2018 IEEE 3rd International Conference on Intelligent Energy and Power Systems (IEPS)
Energy Efficiency of Microgrid Implementation with
Solar Photovoltaic Power Plants
Dmitry Tugay, Serhii Kotelevets, Serhii Korneliuk
Dept. of Alternative Power Engineering and Electrical
Engineering
O.M. Beketov NUUE
Kharkiv, Ukraine
tugaydmytro@gmail.com
Abstract— The work with the use of simulation in the software
Matlab / Simulink / SymPowerSystems environment considers
construction of a local Smart Grid energy supply system with
distributed solar power plants. The obtained model allows us to
investigate the work of the intelligent network in any quasi-steady
and transitional modes, including emergency ones. A distinctive
feature of the proposed model is the localization of places for the
installation of power active filter-compensating devices, the use of
which allows providing the necessary quality of electric energy and
achieving the minimum energy losses in the elements of the energy
supply system. According to the results of the simulation, the
comparison of the energy efficiency of the traditional energy supply
system and Smart Grid has been made.
Keywords—energy supply system, microgrid, minimum energy
losses, solar power plant, efficiency, short cut power, useful power
I. INTRODUCTION
The change of the concept of developing modern power
engineering is conditioned by growing interest in renewable
energy sources [1–9]. The most rapid pace of the development
among low-power distributed renewable energy sources is
presented by private solar power plants, which operate both
autonomously, and can be integrated into the industrial
network [7, 9–14].
Structural changes in the electricity market, where the
consumer acquires additional functionalities and partial energy
independence, contributed to the emergence of a new concept
of energy development – Smart Grid. The most significant
feature of Smart Grid is the presence of a bi-directional energy
flow in the elements of the energy supply system (ESS) [3, 10].
Operation of the Smart Grid ESS is conditioned by the
operation of the industrial network, renewable energy sources
and variable load profiles. In the intelligent ESS with small
solar power plants, the combination of such modes causes
some difficulties in implementing an information management
system that would ensure not only high reliability of power
supply but also increase its energy efficiency [11–15].
Therefore, at the pre-design stage, close attention should be
paid to the means of computer simulation to study the work of
the smart ESS in operating and emergency modes.
The aim of the article is to study of the operating modes
and energy efficiency assessment of local microgrid on the
978-1-5386-9545-6/18/$31.00 ©2018 IEEE 275
George Zhemerov, life member, IEEE
Dept. of Industrial and Biomedical Electronics
National Technical University “KhPI”
Kharkiv, Ukraine
zhemerov@gmail.com
basis of distributed solar power plants, power consuming
storage and power filtering devices using simulation tools.
II. LOCAL MICROGRID BLOCK DIAGRAM
As an object for the implementation of microgrid we shall
consider a local energy supply system, which is a collection of
small households, whose electricity is supplied from the
transformer substation along the trunk circuit with a four-wire
0.4 kV cable line (Fig. 1). The useful installed power of the
load is 8, 5, 4.2, 3 and 4 kW respectively. The reactive power
of the load is 7, 4, 2, 2 and 4 kVAr respectively. The third load
contains nonlinear elements. The distances between the loads
are 100 m, and the cable cross-section of the cable lines
starting from the transformer are 10, 6, 4, 2.5 mm2 respectively.
We shall suppose that some distributed households have a solar
power plant nominal power of 10, 5 and 3 kW installed that
can operate in both the network and autonomous modes.
Fig. 1. Microgrid schematics of the local energy supply system
The implementation of microgrid is carried out by
installing specialized power equipment, the work of which is
controlled by the information management system, according
to the state of the ESS at the current time.
Parallel power active filters (PAF) (blocks PAF 1 and PAF 2)
are connected in common connection points of loads. The tasks
solved by installing filters related to ensuring the required quality
of electricity are compensation of reactive power, elimination of
higher harmonic components of network currents and asymmetry
of currents in terms of uneven loading of phases.
At an even distance from the distributed solar power
plants, a system energy-saving storage device (ES) is installed,
 2018 IEEE 3rd International Conference on Intelligent Energy and Power Systems (IEPS)

which is designed to solve two key tasks – performing the

function of the backup power supply in the autonomous
operating modes of the system and the alignment of the load
profile, that is, the elimination of daily peaks and failures in
power consumption. The implementation of these functions,
combined with the installation of power active filters will
minimize losses in the line and elements of ESS [19, 20]. The
results of the performed calculations have shown that,
depending on the parameters of the ESS, the theoretically
possible reduction in the energy losses is in the range of 2 to
15% [19]. If the level of loss reduction is greater than the loss
in the installed equipment, the implementation of these
measures is energy justifiable.
The formed network structure allows implementing
separately the energy supply system of direct current (Fig. 1).
The system energy storage is charged from distributed solar Fig. 2. Daily graphs of solar insolation (E) and load profile (*)
power plants, and in case of full charge, network inverters are
switched on, and renewable sources give energy to the AC The total relative losses power in the three-phase four-weir
network. In offline mode, when the automatic switch in the ESS to the PAF connection is determined by the ratio [19]
beginning of the AC supply lines is open, with the help of a
stand-alone inverter, the sinusoidal voltage is formed with a
frequency of 50 Hz and the energy supply of the loads (
2
ΔPΣ* = ΔPmin* 1 + QRMS 2
* + PpulsRMS * + ) (2)
+ ΔPn* ⋅ (1 + ΔPmin* )2 .
connected to the microgrid is carried out from the system ES.
Independent DC power can be connected by the appropriate
low-power load or electric vehicles, both for recharging on-
board batteries and as additional backup sources. where QRMS* = QRMS/Pusf – relative RMS of the reactive power,
measured on the transformer busbar; PpulsRMS* =
The diversity of the modes of operation of microgrid is PpulsRMS* = PpulsRMS/Pusf – is the relative mean square value of
provided by an additional information level (Fig. 1), which the variable component of the instantaneous active power;
collects information about the status of each element of the Pn* = Pn/Pusf – the relative losses power from the current
system and, in accordance with the priority algorithms, control flow through zero wire. For three-phase three-conductor or
impacts are formed, which are worked out by power symmetric three-phase four-conductor ESS Pn* = 0.
semiconductor converters.
The relative value of the minimum possible losses power
III. ENERGY-SAVING EFFECTS OF MICROGRID IMPLEMENTING can be calculated after the circuit is transformed in Fig. 1 to the
equivalent one, representing a serial connection of an energy
The efficiency of solar power plants depends to a large
source, a Rs resistor and a resistor RL. The Rs resistor imitates
extent on weather conditions and seasons. Therefore, during
the equivalent resistance of the connecting cable line which the
the implementation of microgrid the solar irradiance level in a
load is connected to. The RL resistor imitates an equivalent load
particular region, which is often presented as a daily graph for
resistance, which determines the average useful power of Pusf.
each calendar month, should be taken into account. Also,
For trunk circuit ESS, the relative value of the minimum
during the calendar year, the daily load profile changes. The
possible losses power is determined by two methods – by the
overlay of these two charts allows predicting the share of
short circuit test, carried out on the clamps of the most distant
electricity that will be consumed by microgrid objects from
from the power supply of the load, or by measuring the
the industrial network, taking into account the generation of
equivalent resistance of the ESS at the beginning of the trunk
alternative sources, and hence the reduction of the current
line R(*), which is a function of the variable load profile.
density in the supply cable and the losses power from its flow.
Figure 2 shows a daily graph of solar insolation (E) for a July According to the second method
day, typical for eastern Ukraine [16], and a daily load profile
in the fractions of the maximum installed power P* = P/PH, n P n P
RS
which is typical for household loading [17, 18]. ΔPmin* = = RS  * = RΣ (P* ) * − 1 , (3)
RL i =1 Ri i =1 Ri
An additional energy-saving effect is achieved by using
active power filters that compensate reactive components of
currents and pulsations of instantaneous active power. The total where Ri is the impedance of the i-th load, n – the number of
relative (in fractions of the average useful power of Pusf) losses loads connected to the main line.
power after using the PAF will consist of a relative minimum The equivalent resistance R(*) can be found by
possible losses power Pmin* = Pmin/Pusf and relative losses measuring the instantaneous values of phase currents and
power in PAF PPAF* = PPAF/Pusf voltage on the transformer busbar

ΔPΣ* PAF = ΔPmin* + ΔPPAF * . (1)

276
 2018 IEEE 3rd International Conference on Intelligent Energy and Power Systems (IEPS)

−1 The model of the solar photomodule in Matlab environment

T 
( )
RΣ P* = 3U S2T
 (ua ia + ub ib + uc ic )dt 
0  mathematical model of the solar cell that takes into account the
where US – is the RMS value of phase voltage,  – the period
of repetition, which can equal the period of measurements
repetition.
Taking into account the fact, that the maximum possible
efficiency of the three-phase ESS is uniquely determined by
the ratio of the resistive short-circuit power to the average
useful load power kSC = PSC/Pusf [19, 20] the following can be
written
  −1  
2
 −1  
n temperature coefficient of short-circuit current –
k SC = 0.25 − RΣ (P* ) ⋅ P*  (Ri ) − 0.5 
. (5) Ksc = +0,66% / ° ; temperature coefficient of maximum power

i =1

 
– Pm = –0.44% / ° C; efficiency – 18.31%. The layout of the

 
Figure 3 shows the dependence of the kSC coefficient on the
load factor P*, constructed according to (4) for the ESS in
Fig. 1 for the above parameters. The figure shows that under
the considered load profile kSC varies widely ranging from 53
to 12.
Fig. 3. Dependence of the kSC coefficient on the load factor P*
IV. MATLAB-MODEL OF MICROGRID WITH SOLAR POWER PLANTS
The study of the control algorithms and operating modes
of microgrid at the design stage is carried out conveniently
using computer simulation tools, for example, in the Matlab /
Simulink / SymPowerSystems environment. Figure 4 shows
the structure of the microgrid computer Matlab-model, which
corresponds to the ESS scheme for Fig. 1.
The model consists of three Solar Station units simulating
distributed solar power stations with a rated power of 10, 5
and 3 kW, an Energy Storage unit simulating a system drive,
five Load 1-Load 5 units simulating linear active-reactive and
non-linear loads, and two PAF 1, PAF 2 blocks imitating
parallel power active filters. Consider the structure of the main
blocks of the model.
, (4) is created using standard SolarCell blocks of the
SimElectronics library, which reproduce a detailed
features of its electrical and thermal characteristics (Fig. 5) [3].
In the Matlab-model a monocrystalline solar photomodule
CHN200-72M is implemented with the following
characteristics: maximum power – Pm = 200 W; number of
elements per module – N = 72; open circuit voltage Uoc = 45 V;
short circuit current – Isc = 5.8 A; voltage at the point of
maximum power tracking – Ump = 38.8 V; current at the point
of maximum power tracking – Imp = 5.15 A; maximum voltage
of the system of direct current – UDCmax = 1000 V; temperature
under normal conditions – NOCT = 45 °  ± 2 ° ; temperature
−1 coefficient of open circuit voltage – Koc = –0.34% / ° ;
solar power plant is carried out by pair wise inclusion of
photomodules into parallel cells, the number of which is
determined by the nominal power of the station.
V. ENERGY EFFICIENCY CALCULATION OF MICROGRID
IMPLEMENTATION
Figure 6 shows a comparison of the traditional ESS with a
unidirectional flow of energy from the microgrid, after its
model implementation at P* = 1. In the power supply of the
considered loads from the industrial network in the curves of
instantaneous network currents, there are higher harmonics,
and the basic harmonics of the currents are lagging behind the
corresponding instantaneous voltage, which indicates the
presence in the system of calculated reactive power.
After microgrid implementation, from the industrial
network point, the smart energy supply system is considered
as a low-power load that has a purely resistive character.
Figure 7 shows the efficiency daily change of the ESS,
calculated using the Matlab-model, for three cases: the
traditional ESS with unidirectional energy flow (); microgrid
ESS (MG); for ESS with disconnected solar power plants and
ideal PAFs (PPAF = 0) (max). Fig. 7 and fig. 2 shows that the
implementation of microgrid can significantly increase the
efficiency of the ESS, especially for the intervals of the day
when high levels of solar insolation (6 – 8 hours) are observed
during high load factors. According to the accepted parameters
of ESS, the efficiency increases by 6%. During the night-time
failure of the load profile (1–5 hours), the use of PAF leads to a
decrease in the efficiency of the ESS, but this affects the
integral daily efficiency insignificantly. An increase in the
efficiency of microgrid at this time interval is achieved by
using the power of the system drive or by disconnecting the
PAF.
277
 2018 IEEE 3rd International Conference on Intelligent Energy and Power Systems (IEPS)

Fig. 4. Microgrid Matlab-model with solar power plants

Fig. 5. Structural diagram of the Matlab model of the solar photomodule

Fig. 6. Instantaneous values of voltages and currents measured on TP busbar: a – traditional ESS; b – microgrid

278
 2018 IEEE 3rd International Conference on Intelligent Energy and Power Systems (IEPS)
Fig. 7. Dependence of the kSC coefficient on the load factor P*
It should be noted that due to the use of PAF the reduced
total harmonic distortional coefficient (THD) decreases from
10% to less than 3%.
VI. CONCLUSIONS
At the design stage of the microgrid ESS it is necessary to
compare energy efficiency indicators before and after its
implementation. The specification of the energy efficiency
indicators estimation is based on comparison of daily profiles
of loading and of solar insolation in the course of the calendar
year.
The implementation of microgrid with solar power plants
allows to increase the efficiency of the ESS. The reserve for
increasing the efficiency through the implementation of
microgrid has two components, the first one is related to the
normalization of the power consumption mode, and the second
one to the optimization of the structure of the network, when
the distances between energy sources and consumers are
reduced, and the density of the network energy flow and trunk
line decreases. Moreover, the second component makes a more
significant contribution to increasing the efficiency of the
energy supply system.
REFERENCES
[1] A. Huang, G. Heydt, S. Dale, J. Zheng, M. Crow, “Energy internet –
future renewable electric energy delivery and management (FREEDM)
systems,” IEEE Power Electronics Society News letter, Fourth Quarter,
pp. 8-9, 2008.
[2] IEEE Standart for Interconnecting Distributed Resources With Electric
Power Systems, IEEE Std. 1547.2-2008, 2008.
[3] B.S. Stogniy, O.V. Kyrylenko, S.P. Denysyuk, “Intelligent electricity
networks of energy systems and their technological support”, Technical
Electrodynamics, no. 6, pp. 44-50, 2010.
[4] H. Xia, Q. Li, R. Xu, T. Chen, J. Wang, M. A. S. Hassan, M. Chen,
“Distributed Control Method for Economic Dispatch in Islanded
Microgrids with Renewable Energy Sources”, IEEE Access, vol. 6, pp.
21802-21811, 2018.
279
[5] X. Liang, ‘‘Emerging power quality challenges due to integration of
renewable energy sources’’, IEEE Trans. Industry Applications, vol. 53,
no. 2, pp. 855-866, 2017.
[6] A. U. Rehman, S. Zeb, H. U. Khan, S. S. U. Shah, A. Ullah, “Design and
operation of microgrid with renewable energy sources and energy
storage system: A case study”, In Engineering Technologies and Social
Sciences (ICETSS), 2017 IEEE 3rd International Conference, pp. 1-6,
2017.
[7] R. Kamphuis, S. Galsworthy, M. Stifter, T. Esterl, S. Kaser, S.
Widergren, M. Renting, “Integrating demand flexibility with distributed
generation–renewable energy sources (DG-RES) at the residential
household and commercial customer level in electricity grids”, CIRED-
Open Access Proceedings Journal, vol. 2017, iss. 1, pp. 1827-1830,
2017.
[8] B. Jie, T. Tsuji, K. Uchida, “Analysis and modelling regarding
frequency regulation of power systems and power supply–demand-
control based on penetration of renewable energy sources”, The Journal
of Engineering, vol. 2017, iss. 13, pp. 1824-1828, 2017.
[9] A. Bagheri, V. Vahidinasab, K. Mehran, “A novel multiobjective
generation and transmission investment framework for implementing
100% renewable energy sources”, IET Generation, Transmission &
Distribution, vol. 12, iss. 2, pp. 455-465, 2018.
[10] J. Praß, J. Weber, S. Staub, J. Bürner, R. Böhm, T. Braun, J. Franke
“Smart energy and grid: novel approaches for the efficient generation,
storage, and usage of energy in the smart home and the smart grid
linkup”, Smart Cities: Foundations, Principles, and Applications, pp.
575-604, 2017.
[11] IEEE Std 929-2000, ‘‘IEEE Recommended Practice for Utility Interface
of Photovoltaic (PV) Systems’’, 2000.
[12] R. K. Varma, M. Salama, “Large-Scale Photovoltaic Solar Power
Integration in Transmission and Distribution Networks”, 2011 IEEE
Power and Energy Society General Meeting, pp. 1-4, 2011.
[13] A. Tuohy, J. Zack, S. E. Haupt, J. Sharp, M. Ahlstrom, S. Dise, E.
Grimit, C. Mohrlen, M. Lange, M. G. Casado, J. Black, M. Marquis, and
C. Collier, “Solar forecasting: Methods, challenges, and performance,”
IEEE Power and Energy Magazine, vol. 13, no. 6, pp. 50–59, 2015.
[14] R. K. Varma, and M. Salama, ‘‘Large-Scale Photovoltaic Solar Power
Integration in Transmission and Distribution Networks’’, 2011 IEEE
Power and Energy Society General Meeting, pp. 1-4, 2011.
[15] A. Tripathi, K. B. Sahay, “Modeling & simulation of proposed grid
connected 10 MW solar PV array power plant at Lucknow”,. In Power
Electronics, Intelligent Control and Energy Systems (ICPEICES), IEEE
International Conference on, pp. 1-5, 2016.
[16] “Prediction Of Worldwide Energy Resource” Available at
https://power.larc.nasa.gov accessed: 10 May 2018.
[17] L. He, N. Liu, “Load profile analysis for commercial buildings
microgrids under demand response”, In Industrial Electronics and
Applications (ICIEA), 2017 12th IEEE Conference on, pp. 461-465,
2017.
[18] C. Wagner, C. Waniek, U. Häger, “Modeling of household electricity
load profiles for distribution grid planning and operation”, In Power
System Technology (POWERCON), 2016 IEEE International
Conference on, pp. 1-6, 2016.
[19] G. Zhemerov, N. Ilina, D. Tugay, “The theorem of minimum energy
losses in three-phase four-wire energy supply system”, IEEE, In
Intelligent Energy and Power Systems (IEPS), 2016 2nd International
Conference on, pp. 1-3, 2016.
[20] M.Yu. Artemenko, L.M. Batrak, V.M. Mykhalskyi, S.Y. Polishchuk,
“Analysis of possibility to increase the efficiency of three-phase four-
wire power system by means of shunt active filter”, Technical
Electrodynamics, no.6, pp. 12-18, 2015.