Introduction to Smart Grid

Prof. N. P. Padhy
Department of Electrical Engineering
Indian Institute of Technology, Roorkee

Lecture – 33
Simulation and Case Study of AC-DC Hybrid Microgrid

Welcome you all to the online NPTEL course on smart grid, today we will be a focusing
on a case study, with respect to AC-DC smart grid and, we will try to take up an
application, where we will try to explore the benefit of having a AC-DC smart grid. So,
that the main objective here is to highlight the advantage of a AC-DC hybrid grid against
the conventional AC grid.

So, in this context we will take up an case study, that is coordinated voltage controls
scheme applications and we will see, how this coordinated voltage control or CVC
scheme can be adopted well, with our AC-DC hybrid grid instead of a conventional AC
grid. And perhaps we will take advantage of the presence of a DC grid DC microgrid
along with a AC distribution system.

Now, first of all what is voltage regulation I mean or voltage control and, what are those
devices, what kind of schemes are in practice and, during different state of operation,
how do they react and what sort of you know modification, or improvement is being
expected looking into a merger of AC-DC smart grid.

Now, first of all if you concentrate on voltage regulating devices, we can just focus on
OLTC DSTATCOM and all the converters connected to my renewable energy systems.
(Refer Slide Time: 02:19)

Now, if you come focus on coordinated voltage control, or CVC scheme the main
objective of the CVC is to optimally utilize different compensating device to increase
reactive power reserve, means the objective here is how to increase the reactive power
available within the system to take care of the voltage control.

Now, mean time we can also try to achieve, the post fault voltage recovery time can be
reduced as soon as possible may be or the time the post fault voltage recovery time can
be reduced with the help of CVC scheme. Now, there are two major type of CVC one is
decentralized and the other one is the centralized, centralized where we have
communication system in place, decentralized where we do not have any communication
among the devices.

Now, focusing on a simple test system, so, we have considered I triple E 33 bus
distribution system, which is as conventional efficiency.
(Refer Slide Time: 03:32)

Whereas, we have connected a DC microgrid, we have connected a DC micro grid at a

particular bus number 5, bus number 5 and this is the bus where we have connected a DC
micro grid and, we also have connected a DSTATCOM at bus number 22 and we also
have connected a DFIG at bus number 22.

And we do have a OLTC at bus number 1 so, what are the major changes we have
incorporated, we have taken I triple E 33 bus system and we introduced OLTC at bus
number 1, DSTATCOM as well as DFIG wind generator at bus number 22 and at bus
number 5 we have introduced a DC microgrid. So, excluding all those devices and the
rating the rest loading and the dimension as well as the variable, the parameters of the
lines and cables remains same as far the I triple E system.

Now, voltage regulating devices like OLTC DSTATCOM etcetera, as well as DFIG
converters are operated in a decentralized manner; that means, all this system, if you see
all the devices do operate without any communication means those devices do try to you
know improve the voltage profile at that particular bus where it has been connected
through different schemes.

But they do not communicate to each other for the overall system benefit at large, and
hence they perhaps regulate the voltages in their own way at the respective buses, but
they do not look at other buses. Real time voltage information at bus number 1, 5 and 22
are provided to the corresponding voltage regulating devices connected at bus number 1
5 and 22.

So, over ally what I what we wanted to focus that this devices do control the voltage at
the respective buses like bus number 1, 22 and 5 but they do not really communicate
among themselves to see, what kind of scenario other buses do face. There are different
voltage regulating devices and the very common is OLTC.

(Refer Slide Time: 06:11)

OLTC changes its step position if the voltage of the first bus vary within plus, or minus
10 percent of it is nominal value, that is if the voltage is varying between 0.9 to 1.1, and
then they can you know take action is the voltage variation at the regulated bus of OLTC,
crosses the predefined dead bend if it crosses between 0.9 to 1.1 minimum as well as
maximum, then the OLTC try to change its tap positions.
(Refer Slide Time: 06:50)

Now, similarly the DFIG and the DSTATCOM, what we have done those are the
additional devices have been placed to look into a to take care of the system reactive
power scenario. So, a 1.1 megawatt rated DFIG based wind system is connected at bus
number 22 through a 0.69 k V slash 12.66 k V step up transformer. A.8 MVA
DSTATCOM is also attached at the same bus, furthermore a pair of switchable reactive
load each rated with 0.4 megawatt and 0.2 MVAr also connected at bus number 22.

So, these are the three things which has been additionally connected to my I triple E 33
bus system at bus number 22.
(Refer Slide Time: 07:57)

Now, we also have you know included a DC micro grid at bus number 5, where at bus
number 5 the DC microgrid is connected through 2 MVA rated DC AC converter, it
includes a PV source operating 1.8 megawatt operated at its MPPT with help of
unidirectional DC, DC convertor.

A stack of lead acid batteries with capacity 0.4 megawatt is also connected at the same
bus. A pair of switchable DC load rated with 1.7 megawatt is also connected at the same
bus, along with this a pair of switchable AC loads with each of the rating 0.12 megawatt
and 0.06 MVAr also connected at bus number 5.

Apart from the change in load rating adopted in bus number 22 and 5, the rest of the
parameter as I mentioned earlier now similar to I triple E 33 bus distribution system.
Those parameters, which has been chosen for this analysis need not be a fixed variable,
just for a case study we have chosen this parameter, but you are very open to choose your
own parameter own settings and carry out the similar simulation.

(Refer Slide Time: 09:25)

Now, in any system we do have come across three different states are the first one is
steady state and, then dynamics state and finally, or transient state and now what kind of
you know reactive power in voltage scenario during all those states a first of all if you
consider the state number one, where actually we say the steady state at which small
loads are being you know switched on the variation is not significant very you know
minor change on your load. And the reactive power supplied by the fast acting
converters, which takes part in the contingency situation are less than 40 percent of there
a rated capacity or available rating whereas, the voltage magnitude of the regulated buses
are within 0.9 to 1.1.

But if you move to a dynamic state where actually any sudden load changes will fall
under this stage two category, the reactive power support required from the fast acting
converters are more than 40 percent of their available capacity during the steady state it
is below 40 percent and during dynamic state, we expect it is to be more than 40 percent
on their valuable rating.
(Refer Slide Time: 10:24)

In order to build the reactive power reserve the reactive power support of these devices
are hard limited 40 percent of the limit ok.

So, we limit hard at 40 percent whereas, the bus voltages are the magnitudes are
regulated between 0.9 to 1.1, the final one which is a very important the transient state

(Refer Slide Time: 11:13)

When three phase to ground fault, or three phase to ground temporary fault is triggered
for a duration of 150 millisecond, the voltage of the regulated bus suddenly falls below
0.9 per unit. And hence the system enter into a transient state here, the fast acting
converters are allowed to inject reactive power, up to their maximum capacity as
maximum as 100 percent.

(Refer Slide Time: 11:37)

Now, if you look into this two different flow charts, the CVC schemes are not very new
they are very I mean very old and being practiced across the world and, where as the
conventional scheme you can see that you know, it is bus simple algorithm whereas, in
the proposed CVC scheme what I am going to talk about today is slightly different where
we taken care of the handling of different converters. So, that you know the overall
performance of the system is being improved and, further if you see what are the
challenges with the existing CVC scheme. Now, the current limitations of the CVC
scheme coordinated voltage control schemes.
(Refer Slide Time: 12:32)

Given the importance for the speed of operation of devices only because it really takes
care or it give importance to those devices where past, does not check for the availability
of the devices for particular operating conditions, all the fast acting devices are
considered for injecting reactive power simultaneously along with slow devices.

So, what we do when it is from stage one or stage two or stage three all the reactive
power injected devices are instructed to you know inject as maximum, they can and in
that process the fast acting devices are always you know regularly participating in
injecting reactive power and, the slow devices unfortunately may not be able to
contribute by the time all the fast acting devices do contribute, whatever major
contribution they can make.

But that is not a good idea in the proposed model what we have suggested, utilized the
devices more effectively even though OLTC, DSTATCOM, DFIG, DC micro grid all of
them have to participate simultaneously, by which it improves the voltage profile of the
system at large, it improves the fault ride through capability of the devices, it also
improves the transient conditions of the systems.

(Refer Slide Time: 13:59)

Now, how it is being proposed here in the first stage, identifying the operating conditions
of the gird. Based on the real time voltage information from the voltage regulating buses
and, the reactive power injected by the fast acting converters a particular operating
condition of the grid is identified. And then in the second stage checking for the
availability of the devices and assigning the master and slave roles.

So, who will act as a master and who will act as a slave depending upon the fast acting,
or slow acting devices. It is required to first check for the availability of the device, for a
particular operating condition before assigning any role to it for example, OLTC and DC
micro grid are you know used during steady state and dynamic state whereas, the wind
generators and this DSTATCOM comes are available throughout all the three states.

Now, the final state introducing the time delay in operation for example, you want some
slow device to be you know acted before the fast devices, than a delay in action on those
fast devices can be incorporated. So, that the slow devices can you know inject the
reactive power and further the fast acting devices can come to action.
(Refer Slide Time: 15:25)

So, as an intentional time delay is inserted for all fast acting devices like WGSC and DC
microgrid DCMG converter along with the DSTATCOM, this ensure that the slow acting
of OLTC would complete it action.

So, as we discussed earlier when all those fast acting devices come to action, then the
slower acting devices may not be able to contribute much. So, please allow the slow
device like OLTC to act first and, then the fast devices can come and follow them. After
which the fast acting devices are allowed to operate. The minimum delay time is
calculated for each and every bus, where the fast acting converters are connected. If the
waiting period is less than this delay the time, then the OLTC might not complete its
action.
(Refer Slide Time: 16:18).

And this is what the proposed CVC scheme, because most of the control schemes we
have discussed in past. So, I am not taking lot of time here, but is your whole you can see
that this is basically DC-AC bidirectional converter implemented with the proposed CVC
scheme and whereas, we talk about actually the modelling for a DC grid bus voltage and,
this talk about the generated duty ratio for the bidirectional DC DC convertor of the
battery and, the final one talking about the track MPPT of PV within DC microgrid.

So, you can see different control algorithms, or control schemes or the strategy being
developed to execute the proposed case study. Now, in the result and discussion section,
we will see what kind of outcome it is because by adding a time delay and, allowing all
the you know devices to act in a different phase of time not simultaneously through
which the slow devices do not contribute much.

And hence with the proposed CVC scheme, we will see what kind of advantage of merit
has been achieved.
(Refer Slide Time: 17:32)

In order to validate the effectiveness of the proposed scheme, the system is tested under
all the three operating states of the grid, the modified I triple E 33 bus distribution system
is simulated in the real time digital simulator RTDS platform. And we have consider the
grid rated voltage of 33 k V and frequency of 40 hertz and the OLTC substation
transformer of rating 10 MVA primary voltage 33 kV, secondary voltage 12.66 k V and
the leakage reactance assume to be 0.15 per unit.

The DSTATCOM capacitors 2000 microfarads and DC link voltage of 600 volts and
windshield speed considered to be a 12 minutes 12 meter per second. The PV parameters
of a DC microgrid its 100 ohms shunt resistance series resistance of 0.5 short circuit
current of 8.85 ampere open circuit voltage of 37.6 volt and, solar irritation of 1000 and
number of panels in series and parallel 24 and 300 power rating of each panel is 250 watt
the battery lead acid battery of four 100 volts and DC DC converter of 0.5 megawatt.

So, this is basically the system parameters being discussed before we proceed further
simulation.
(Refer Slide Time: 19:15)

And during the RTDS simulation we have seen, this steady state operation for the PV
output and for battery power, real power supplied by the DC microgrid and, reactive
similarly the reactive power supplied by the DC microgrid and, this is the DC load and
this the DC grid voltage parameter. So, during steady state operation of DC microgrid all
you know parameters being interest whereas, steady state and dynamic state operation of
various devices at different AC bus of the distribution system, we have seen the real and
reactive power loaded bus number 5 and 22.

Whereas this is OLTC tap positions how they are keep on changing real power and
reactive power by the DC microgrid and, reactive power supplied by the DSTATCOM
and real power and reactive power supplied by the wind generator. So, this all a
simulation results for your reference.
(Refer Slide Time: 20:11)

And perhaps we can say the blue line represents the proposed CVC and red one is the
existing CVC and the black one is without CVC.

So, we compare if you do not have the CVC scheme, how does it behaves and if you
have based on the existing practice how the scenario would be and the proposed one so,
all three have been compared the first you know, if you see the bus number 22 and this is
at bus number one and bus number 5.

So, this is how actually we have traced at different location bus number 22 where we had
both DSTATCOM and wind generator, bus number 5 we had a microgrid and bus
number one we had a OLTC. Now, the state numbers one and two that is steady state and
dynamic states have been, you know considered and the voltage profile of regulated
buses are being plotted.

Whereas in this diagram it is the state number 3, that is the transient state which is very
important from the state one and you can see the state three is being created at each and
every bus, we can see the voltage profile of regulated buses. Now, the state three, where
we are during the transient state the operation of voltage of regulated devices are being
plotted. So, during the state three these are the bus positions and these are the device
positions OLTC DC micro grid and wind generator and my DSTATCOM, RTDS
simulation results.
(Refer Slide Time: 21:53)

Where the voltage profile of 33 bus are being plotted and with both steady state, as well
as dynamic state. So, the first one is during steady state and the second one is during a
dynamic states are being plotted.

(Refer Slide Time: 22:25)

Now, let us see what exactly happened during the state number 1, that is steady state
conditions, how the DC microgrid operated or behaved as per the proposed CVC during
this state the OLTC and the DC microgrid converter act as a master whereas, the
DSTATCOM and wind generator act as a slave. The results of the DC microgrid with
proposed CVC scheme are shown in previous figures, where initially the DC microgrid
is not feeding any real power to the utility, the power supplied by the PV is more than
required by the DC load.

And hence the battery enters into charging mode to regulate the DC grid voltage at, the
interval t 1 as you have seen interval t 1 DC microgrid starts feeding real power to the
utility and, in this scenario the DC bus voltage regulation is achieved by supplying
excess power from the DC microgrid to the utility grid.

(Refer Slide Time: 23:27)

Now, during the interval 42 the PV output is decrease, due to fall in PV irradiation and
hence the real power supply to the utility reduces to zero.

During the interval t 3, where there is a switching of additional DC load within the DC
microgrid and the deficit power is supplied by the utility grid. Thus in spite of all the
switching dynamics the controller of various converters of DC microgrid perform
satisfactory, the DC grid voltage is regulated at 1200 volt during all modes.
(Refer Slide Time: 24:05)

Now, in case of state one and two both steady state and dynamic state conditions when
the operation of various voltage regulation regulating devices, the objective of the
proposed CVC scheme during steady state condition is to increase the reactive power
reserve of the DSTATCOM and the wind generator and, also to improve the voltage
profile of overall distribution system.

Now, in case of without CVC all the voltage regulating devices rush towards regulating
the voltage of their respective bus, simultaneously without any concern of the reactive
power reserve. Now, as we discussed like you know during emergency, or if you do not
have a proper CVC scheme, then they all try to you know help the system at their own
locations and perhaps without looking into how much reserve, they carry with them
though there is a best utilization of DC microgrid converter the OLTC is poorly utilized
with only one switching.
(Refer Slide Time: 25:08)

In case of existing CVC scheme though there is an additional switching of OLTC, the
DC micro grid converter is poorly utilized, because of all the fast acting converters along
with OLTC are operated simultaneously. Once DC micro grid DSTATCOM wind
generator reaches 40 percent of the rating, they are stop from injecting reactive power
further.

The major problem is that they all acts simultaneously the moment, they reached to their
40 percent capability they have not mono more allowed to act; however, the proposed
CVC scheme fully utilized both OLTC and DC microgrid converter, by providing an
intentional time delay in operating the DC micro grid converters thus the voltage profile
of the regulated bus 5 is improved by an amount delta v as shown in the previous
diagrams.
(Refer Slide Time: 26:08)

Now, the voltage profile of the entire distribution system is improved in case of the
proposed scheme, that can easily be verified the reactive power reserve of the
DSTATCOM and the wind generator. In the case of the proposed CVC scheme shows as
significant increase by a maximum amount of delta q D of 40 percent and, delta q W of
40 percent both DSTATCOM, as well as wind generator to 40 percent and by a minimum
amount of 14 percent and 15.33 percent respectively compared to the existing CVC
schemes.

Furthermore the reactive power reverse of the proposed CVC scheme is increased by a
maximum amount of 59 percent and 56 percent by both DSTATCOM, as well as wind
generator and by a minimum amount of 19 percent and 15.33 percent respectively
opposed to without CVC scheme respectively by imparting time delay in their operation.
(Refer Slide Time: 27:13)

Now, state number two dynamic state condition in order to study the performance of
proposed CVC scheme under dynamic state condition load at bus number 5 and 22 are
increased by 100 percent. So, to introduce a sudden load change 100 percent load change
have been introduced at both bus number 5 and 22.

At instant t 4 as a consequence the voltage at these buses fall down from 1 per unit due to
which the DSTATCOM and WGSC increase the reactive power injection after passing
through a delay period. Immediately after reaching 40 percent of the capacity they are
force fully stopped from injecting reactive power, in order to retain the same reactive
power reserve so, that the existing CVC schemes.

(Refer Slide Time: 28:10)

However the reactive power reserve of DSTATCOM and WGSC, in case of the proposed
CVC scheme is higher than that of without CVC by an amount 37.7 percent and 52
percent respectively for both the DSTATCOM and wind generators. In spite of additional
switching of OLTC in the existing scheme the voltage profile of the overall distribution
system with propose scheme is better and, which has been observed in the previous
pictures.

(Refer Slide Time: 28:40)

Now, let us move to the transient state conditions a three phase to ground fault
introduced into the system for a duration of 150 milliseconds due to which the voltage
magnitude of the overall system falls below 0.9 per unit.
Now, during this state both WGSC and DSTATCOM play the master role and inject
reactive power up to their maximum capacity, now it is reverse not necessary for those
WGSC and DSTATCOM currently, now act as a masters and try to evacuate all the
reactive power available to them. The OLTC and DC micro grid will be made slave and,
hence does not take part in the reactive power injection under transient state conditions.
The DC microgrid is isolated from the utility grid, and operate in an islanding mode of
and the feed the local DC loads.

The guage for deciding the efficiency of the control scheme is the ability by which the
system recovers from the transient state at the earliest. So, the very interesting part here
that you know we try to take advantage in the presence of DC microgrid, along with
other devices like DSTATCOM and the wind generator. And sometimes we force the
OLTC to act in the beginning, but if it is emergency then we do not allow OLTC to
action in the initial stage. So, the control strategy the propose CVC certainly, you know
takes care of many other issues those have not been identified or addressed in case of the
existing CVC.

(Refer Slide Time: 30:28)

Ah it can also be observed that as soon as the fault is cleared the regulated bus start to
regain the voltage however, due to the increase in reactive power reserve in case of
propose scheme, the system quickly reaches 0.9 per unit by a duration of delta t 1
compared to the existing CVC scheme and delta 2 compared to the without CVC. So,
that time of recovering got reduced with the propose CVC scheme compared to the other
schemes.

Thus the proposed CVC scheme needs less fault voltage recovery time in contrast to the
existing schemes.

(Refer Slide Time: 31:07)

Further it can also be observed that after entering into steady state the system reaches 1
per unit much faster in comparison to other schemes, due to the better utilization of the
DC microgrid converter and OLTC. In addition to this the propose scheme has better
voltage regulation compared to existing schemes, after entering into steady state from the
transient state.

The operation of various voltage regulating devices is displayed from which it is evident,
that once the fault is cleared the system rebuilds the reactive power reserve. And to
conclude what we wish to highlight the propose CVC coordinated voltage control
scheme.
(Refer Slide Time: 32:00)

Increases the reactive power reserve of the DSTATCOM, as well as WGSC, and also
improve the voltage profile of the overall distribution system during normal conditions.

(Refer Slide Time: 32:19)

Furthermore the proposed scheme is also targeted to reduce the post fault voltage
recovery period of the system. An intentional time delay is introduced for all the fast
acting convertor for achieving the set of objectives.

The OLTC and DC microgrid converters are effectively utilized to improve the overall
voltage profile of the distribution system, during steady state and dynamic state
conditions. It was observed that the propose CVC scheme shows improved performance
in comparison to the existing CVC schemes, in terms of reduction in post called voltage
recovery time.

(Refer Slide Time: 33:11)

These are the following references can be used, this is a very interesting case study
where you can see the benefits of having AC-DC smart grid together to address the
voltage control scheme.

Thank you.