy Market (EEM), Porto, 2016, pp. 1-5. management for a power-split multi-source fuel cell vehicle based on
optimal source sizing and regenerative braking,” 2016 IEEE 16th
[2] R. Araneo and M. C. Falvo, “Simulation of a ESS in a prosumer power- International Conference on Environment and Electrical Engineering
plant with a PV system and an EV charging station,” 2016 IEEE 16th (EEEIC), Florence, 2016, pp. 1-6.
International Conference on Environment and Electrical Engineering
(EEEIC), Florence, 2016, pp. 1-5. [15] C. Abbey, K. Strunz and G. Joos, “A Knowledge-Based Approach for
Control of Two-Level Energy Storage for Wind Energy Systems,” in
[3] M. C. Falvo, U. Grasselli, M. Manganelli and A. Modesto, “Small scale IEEE Transactions on Energy Conversion, vol. 24, no. 2, pp. 539-547,
ESS for LV prosumers: An economic feasibility and sensitivity June 2009.
analysis,” 2015 IEEE 15th International Conference on Environment and
Electrical Engineering (EEEIC), Rome, 2015, pp. 950-955. [16] J. Cao and A. Emadi, “A New Battery/UltraCapacitor Hybrid Energy
Storage System for Electric, Hybrid, and Plug-In Hybrid Electric
[4] G. Graditi, M.G. Ippolito, R. Rizzo, E. Telaretti, and G. Zizzo, Vehicles,” in IEEE Transactions on Power Electronics, vol. 27, no. 1,
“Technical-Economical Evaluations for Distributed Storage pp. 122-132, Jan. 2012.
Applications: An Italian Case Study for a Medium-Scale Public
[17] A. Etxeberria, I. Vechiu, H. Camblong, J. M. Vinassa and H. Camblong,
Facility”. Paper presented at 3rd Renewable Power Generation
“Hybrid Energy Storage Systems for renewable Energy Sources
Conference - RPG 2014, Napoli.
Integration in microgrids: A review,” 2010 Conference Proceedings
[5] Y. Wang, X. Lin and M. Pedram, “Adaptive Control for Energy Storage IPEC, Singapore, 2010, pp. 532-537.
Systems in Households With Photovoltaic Modules,” in IEEE
Transactions on Smart Grid, vol. 5, no. 2, pp. 992-1001, March 2014. [18] M. E. Glavin, P. K. W. Chan, S. Armstrong and W. G. Hurley, “A stand-
alone photovoltaic supercapacitor battery hybrid energy storage system,”
[6] E. Romero-Cadaval, G. Spagnuolo, L. G. Franquelo, C. A. Ramos-Paja, 2008 13th International Power Electronics and Motion Control
T. Suntio and W. M. Xiao, “Grid-Connected Photovoltaic Generation Conference, Poznan, 2008, pp. 1688-1695.
Plants: Components and Operation,” in IEEE Industrial Electronics
Magazine, vol. 7, no. 3, pp. 6-20, Sept. 2013. [19] C. R. Akli, X. Roboam, B. Sareni and A. Jeunesse, “Energy
management and sizing of a hybrid locomotive,” 2007 European
[7] M. J. E. Alam, K. M. Muttaqi and D. Sutanto, “Mitigation of Rooftop Conference on Power Electronics and Applications, Aalborg, 2007,
Solar PV Impacts and Evening Peak Support by Managing Available pp. 1-10.
Capacity of Distributed Energy Storage Systems,” in IEEE Transactions
on Power Systems, vol. 28, no. 4, pp. 3874-3884, Nov. 2013. [20] O. Palizban and K. Kauhaniemi, “Power sharing for distributed energy
storage systems in AC microgrid: Based on state-of-charge,” 2015 IEEE
[8] M. Sechilariu, B. Wang and F. Locment, “Building Integrated PES Asia-Pacific Power and Energy Engineering Conference
Photovoltaic System With Energy Storage and Smart Grid (APPEEC), Brisbane, QLD, 2015, pp. 1-5.
Communication,” in IEEE Transactions on Industrial Electronics, vol.
60, no. 4, pp. 1607-1618, April 2013. [21] J. Zou, C. Peng, J. Shi, X. Xin and Z. Zhang, “State-of-charge
optimising control approach of battery energy storage system for wind
[9] M. Momayyezan, D. B. W. Abeywardana, B. Hredzak and V. G. farm,” in IET Renewable Power Generation, vol. 9, no. 6, pp. 647-652,
Agelidis, “Integrated Reconfigurable Configuration for 8 2015.
Battery/Ultracapacitor Hybrid Energy Storage Systems,” in IEEE
[22] M. A. Guerrero-Martínez, E. Romero-Cadaval, F. Barrero-González, M.
Transactions on Energy Conversion, vol. 31, no. 4, pp. 1583-1590,
Dec. 2016. I. Milanés-Montero and E. González-Romera, “Overview of medium
scale energy storage systems,” 2009 Compatibility and Power
[10] H. Chen, T. N. Cong, W. Yang, C. Tan b, Y. Li, Y. Ding, “Progress in Electronics, Badajoz, 2009, pp. 93-100.
electrical energy storage system: A critical review,” Progress in Natural
[23] H. Yin, W. Zhou, M. Li, C. Ma and C. Zhao, “An Adaptive Fuzzy
Science, vol. 19, 2009, pp. 291–312.
Logic-Based Energy Management Strategy on Battery/Ultracapacitor
[11] E. Romero-Cadaval, B. Francois, M. Malinowski and Q. C. Zhong, Hybrid Electric Vehicles,” in IEEE Transactions on Transportation
“Grid-Connected Photovoltaic Plants: An Alternative Energy Source, Electrification, vol. 2, no. 3, pp. 300-311, Sept. 2016.
Replacing Conventional Sources,” in IEEE Industrial Electronics
Magazine, vol. 9, no. 1, pp. 18-32, March 2015. [24] M. A. Guerrero-Martínez, E. Romero-Cadaval, M. I. Milanés-Montero,
V. M. Miñambres-Marcos and F. Barrero-González, “Aplicación de
[12] H. Beltran, E. Bilbao, E. Belenguer, I. Etxeberria-Otadui and P. Sistemas de Almacenamiento de Energía con Supercondensadores en
Rodriguez, “Evaluation of Storage Energy Requirements for Constant Plantas de Generación Fotovoltaica,” Seminario Anual de Automática,
Production in PV Power Plants,” in IEEE Transactions on Industrial Electrónica Industrial e Instrumentación 2009. Actas de Congreso,
Electronics, vol. 60, no. 3, pp. 1225-1234, March 2013. Leganés, 1-3 July 2009.
[13] H. Zhou, T. Bhattacharya, D. Tran, T. S. T. Siew and A. M.
Khambadkone, “Composite Energy Storage System Involving Battery

976·1·5090-4639·31171$31,00 C 2017 IEEE 20171nlemalional Young Engineers Forum (YEF~ECE )

Caparica, Portugal, May 5, 2017
35