2019 1st Global Power, Energy and Communication Conference (IEEE GPECOM2019), June 12-15, 2019, Cappadocia, Turkey
A Consecutive Energy Management Approach for a
VPP Comprising Commercial Loads and Electric
Vehicle Parking Lots Integrated with Solar PV Units
and Energy Storage Systems
Arian Zahedmanesh Kashem M. Muttaqi
Student Member, IEEE Senior Member, IEEE
School of Electrical, Computer and School of Electrical, Computer and
Telecomminications Engineering, Telecomminications Engineering,
University of Wollongong University of Wollongong
Wollongong, Australia Wollongong, Australia
az617@uowmail.edu.au kashem@uow.edu.au
Abstract— Renewable energy resources (RES), such as
photovoltaic (PV) solar systems, are environmentally friendly
and can help to meet the fast growing load demand. Further,
electric vehicle (EV) is a reliable alternative for the
conventional vehicles given its technical and environmental
advantages. However, due to the intermittent nature of the
solar PV output, and also the electric and the economic issues
that can be caused by the EV mass charging, a suitable
paradigm is necessary for a better integration of EV charging
with PV systems in distribution power systems. In this paper,
a model of a virtual power plant (VPP), comprises the PV
system and the EVs’ parking lot, connected to a distribution
power grid is developed. A consecutive energy management
(CEM)-based methodology is presented to assign an organized
and economic energy management for the VPP and to address
the electric constraints for the power systems. Both the electric
and economic impacts of the VPP and the efficacy of the CEM
method are assessed through different case studies. The results
from the case studies show that the proposed CEM method can
decrease the cost of energy to the VPP and facilitate the
integration of the VPP in power systems under varying system
conditions.
Keywords— Electric vehicle (EV), energy management,
energy storage system (ESS), solar photovoltaic (PV) system,
virtual power plant (VPP)
I. INTRODUCTION
Due to the limitations and environmental drawbacks of
fossil fuels for electricity generation and transportation and
also the fast-growing demand in energy markets, it is
essential to consider other reliable and environmentally
friendly energy resources. Consequently, the use of
renewable energy sources (RESs) has been increasingly
growing, and RES appear to be the future prevalent
technologies given their continuously reducing cost and
increasing efficiency. However, given the unpredictable and
intermittent nature of RESs, certain challenges still need to
be addressed. Likewise, electric vehicles (EVs) are feasible
solutions for future transportation systems. Nonetheless, in
order to expand their applications, major technical and
economic challenges remain to be addressed. In this context,
the integration of commercial/residential buildings and EV
parking lots with RESs and/or energy storage systems
(ESS), as a virtual power plant (VPP) operated by a system
operator or a VPP controller has been endorsed by many
studies[1]-[6], because: (i) the RESs can be used to supply
energy demand required by the EV chargers and other local
loads, specifically during peak hours with high electricity
rates, and (ii) such a configuration offers more flexibility for
This work was supported by the Australian Research Council (ARC).
978-1-5386-8086-5/19/$31.00 ©2019 IEEE
242
Authorized licensed use limited to: Universidad del Valle. Downloaded on October 02,2020 at 01:48:21 UTC from IEEE Xplore. Restrictions apply.
Danny Sutanto
Senior Member, IEEE
School of Electrical, Computer and
Telecomminications Engineering,
University of Wollongong
Wollongong, Australia
soetanto@uow.edu.au
the system operator to mitigate the intermittency of RESs
and also to take advantage of lower electricity prices at
night. For example during midday when the outputs from
the PV is higher, EVs and ESS can be charged to absorb the
excessive energy produced by the PV solar system and this
stored energy can flow back to the power system when
necessary.
In [1], a cost reduction has been demonstrated by
deploying a hybrid algorithm for a photovoltaic (PV)
integrated with an EV charging station and an ESS. The
focus of the study in [1], is on the energy management of
the ESS, while the degradation cost of the ESS has been also
taken into account. The proposed energy storage
management in [1], consists of three different operations
and the system operates between the deterministic and rule-
based approaches based on real-time data. In order to
decrease the operational cost and to optimize the offer/bid
strategies for an ensemble with charging stations coupled
with ESS in the electricity market, a scheduling framework
has been presented in [2]. The degradation costs of the ESS,
the stochastic load of the charging station and the
uncertainties of the electricity prices have been also taken
into consideration for the proposed framework in [2]. In [3],
the authors have proposed a two-stage economic operation
framework for an EV parking deck with rooftop PV panels,
where the first stage includes a stochastic method to specify
the marginal price considering the indeterminacy of the
solar energy generation. The second stage is to deal with the
varying load of the EVs’ charging. According to the case
studies in [3], such a two-stage approach can offer a more
reliable operation to mitigate the uncertainties of the
distributed energy sources and EV charging load and it can
increase the total revenue for the system’s operator. Similar
approaches have been suggested to diminish the
uncertainties associated with the EV charging and RESs in
[4], [5]. In [4], two different two-stage control algorithms
have been proposed for the energy management of the office
buildings with EVs and PV panels, where the first stage of
the algorithms is for the day-ahead planning and the second
stage is for the real-time operation to reduce the operation
cost by coordinating the EV charging. The discharge of the
EVs’ batteries and a stand-alone battery system has been
also proposed to counteract the mismatch between the two-
stages in [4]. The authors in [5] have presented a four-stage
control algorithm for a charging station in a commercial
building with PV energy production, that is also deployed
with a fixed battery ESS. The control objective is to
minimize the operating costs taking into account the
 2019 1st Global Power, Energy and Communication Conference (IEEE GPECOM2019), June 12-15, 2019, Cappadocia, Turkey

uncertainties of the system. Also, since the bidirectional

power flow has been considered for the EV chargers,
another objective of the control algorithm is to balance the
supply and demand through the discharging of the EV and
the ESS. Even though the four stage algorithm results in
better resiliency against uncertainties, it increases the
complexity of the system. The application of a mixed
integer linear programming (MILP) technique for the EV
charging using PVs has been also demonstrated in [6]. The
technique employs strategies for market participation of the
PV and EVs with vehicle to grid (V2G) capability in the
electricity grid and the adoption of EVs in order to provide
ancillary services with the reserve capacity of EV’s
batteries. In the study in [6], the benefits of V2G services Fig. 1. The proposed configuration for the VPP.
have been compared with the degradation costs of the EV
batteries. Furthermore, the utilization of an integrated single A. Daily Schedule
converter for both the EVs and PV systems has been
suggested given that such a structure can provide better In the first hierarchy, the VPP controller participates in
efficiency and allows the EV charging and the PV the energy market based on the initial estimation of the
generation schedules to be optimally combined [6]. electric demand of the commercial area and the parking lot,
However, realization of such an infrastructure entails more and the forecast of the generated energy of the PV system.
advanced control systems and the need for modification of The main functions of this hierarchy are to supply the actual
the existing charging equipment and the PV inverters. energy demand of the VPP and also to minimize the energy
cost. In order to address the prediction errors regarding the
As a result, in the majority of the existing studies on the generated power of the PV system and the power demand of
integration of EV charging and RESs, the main function of the parking lot and the commercial area, the VPP controller
the control algorithm has been to decrease the operational would control the rate of charging/discharging of the ESS
cost, whilst no further scrutinies have been presented and shift the charging demand of the EVs using the receding
regarding the electric impacts of such entities or the need to horizon technique (RHT) as presented in [7]. The
meet the power constraints in the distribution networks. In scheduling of the ESS and also the parking lot will provide
addition, stringent and realistic models capable of flexibility for the VPP controller.
illustrating the varying load of the EV charging and the
produced energy of RESs are scant in the existing literature. The main objective is to minimize the energy cost for
the VPP controller, , which is described as follows:
In this paper a consecutive energy management (CEM)-
based methodology is presented which assigns an organized
and economic planning of energy consumption/production = min ( ℓ, + ℓ, − ℓ, ), ∀ℓ ∈ ℒ (1)
for a VPP, so that the energy management of the VPP is
operated by a VPP controller. The key objective of this In (1), ℓ, is the degradation cost of the ESS and
study is to devise a framework for the participation of the and are defined as (2) and (3), respectively.
ℓ, ℓ,
VPP controller in energy markets under varying system
conditions and uncertainties, while facilitating the = ∙ ∙ , ∀ℓ ∈ ℒ, ∀ ∈ (2)
ℓ, ℓ,
integration of the VPP in power systems by supporting = ∙ ∙ , ∀ℓ ∈ ℒ, ∀ ∈ (3)
ℓ, ℓ,
ancillary services. Also, both the electric and monetary
where, ℓ is the index of the VPP, is the index of the time-
effects of the VPP and also the efficacy of the proposed
CEM-based methodology are assessed by means of steps and is the final time-step of the optimization time
/
experimental studies and simulations. horizon. Also, [$/kWh] are the purchase and
selling prices of electricity and [h] is the duration of each
II. THE PROPOSED CONSECUTIVE ENERGY MANAGEMENT time-step, which is considered to be an hour. Moreover,
( )
The structure for the VPP includes a commercial area ( ℓ, ∙ ) is equal to the energy purchased/sold from
with an EV parking lot integrated with a PV generation the electricity market at time-step .
system and a battery ESS. The schematic of the VPP is
shown in Fig. 1. The proposed CEM method includes The following constraints should be fulfilled for the
hierarchies for the energy management and for providing VPP:
ancillary services. The first hierarchy of the CEM includes = + , ∀ℓ ∈ ℒ, ∀ ∈ (4)
ℓ, ℓ, ℓ,
a daily scheduling approach to provide the electric energy
for the consumers in the commercial area together with the ℓ, = ℓ, + ℓ, , ∀ℓ ∈ ℒ, ∀ ∈ (5)
charging demand of the parking lot and also to maximize where, ℓ, [kW] and ℓ, [kW] are the power demand of
the monetary profit of the VPP controller. The second the building and the parking lot, ℓ, [kW] is the forecasted
hierarchy contributes to the technical requirements of the
power system that will be achieved by providing the total electric power generated by the PV system, ℓ, is
reactive power compensation (RPC) by the VPP. the part of the PV generated power for supplying the
building and the parking lot consumption and ℓ, is the
part of the PV generated power that VPP controller can offer
to the electricity market as the surplus energy of the VPP,
for time-step .

243
Authorized licensed use limited to: Universidad del Valle. Downloaded on October 02,2020 at 01:48:21 UTC from IEEE Xplore. Restrictions apply.
 2019 1st Global Power, Energy and Communication Conference (IEEE GPECOM2019), June 12-15, 2019, Cappadocia, Turkey

The VPP controller should provide the energy demand respectively. Also, , , , , ( ) and
of the building and the parking lot as described in (6):
( ) are the maximum allowable charging/discharging

ℓ, + ℓ, + ℓ, = ℓ, + ℓ, + ℓ, power of the ESS, the minimum/maximum allowable SoC

(6) of the ESS and the efficiency of charging/discharging of the
∀ℓ ∈ ℒ, ∀ ∈
ESS, at time-step for VPP ℓ, respectively.
where, ℓ, [kW] and ℓ, [kW] are the charging and
discharging power of the ESS at time-step for VPP ℓ, B. Reactive Power Compensation (RPC)
respectively. Following the adoption of the first hierarchy, given the
coordination of the PV generated power and the
The following constraints should be fulfilled for EV
charge/discharge of the ESS and also by scheduling the load
charging in the parking lot:
demand, the energy cost to the VPP controller can be
minimized. However, in order to address the electric
∙ ∙ =ℰ , ∀ ∈ (7) constraints of the power system, the VPP can provide the
,
RPC for the power grid in the second hierarchy. This service
is feasible for the VPP controller because: (i) it doesn’t
intervene with the normal operation of the VPP, where the
= ( , )+ ℓ, , ∀ℓ ∈ ℒ, ∀ ∈ (8) VPP controller minimizes the energy cost by supplying the
ℓ,
loads and selling energy to the grid and (ii) the RPC is less
≥
,
, ∀ℓ ∈ ℒ, ∀ ∈ (9) complex in comparison to the active power control since the
ℓ, ℓ,
( )
RPC can be achieved for each time-step, independently.
where, , , [kW], ℰ [kWh] and [h]
are the charger efficiency, the charging power of EV at Considering the fact that the converters of the PV and
time-step , the total energy demand to fully charge EV , ESS are capable of reactive power control, after solving the
first hierarchy, the RPC capacity will be determined for the
and the plugin/plug-out times of EV , respectively.
PV and ESS. Afterward, every VPP controller will inform
Moreover, ℓ, [kW] is the total power demand of the these values to the distribution system operator (DSO).
,
arriving EVs at time-step , ( ℓ, = 0), whilst ℓ, Subsequently, the required amount of the RPC will be
[kW] is the estimated total power demand of the EVs for determined based on the objective function in (18) together
time-step , that is specified according to the statistical data. with the constraints in (19), (20) and (21), and the nonlinear
is the number of the EVs in the parking lot at time-step power flow constraints in (22) and (23) in order to minimize
. the power losses in the power system:
An important factor to be taken into account is the = min ∙ , ∀ ∈
, (18)
degradation cost of the ESS. Different models have been ∈ℬ
developed to represent the effect of several
charge/discharge cycles on the lifetime of the battery ESS. ≤ , ≤ , ∀ ∈ ℐ, ∀ ∈ (19)
ℐ
The degradation cost of the ESS can be modelled by the
, ≤ , ∀ ∈ (20)
following cost function [8]:

1 , ≤ , ∀ ∈ ℬ, ∀ ∈ (21)
ℓ,
ℓ, = ∙ ℓ ∙| |∙ , ∀ℓ ∈ ℒ (10) ( ) ( )
100 ℓ , = ( , , cos + , − ,
∈{ , , }
where, is a coefficient to represent the lifetime of the ( ) ( )
− , , cos + , − , ),∀ ∈ (22)
battery as the number of cycles, ℓ is the cost of the battery
( ) ( )
ESS [$/kWh] and ℓ [kWh] is the battery capacity of , = ( , , sin + , − ,
ESS ℓ. ∈{ , , }
( ) ( )
The following constraints are essential for the ESS: − , , sin + , − , ),∀ ∈ (23)
where, , = , , ,
( ), , and are
ℓ, = ℓ, + ∙[ ∙ ℓ, −
ℓ the voltage at bus , the minimum/maximum allowed
(1/ ) ∙ ℓ, ], ∀ℓ ∈ ℒ, ∀ ∈ (11)
bounds for the rms voltages, the apparent power of bus at
ℓ, = ℓ, + ℓ, , ∀ℓ ∈ ℒ, ∀ ∈ (12) time-step and the rating power of the transformer, also,
0 ≤ ℓ, ≤ ∙ , ∀ℓ ∈ ℒ, ∀ ∈ (13)
, and are the current and the rating current of
,
0 ≤ ℓ, ≤ (1 − ) ∙ , , ∀ℓ ∈ ℒ, ∀ ∈ (14)
branch between bus and bus , respectively. Further, p
≥ ℓ, , ∀ℓ ∈ ℒ, ∀ ∈ (15)
≤ is the index of the electric phases and , = ,
ℓ, , ∀ℓ ∈ ℒ, ∀ ∈ (16)
ℓ, = ℓ, , ∀ℓ ∈ ℒ, ∀ ∈ (17) , and , , are the resistance, the impedance, the active
where, = 0, 1, is an auxiliary variable, that enforces the and the reactive power flow of branch at time-step ,
fact that the ESS cannot be charged and discharged at the respectively.
same time, ℓ [%] is the state of charge (SoC) of the
III. CASE STUDIES
ESS, ℓ, and ℓ, are the part of the ESS
discharging power to supply the building together with the A. Modelling the Distribution Power System
parking lot and the part of the ESS discharging power that In order to analyse the electric impacts of the proposed
VPP controller can offer in the electricity market, VPP on the distribution power systems, a power network is

244
Authorized licensed use limited to: Universidad del Valle. Downloaded on October 02,2020 at 01:48:21 UTC from IEEE Xplore. Restrictions apply.
 2019 1st Global Power, Energy and Communication Conference (IEEE GPECOM2019), June 12-15, 2019, Cappadocia, Turkey

modelled based on a standard IEEE medium voltage (MV)
test feeder, so that the balanced distribution of the loads at
all the phases is taken into account [9]. The single-line
diagram of the modelled network is illustrated in Fig. 2.
extracted for the EVs [15], [16]. To this aim, experimental
studies, that include the deriving of the harmonic and inter-
harmonic charging currents, are carried out for an EV. The
specifications of the EV adopted for the field studies are
shown in Table II. In the case studies, it is considered that 9
electric vehicle supply equipment (EVSE) of the
ChargePoint CT4000 type, located in the parking lot,
provide the simultaneous charging of 18 EVs with level 2
AC charging. Fig. 3 illustrates the charging of the EVs with
such a commercial infrastructure.
TABLE II
THE SPECIFICATIONS OF THE EV IN THE CASE STUDIES

Model BMW i3
Rated Battery Capacity 94 Ah
Rated power of EV charger 7.4 kW

Fig. 2. The single-line diagram of the modelled network.

The modelled network includes 12 MV load buses and

there are 16 low voltage (LV) nodes at all phases of each
MV bus, hence each MV bus totally supplies 48 commercial
and residential consumers. Besides these loads, the VPP is
also considered to be located at Bus 12. This location is
chosen based on the voltage sensitivity index introduced in
[10], where the change of active and reactive power at this
bus has the highest impact on the voltage magnitude at the
other load buses.
In addition, the load modelling technique described in
[11], is implemented to duplicate the load profile of the Fig. 3. The commercial charging equipment adopted for the case studies.
commercial area in the VPP and the other buses in the power
network. Further, the produced energy of the PV system at In order to better illustrate the load behaviour of an EV,
different times is modelled based on the method discussed the distorted currents drawn by the EV charger, in the
in [12], while the maximum solar irradiance level is derived reduced level 2 mode, are shown in Fig. 4.
based on the statistical information for the Sydney region in
Australia. The rating of the distribution transformer, in
addition to the rating of the PV system, the capacity and the
maximum charging/discharging rate of the battery ESS in
the VPP, are given in Table I.
TABLE I
THE RATINGS OF THE DISTRIBUTION TRANSFORMER AND THE VPP

Distribution transformer VPP

Rated capacity 5000 kVA PV output 114 kW
Voltage 115/4.16 kV ESS capacity 84.50 kWh
R 1% ESS maximum
52.80 kW
X 8% charge/discharge

The electric model of linear and nonlinear loads at the

fundamental frequency are developed as constant power
loads. At harmonic frequencies, linear loads are modelled as
impedance loads and harmonic-producing loads are
modelled as constant current sources [13]. Also, the models
for the PV and the ESS systems are developed in accordance
to the studies adopted in [14].
B. Modelling the Electric Load of the Parking Lot
In this paper, a model that can replicate the AC charging Fig. 4. The harmonic currents drawn by the EV charger in the reduced
of EVs in a parking lot, is developed using the load analysis level 2 mode.
of the EV commercial charging equipment and also the
available statistical data regarding the EVs’ In Fig. 4, the ninth, fifth and third harmonic orders have
arrival/departure patterns in a real parking lot. Given that the the highest amplitudes, whilst the total harmonic current
charging current of an EV is controlled during the constant distortion factor (THDI) is 2.94%. Likewise, the harmonic
current (CC) mode of the AC charging, which occurs during currents for the other charging power levels are extracted
the major part of the AC charging period, the same electric through the experiments. For the energy demand of the
model that is used for the harmonic-producing loads, can be parking lot, the arrival of the EVs can be modelled using a

245
Authorized licensed use limited to: Universidad del Valle. Downloaded on October 02,2020 at 01:48:21 UTC from IEEE Xplore. Restrictions apply.
 2019 1st Global Power, Energy and Communication Conference (IEEE GPECOM2019), June 12-15, 2019, Cappadocia, Turkey

Poisson distribution, in addition, the required charging
demand of each EV can be modelled according to the
historic data for the driven distance of the EVs [11], [17].
Fig. 5 shows an example of the distorted current at the point
of common coupling (PCC) for the parking lot that is
achieved by the aforementioned approach for modelling the
EVs’ charging load.
the real energy markets in Australia. In order to better
compare the net cost to the VPP controller, the degradation
cost of the battery ESS is also considered for the second
scenario. Moreover, the levelized cost of energy (LCOE)
(the cost of producing the PV energy per kWh) is considered
for the 24 hour horizon and it is presented in Table III [19].
TABLE III
THE COST OF ELECTRICITY FOR THE CASE STUDIES

Without the CEM method With the CEM method

Total cost for
Total cost for $241.18
$1220.80 the VPP
the VPP
LCOE $102.57

Table III shows that even considering the LCOE of the

PV system and the degradation cost of the battery ESS, the
net cost to the VPP controller is $343.75 that is much less
for the second scenario. This validates the effectiveness of
the CEM method for minimizing the cost to the VPP
controller.
Fig. 6 shows the minimum RMS values of the bus
Fig. 5. The load current at the PCC of the parking lot. voltages and Fig. 7 shows the total harmonic distortion
factor for the voltage (THDV) at Bus 12 for both case
IV. RESULTS AND DISCUSSIONS studies.
In order to study the impacts of the proposed scheme for
the VPP, and also to verify the effectiveness of the CEM
method, power flow studies are carried out for the test power
network for the following case studies: (i) the case that there
is no PV system neither ESS in the VPP and all of the loads
and EVs are supplied without adopting any energy
management methods, and (ii) the case that the CEM based
method is adopted for planning of energy
consumption/production of the VPP. In the scenario with
the CEM method, it is also assumed that the VPP controller
will receive the necessary information regarding the
charging characteristics of the EVs from the EV owners. Fig. 6. The RMS values of the bus voltages for the case studies.
Moreover, the VPP controller has the forecast of the PV
output power, the charging demand of the EVs and the
energy demand of the commercial area. For this objective,
the forecasted data are obtained in such a way that the mean
absolute percentage error (MAPE) and the root mean
squared error (RMSE) of these data are compatible with the
available forecasting techniques [18].
The optimization time horizon is five hours for the RHT,
hence the CEM method assigns the optimized energy
scheduling for the next five hours, however only the energy
schedule for the operational time-step ( =1) are acceptable.
The required SoC of all the EVs is considered to be 90%,
also the minimum and maximum allowable and the initial Fig. 7. The value of THDV for Bus 12.
SoC levels of the ESS are assigned to be 10%, 95% and
50%, respectively. In order to guarantee that the rms values As shown in Fig. 6, for the scenario where the CEM
of the voltages would remain within the accepted margins method is utilized for the VPP, the rms values of voltages at
according to the relevant standards, Vmin and Vmax are all buses are in compliance with the assigned limits,
assigned to be 0.95 pu and 1.05 pu, for the second hierarchy therefore the adoption of the CEM method results in the
of the CEM method, respectively. The electric models of the mitigation of the under-voltages in the distribution power
loads and the parking lot for each time-step are based on the network. In the case of the voltage distortions, the
10 minute measurements. In order to solve the problem in application of the CEM method for the VPP, would not
(1), the solver for the linear programming in MATLAB affect the total distortion at Bus 12. This shows that the
software is adopted. The optimization problem in (18) for nonlinear loads, e.g. TV, laptop, etc. are the predominant
the second hierarchy is nonlinear and hence, the genetic sources of the distorted current at Bus 12.
algorithm technique is adopted. Whilst, other heuristic or In Fig. 8, the profile for the active power flow at the VPP
the gradient-based techniques could be also implemented. bus is shown for both case studies. Also, the SoC of the ESS
The costs of the electricity are presented for both case and the real and forecasted output power of the PV system
studies in Table III, whilst the electricity prices are based on are illustrated in Fig. 9.

246
Authorized licensed use limited to: Universidad del Valle. Downloaded on October 02,2020 at 01:48:21 UTC from IEEE Xplore. Restrictions apply.
 2019 1st Global Power, Energy and Communication Conference (IEEE GPECOM2019), June 12-15, 2019, Cappadocia, Turkey
Fig. 8. The profile for the active power flow at the VPP bus.
Fig. 9. The SoC of the ESS and the real and forecasted output power of
the PV system.
Fig. 8 shows that the profile of the active power flow is
managed for the case study where the CEM based is applied
for scheduling the energy management. Furthermore, for
this case study, given the integration of the PV system, the
VPP controller can offer the excessive energy in the
electricity market, during the hours with high solar
irradiation, see Fig. 8. In Fig. 9, the predetermined limits for
the SoC are not violated during the 24 hour horizon, also the
SoC values at the end of the 24 horizon is the same as the
initial SoC of the ESS battery. In addition, according to the
results of the case studies, all of the EVs in the parking lot
are fully charged before the departure time as specified by
the EV owners. Also, all of the constraints for the CEM
beside the electric constraint for the distribution power
network are fulfilled, and the electricity cost is decreased
significantly for the VPP controller for the scenario that the
CEM method is applied.
V. CONCLUSION
This paper has investigated the electrical and economic
feasibility of a virtual power plant (VPP) comprising
commercial loads, EV parking lots, solar PV units and
energy storage systems. A consecutive energy management
(CEM)-based methodology has been presented that assigns
the economic planning of energy consumption/production
for the VPP and also addresses the electric requirements of
the power system, independently. An acquisitive model has
been developed for the VPP and the EV charging in the
parking lot and both electric and monetary indices are
derived and analysed for the test system. According to the
results, the CEM can effectively satisfy the determined
objectives under the uncertainties. Moreover, by adoption of
the proposed CEM method for the VPP in the distribution
power systems, the energy cost to the VPP controller has
been minimized, while all the electric constraints are met
and the voltage quality is improved.
247
Authorized licensed use limited to: Universidad del Valle. Downloaded on October 02,2020 at 01:48:21 UTC from IEEE Xplore. Restrictions apply.
REFERENCES
[1] K. Chaudhari, A. Ukil, K.N. Kumar, U. Manandhar, and S.K.
Kollimalla, “Hybrid optimization for economic deployment of ESS in PV-
integrated EV charging stations”, IEEE Trans. on Industrial Informatics,
vol. 14, no. 1, Jan. 2018, pp. 106-115.
[2] M. R. Sarker, H. Pandžić, K. Sun, and M. A. Ortega-Vazquez,
“Optimal operation of aggregated electric vehicle charging stations
coupled with energy storage,” IET Generation, Transmission &
Distribution, 2017, pp. 1127-1136.
[3] Y. Guo, J. Xiong, S. Xu, and W. Su, “Two-stage economic operation
of microgrid-like electric vehicle parking deck,” IEEE Trans. on Smart
Grid, vol. 7, no. 3, pp. 1703-1712, 2016.
[4] D. Wu, H. Zeng, C. Lu, and B. Boulet, “Two-stage energy
management for office buildings with workplace EV charging and
renewable energy,” IEEE Trans. on Transportation Electrification, vol. 3,
no. 1, Mar. 2017, pp. 225-237.
[5] Q. Yan, B. Zhang, and M. Kezunovic, “Optimized operational cost
reduction for an EV charging station integrated with battery energy
storage and PV generation”, IEEE Trans. on Smart Grid, vol. 10, march
2019.
[6] G.R.C. Mouli, M. Kefayati, R. Baldick, and P. Bauer, “Integrated PV
charging of EV fleet based on energy prices, V2G and Offer of Reserves”,
IEEE Trans. on Smart Grid, 2017.
[7] M. Rahmani-Andebili and M. Fotuhi-Firuzabad, “An adaptive
approach for PEVs charging management and reconfiguration of electrical
distribution system penetrated by renewables,” IEEE Trans. on Industrial
Informatics, vol. 14, no. 5, May 2018.
[8] M.A. Ortega-Vazquez, “Optimal scheduling of electric vehicle
charging and vehicle-to-grid services at household level including battery
degradation and price uncertainty,” IET Gener. Transm. Distrib., 2014,
vol. 8, pp. 1007–1016.
[9] IEEE 13 Node Test Feeder. Test feeders. [Online]. Available:
http://sites.ieee.org/pes-testfeeders/resources/.
[10] B. Bakhshideh Zad, H. Hasanvand, J. Lobry, and F. Vallée, “Optimal
reactive power control of DGs for voltage regulation of MV distribution
systems using sensitivity analysis method and PSO algorithm,”
International Journal of Electrical Power & Energy Systems (ijepes),
2015.
[11] A. Zahedmanesh, D. Sutanto, and K. M. Muttaqi, “Analyzing the
Impacts of Charging Plug-in Electric Vehicles in Low Voltage
Distribution Networks: A Case Study of Utilization of Droop Charging
Control System Based on the SAE J1772 Standard,” 2017 Australasian
Universities Power Engineering Conference (AUPEC), Melbourne,
Australia, 2017, pp. 1-6.
[12] R. Torquato, Q. Shi, W. Xu, and W. Freitas, “A Monte Carlo
simulation platform for studying low voltage residential networks,” IEEE
Trans. on Smart Grid, vol. 5, No. 6, Nov. 2014.
[13] N. Yang, and M. Le, “Three-phase harmonic power flow by direct Z
BUS method for unbalanced radial distribution systems with passive
power filters,” IET Generation, Transmission & Distribution, vol. 10,
2016.
[14] J. H. Enslin and P. J. Heskes, “Harmonic interaction between a large
number of distributed power inverters and the distribution network,” IEEE
Trans. Power Electron., vol. 19, no. 6, pp. 1586–1593, Nov. 2004.
[15] A. Zahedmanesh, K. M. Muttaqi, and D. Sutanto, “Direct control of
plug-in electric vehicle charging load using an in-house developed
intermediate control unit,” 2018 IEEE Industry Applications Society
Annual Meeting, 2018.
[16] A. Zahedmanesh, K. M. Muttaqi, and D. Sutanto, “Direct control of
plug-in electric vehicle charging load using an in-house developed
intermediate control unit,” IEEE Trans. on Industry Applications, vol. 55,
pp. 2208-2218, 2019.
[17] S. Wang, S. Bi, Y.A. Zhang, J. Huang, “Electrical Vehicle Charging
Station Profit Maximization: Admission, Pricing, and Online
Scheduling,” IEEE Trans. on Sustainable Energy, vol. 9, Oct. 2018.
[18] A. O'Connell, D. Flynn, and A. Keane, “Rolling Multi-Period
Optimization to Control Electric Vehicle Charging in Distribution
Networks,” IEEE Trans. on Power Systems, vol. 29, pp. 340-348, 2014.
[19] Solar Report Jan. 2018 Australian Energy Council. [Online].
Available: https://www.energycouncil.com.au/media/11188/australian-
energy-council-solar-report_-january-2018.pdf