Protection and Design of Distributed Energy Generation System (Microgrid)
Protection and Design of Distributed Energy Generation System (Microgrid)
Protection and Design of Distributed Energy Generation System (Microgrid)
www.ijiset.com
ISSN 2348 7968
I. INTRODUCTION
Due to the increased concern over global climate change,
the demand for clean sustainable energy sources has increased
greatly. One of the major problems with these sources, such as
wind and solar, is integrating them into the larger power grid.
One proposed way is to have distributed renewable generation
sources integrated into a microgrid. A microgrid is defined as
a low to medium voltage network of small load clusters with
Distributed Generation (DG) sources and storage [1]. It can
operate connected with the larger grid or islanded in the event
of a grid fault. A microgrid is controlled by a single controller
and is viewed as a single load or generation source by the
larger system. It can be operated in industrial, commercial,
and/or residential areas. One major challenge with operating a
distribution level microgrid with Renewable Energy Sources
(RES) connected to the system with inverters is protection
against faults. In this paper we analyze fault currents on a
larger system of distribution network than that used in [1], [2].
Our system is protected with a standard distribution
protection scheme. Our
objective is to see how the protection will need to be changed
to facilitate microgrid operation with the inclusion of DG
sources and islanding capabilities. Our system was built and
simulated in Matlab Simulinks SimPowerSystems..
286
IJISET - International Journal of Innovative Science, Engineering & Technology, Vol. 1 Issue 6, August 2014.
www.ijiset.com
ISSN 2348 7968
B. Models of DG Sources
The inverter connected to the solar arrays is a current
converter made with a three-phase IGBT/Diode Bridge
controlled by a PWM generator with a voltage/current
controller. The PWM has a carrier frequency of 4,320 Hz.
This is connected in series to an LC low-pass filter for
improved power quality. An additional LC filter was used in
the input to maintain the quality of the dc input signal. The
input is any dc source rated between 50 kW and 570 kW and 5
kV. The output is a three-phase 12.47 kV system. Maximum
allowable current is limited to 90 A, approximately twice the
maximum power current. This is an adaptation of the inverter
used in [4].
The diesel motor model used in this study is a modified
version of the one used in [5] and is connected to a 300kW/12.47-kV synchronous generator Simulink block.
The wind turbines are based on the wind turbine Simulink
model demonstrated in [6]. This turbine is connected to a
100-kW and 400-kW squirrel-cage induction generators. The
turbines are given constant wind input needed for maximum
capacity generation.
The model for the PV modules is a simple single diode
approximation with two resistors as shown in Fig. 3. This
circuit is used to model a Sharp ND-Q0E2U 160-W module
with I-V characteristics obtained from [7]. The resistor values
are computed with the equations given in [8]. Though this
model is not the most accurate, we feel it is sufficient for this
study as the current limiting features of the inverter will
dominate the dynamics in fault simulation. This module
model is used to construct 564-kW arrays.
The DG sources are placed in desirable areas, with the
diesel generator supplying a balanced three-phase load which
will most likely be industrial and have a backup power supply.
The solar arrays are all, with the exception of the array at bus
12, placed with loads to simulate the aggregate effect of
rooftop solar panels on homes and businesses. The wind
turbines are placed on buses without loads as usually large
turbines are placed in wide open areas or hilltops near
transmission.
C. Modeling of PV Array
PV arrays are built up with combined series/parallel
combinations of PV solar cells, which
are
usually
represented by a simplified equivalent circuit model such as
the one given in Fig. 4.1 and/or by an equation as in (1).
Fig. 2
The PV cell output voltage is a function of the photocurrent
that mainly determined by load current depending on the
solar irradiation level during the operation.
Equation I
A possible model of a photovoltaic array and inverter [4] was
researched. After
verifying the theory its features were
modeled in Simulink. This proved moretime consuming and
complicated than intended. Like the theory, the modelworked
both under load and no load conditions.Figure 3.4 shows the
model of the PV array system. The blocks of mostinterest
are shown on in the top layer as seen in the diagram
Fig. 3
D. Modeling of Wind System
Fig 1
IJISET - International Journal of Innovative Science, Engineering & Technology, Vol. 1 Issue 6, August 2014.
www.ijiset.com
ISSN 2348 7968
Fig 5
Fig 4
E. Modeling of Diesel System
A plant consisting of a resistive and motor load (ASM) is fed
at 12.47KV from a distribution 32 kV network through a 6
MVA 32/12.47 kV Wye-Delta transformer and from an
emergency synchronous generator/ diesel engine unit (SM).
The 32 kV network is modeled by a simple R-L equivalent
source (short-circuit level 1000 MVA) and a 5 MW load. The
asynchronous motor is rated 2250 HP, 2.4 kV and the
synchronous machine is rated 3.125 MVA, 2.4kV. The SM
excitation is performed by the standard excitation block
provided in the machine library. The diesel engine and
governor system are modeled by a Simulink block. Initially,
the motor develops a mechanical power of 2000 HP (1.49
MW) and the diesel generator is in standby, providing no
active power. The synchronous machine excitation system
controls the 2400 V bus B2 voltage at 1 pu. At t = 0.1 s, a
three-phase to ground fault occurs on the 32 kV system,
causing opening of the 32kV circuit breaker at t = 0.2 s
Fig. 6
In this study, we are interested in evaluating the maximum and
minimum fault currents at each bus in the microgrid. The
system was built in Matlab Simulink using the
SimPowerSystem toolbox. The system is simulated using
Simulinks ode3 with a fixed time step of 1 microsecond. The
four major faultssingle line to ground, line to line, double
line to ground, and three phase faultsare initiated in the
system 0.1 second after the system had reached steady state
and is sustained for another 0.4 second. This allows the rms
values of the symmetric fault currents to be measured. The
fault impedance is chosen as 1m. The system is only
simulated in the islanded mode as the fault currents at each
bus for the original system configuration are given in [3].
Faults with DG source contributions in grid-connect mode
can therefore be easily computed using superposition. Fault
currents are measured on both the high (closest to substation)
and low (farthest from the substation) side of the fault as most
locations had generation on both sides. Currents are also
measured at the high and low buses, (where high and low
mean the same as above) that would need to have backup
protection if the devices at the particular bus should fail.
288
IJISET - International Journal of Innovative Science, Engineering & Technology, Vol. 1 Issue 6, August 2014.
www.ijiset.com
ISSN 2348 7968
Fig 8
Fig 7
B. Wind System Results
The wind speed (10m/s) is such that the wind turbine
produces enough power to supply the load. The diesel
generator (not simulated) is stopped and the synchronous
machine operates as a synchronous condenser with its
mechanical power input (Pm) set at zero. The example
illustrates the dynamic performance of the frequency
regulation system when an additional 25 kW customer load is
switched on.
When simulation is run, voltages, currents, powers,
asynchronous machine speed and system frequency on the
two scopes are observed. Initial conditions (x Initial vector)
have been automatically loaded in workspace so that
simulation starts in steady state.
As the asynchronous machine operates in generator mode,
its speed is slightly above the synchronous speed (1.011 pu).
Fig 9
289
IJISET - International Journal of Innovative Science, Engineering & Technology, Vol. 1 Issue 6, August 2014.
www.ijiset.com
ISSN 2348 7968
Fig 10
IV. CONCLUSIONS OF MICRO-GRID MODELLING
In this final chapter the project plan, aims and outcomes
will be highlighted and summarized. The project is discussed
in terms of what it aims and how it could contribute to the
power industrys needs. It also explores how the project could
be extended or improved and how this might be done. This
includes what can be done in the future to understand
micro-grid behavior, with the goal of micro-grids use and
commercial development.
A. Project Conclusions
It is hoped that, by making use of the small and varied
sources which comprise it, micro-grids may be able to make a
significant contribution to the power generation and
distribution market For instance, if the sun is out the PV array
may provide power, if its windy the wind turbine, if it is
neither or if more power is needed, mains supply can be used.
The inclusion of batteries in a micro-grid system would also
allow excess power produced to be stored, or alternatively the
excess power could be put into the main grid.
In this way it is expected that micro-grids could reduce
pollution and deliver reliable energy in a variety of situations
as discussed. Micro-grid behavior is on the whole not well
understood. For this reason this project aimed to develop
models suitable for analysis and investigation. The project
aim was to model behavior of micro-grids individual power
sources, and time allowing a micro-grid system.
A final aim was to lay groundwork which would allow
analysis for the further development of a more sophisticated
model. More specifically, it involved modeling a photovoltaic
cell, a wind turbine. To this end the project has been
successful. All models developed will allow for investigation
that will provide an understanding of micro-grids to facilitate
the evolution of a more sophisticated model.
This project was carried out by way of extensive
research, model design, modelling, testing and development.
Each power source model was completed independently
starting with the PV array.
The PV array works as expected fora changing input.
This is a good result that reflects real situations well. The
wind is not yet fully completed. The results thus far are very
good, and it is expected they will improve prior to open day
through further testing and development. These sources will
be connected together to form a micro-grid. It is anticipated at
least two of the three power sources will be connected
together to power a single load.
B. Micro-grid Modelling and the Future
As previously discussed, the goals of this project were
prescribed in terms of how the model developed will be used.
The next step should be to further develop the micro-grid as a
whole. It is important to learn more about how the sources
interact with each other that is do they enhance or interfere
with each other. More specifically their relationship to each
other needs to be defined. I fall goes as anticipated and the
micro-grid system as a whole is developed, the control of the
system will likely be imbedded within the electronics. It could
be possible to use a specialized controller to get a more stable
response and to use each power source more efficiently. This
should certainly be researched and considered once the power
sources interaction and relationship to each other and the
mains has been defined.
Another aspect that could be developed further are the
individual sources within the micro-grid. This could happen
on two levels. The first is the consideration of other variables
for each source. For example, wind speed is not considered
290
IJISET - International Journal of Innovative Science, Engineering & Technology, Vol. 1 Issue 6, August 2014.
www.ijiset.com
ISSN 2348 7968
Microgrid
C. Final Remarks
On the whole, this project The Modelling a Micro-grid
System has been successful. Models which allow for
investigation of the individual power sources behavior have
been developed and it is expected that a micro-grid unit will
be modelled prior to open day. The project was carried out by
doing extensive research and by using a design process to
implement each system individually. Testing and
development through understanding was also a significant
part of this project. The goals of this project have been met
and it is anticipated further research and development will be
carried out on the system, with the goal that micro-grids will
be able to make a valid, greener, contribution to the worlds
growing energy needs.
V. REFERENCES
[1] Microgrid Protection Using Communication-Assisted
Digital Relays,EricSortomme, S. S. Venkata and
JoydeepMitra,IEEE TRANSACTIONS ON POWER
DELIVERY, VOL. 25, NO. 4, OCTOBER 2010
[2] "Fault Analysis and Protection of a Microgrid" E.
Sortomme, Student Member, IEEE, G. J. Mapes, B. A.
Foster, S. S. Venkata, Fellow, IEEE
[3] A Photovoltaic Array Simulation Model for
Matlab-Simulink GUI Environment I. H. Altas1,*
and A.M. Sharaf2
[4] Dynamic Model and Control of a Microgrid with
Passive Loads , M. Popov, H. Karimi, H. Nikkhajoei,
V. Terzija
291