Lec 1

Download as pdf or txt
Download as pdf or txt
You are on page 1of 25

Power Electronics and Distributed Generation

Prof. Vinod John


Department of Electrical Engineering
Indian Institute of Science, Bangalore

Lecture - 1
Course Introduction and Overview

Welcome to this class on Power Electronics and Distributed Generation, there are set of
students in the according room of this class also this is been recorded for the saving to the
internet, so welcome to all of you.

(Refer Slide Time: 00:33)

In this course we will be looking at distributed generation and in particular power electronics
concepts, which are important for distributed generation. Power electronics is a wide area and
it covers a large number of segments, a very wide spread segment is DC to DC converters,
such as switch mode power supplies, which are used in computer power supplies, household
equipment, wide range of a applications, then you have inverters which are extensively used
in industrial applications; especially motor drives.

Then you have power quality equipment, a very common power quality equipment that is
used is a UPS to ensure that you have power availability, even when there is a disturbance in
the grid. Also you have harmonic filters, you have a compensators to compensate VARs,
compensate harmonics to adjust your voltage in the presence of sags, swells, etc. These are
all power quality related segments for power electronics and at the very high power level you
have HVDC, which is high voltage DC systems, flexible AC transmission systems, which are
used at the very high power level.

Then an emerging area which is gaining increasing importance is distributed generation,


where the power converter or any converter in general is used as an interface between your
energy source and the grid. So, you can have power converters, you can have machine based
interfaces and in general you could call it a power conditioner, which is used to interconnect a
distributed generation unit with the electrical grid.

They are many other applications of power electronics in high voltage, pollution control,
plasma generation, many applications in medical imaging. But, focus of this course will be on
the distributed generation and the issues of connecting the distributed generation unit with the
electrical grid.

(Refer Slide Time: 03:09)

If you look at the structure of the course there will be three parts to the course, the first part is
looking at the power system related issues, specifically relating to the distribution system.
And we will address the issue of why we need to look at the distribution system closely and
the many issues on a steady state basis how large a distributed generation source can be
connected with the distribution system. Also transients such as, a big transient is when a fault
occurs on the distribution system and how would your distributed generation source in
behave in the presence of such a transients.
What is the strategy to deal with protection, especially when you have faults and one thing
that can happen, when you have additional source that is in the distribution system is that you
have the potential to form an island. If you look at the traditional distribution system, the
sources, essentially the grid coming from the substation to the feeder and you have a feeder
coming from the substation to the individual loads, there might be distribution transformers
coming to residential areas, commercial areas, industrial areas.

And you are looking at how to interface such systems with the electrical grid and we need to
understand the distribution system related issues. The second part of the course will look at
the issue of comparison and bench marking. And the reason this is important is, you can have
a wide variety of distributed generation sources, you can have wide variety of interconnection
systems, power electronic circuits, different circuits, which can do the interconnection.

And the question is are the type of systems that you are considering is it actually improving
the situation or like are you going backwards? So, it is important to come up with a way of
quantifying how good your interconnection system is, to make sure that you are going in the
right direction. And, so it becomes important to actually come up with a method to evaluate
the value of what you are doing and then make a decision is this change that you are making
taking you in the right direction.

And this becomes an important issue, where often you have to compare the distributed
generation source with the traditional electric grid. And the question is, is it giving you
something better are you are you going in a direction where things are getting worse. The
third part of the course is related to the power electronic design issues, where we will look at
the power electronic circuits for distributed generation applications, we will look at single
phase, three phase, and different topologies.

The basics of the topologies and the components that go into a typical power electronic
circuit. And the component that go in a typical power electronic circuit, they are essentially
inductors, capacitors and switches, you would not intentionally put a resistor in a power
electronic circuit because it is a dissipative source. So, the question is what sort of inductors,
what sort of capacitors, how do you rate them, how do you size them, what sort of
semiconductors, how do you rate them and size them, should go into such an equipment.

They are implications of the selection in terms of both efficiency and reliability and ideally
there is no loss in an inductor, capacitor or switch. But, you always have non-idealities in a
practical circuit, so in a practical circuit you will have losses and there are implications of
loss, one is you lose the generator part that your generating in your generation source in loss.
And you have power efficiency that is one impact, the second impact typically is you have
more loss, it means that you need more thermal management and your compacts can get hard,
which can have reliability implications.

So, the parasitic’s of the power convertor component selection has implications in terms of
both reliability and efficiency. And that has to be considered when you actually do the design,
so that you are taking the overall picture into mind, rather than just “does this component
work?”. Then will also to some limited extend look at modeling and control related issues of
such a system.

(Refer Slide Time: 08:33)

If you look at the typical background required for the course at IISc over here, we have
students who come in for masters degree or for research degrees, they would have already
done courses in power electronics, they would have had courses in power systems. So, at
introductory level you would have a power electronics course, you would have power
systems course, which gives you some exposure to distribution systems, you would have
done courses on switched mode power electronics, you would have done courses which help
model motors, so you would have courses on motor drives.

And the idea of this particular course is to take in aspects of power electronics, some amount
of machines and a power systems. And look at the issues, when you actually interconnect a
distributed generation source with the grid, also at IISc we have courses on details of the
power converters, such as how do you modulate the power converter. You also have courses
on advanced drives, where you look at details about control related modeling issues.

So, the idea was at this particular point the student can look at research related issues or some
particular topic, which you may want to investigate in detail. And hopefully this background
will give you the necessary tool to actually go in the direction of looking at power electronics
and distributed generation.

(Refer Slide Time: 10:19)

So, we talked about three aspects of a this course, so the first aspect was power systems
related. And as we mentioned the fault is a severe transient and you need to actually deal with
a fault in a distributed generation system and one of the basic requirement would be that you
have the existing distribution system with it’s protection philosophy. And you do not want to
modify the existing protection system, when you add in the distributed generation source.

So, suppose you have to rewire the whole distribution feeder, when you connect your solar
inverter that would bring in a big cost to the overall society just, because a person wants to
add a distributed generation source. So, you want your source to work in tandem with the
existing system and you do not want to modify the existing system, when you introduce the
new sources. Some of the background that you would use, when you look at say for example,
fault studies is you would typically do a per-unit analysis to actually evaluate what the fault
levels are, what the voltages and currents in the system are.
So, this would be something that you would have from your power systems background, you
also make use of symmetrical components, extensively do analyze faults. Very common fault
is a single phase fault. So, such methods can be useful when you are analyzing distributed
generation system and it is a impact on the existing system. Then another important scenario
is when you have a additional source in addition to the main grid, which is represents the
feeder coming in the substation.

If the feeder gets open for some reason, may be an upstream breaker is opened, then you have
to consider what happens to the protection system. If now a new source is actually holding up
portion of the feeder as an island, will the protection that was envisaged to work with the
original substation based protection, would it still work when you are operating as a island.
So, these are some of the concerns that traditional utility engineers have, when people want to
connect distributed generation sources, will the existing method of working, will it be
severely impacted by adding the distributed generation unit.

(Refer Slide Time: 13:21)

So, if you look at the structure of the power system, people like to talk about a distributed
generation as something similar to the old centralized computers versus the newer network of
computers; to some extent that comparison might be valid. If you look at the overall power
system, if you look at the sources in the traditional power system, they are shown as the
generators over at the very top. Then you connect the generators to the a meshed transmission
system, where you have power flows, which can be bidirectional from one side to the other
side.

You also have sub-transmission systems, which might be at a lower voltage level compared to
the main power grid, which then connects to distribution systems. And traditionally you did
not have the distributed generators connected to the distribution feeders and the power flow
on your distribution system from the distribution transformer, onwards was unidirectional.
But, now when you have a DG source connected you are looking at the potential for
bidirectional power flow, even on radial system, distribution systems and this has
implications.

(Refer Slide Time: 14:56)

If you look at the voltages and the role of power electronics at the different levels of such a
system, you have at the traditional generator, you might have a the generators operating at 11
kV may be 20 kV. And you step it up to the transmission level with a step up transformer, the
role of power electronics at this level might be in the excitors of large synchronous
generators, there might be some power electronics in the governor systems but, it is not a
major player at this particular point in the generator.

If you look at the transmission systems, the voltages are a quite high, we are talking about
400 kV and when people talk about a ultra high voltage, we are talking about 1200 kV etc.
And at this level you have HVDC systems, FACTS systems, which use very high power
electronics, you are sub-transmission system is at a slightly low voltage and even in this
particular level the role of power electronics would be in HVDC and FACTS at different
voltage levels.

When you look at the distribution traditionally distribution is 11 kV, can also be at 33 kV
level and when it comes down to the consumption point it is typically 415 volts three phase
or 230 volts single phase. So, the traditional role of power electronics has been in loads, later
switch mode power supply, motor drive at the load level at the consumption point. But, when
you are talking about distributed generation, you are talking about potentially connecting it at
the consumption level or at the distribution level.

So, this is essentially the range of areas, where we people looking at distributed generation
solutions. So, if you look at the way of interconnecting distributed generation systems, so
people talk about now having a network of sources connected at the close to the loads. One
thing when you have computer network, the constraints on the flow of information and the
limits on those constraints are quite some challenging.

But, if you look at the constraints, that governed, the connection of distributed generation
sources is not just the information required to control the source, you also have the physical
circuit laws, which govern the power flow, you have the thermal management required to
allow the power to actually come through. So, you have a lot more physical constraints in the
network of energy sources.

So, I would say even though people talk about the analogy, it is actually much tougher
problem, when you are looking at a network of energy sources and still at the stage of infancy
compared to the mature level that the network of information systems are at today
(Refer Slide Time: 18:36)

The concept of unidirectional versus bidirectional power flow has got a lot of implications,
((Refer Time: 18:50)) just again on the previous slide I just want to point out just because you
have a lot of people associate distributed generation with solar energy or wind energy. But,
that is not necessarily true all the time, I mean if you look at large wind farms they might be
even though the individual turbine might be rated at couple of megawatts, the wind farm
might be rated at 250 megawatts or 300 megawatts they would be connect connected to the
sub-transmission or the transmission level.

So, the wind energy needn’t be distributed generation, and similarly if you look at some other
newer solar power plans, they are rated at tens of megawatts especially at the large range. So,
they can be considered as a more like centralized generation, rather than distributed
generation, so when you are talking about the scope of DG systems, you are talking about the
power levels that are compatible with the distribution feeders or at the consumption point.

And it is not about the type of source, it is not about solar energy or renewable energy, often
you can have a diesel generator sitting over here, which is sending power to the load and it’s
a non-renewable fuel. But, it is distributed generation, because of the way in which it is
connected and the fact that it is distributed in two ways, one is it is connected at the
distribution level or at the consumption level. The second thing is geographically distributed
rather than being centralized in large individual point centralized power plants.
So, it is good to have clarity on the difference between the type of source and whether it is
distributed generation or not; just because you have one variety of source it does not mean
distributed generation. So, if you look at the issue of power flow in typical distribution
system, you have simplicity of the system of the distribution system and if you look at a
meshed system, where you have a possibility of power coming in from multiple sources.

(Refer Slide Time: 21:31)

Then you have a scenario, where say you have two sources, suppose you have a fault over
here at the center of the line, then to clave the fault you need to actually open say breaker 1
and breaker 2 to actually clear that fault. Because, just opening one breaker is not sufficient,
you need to disconnect all the sources to actually clear the fault. If you look at the traditional
distribution system, you might have a breaker at CB 3 and say you have a fault somewhere
down on the feeder.

And you have to open just one breaker and you can actually clear the fault on the feeder,
which means that there is lot of simplicity over here, the protection equipment is actually
reduced by half in this particular case, you need just one breaker compared to multiple
breakers in the other scenario. And also as you proceed along this particular feeder, the
ratings reduce as you go further down the feeder as the loads are typically connected
uniformly across the length of the feeder.

So, the rating or something which is much closer over here would be higher, the ratings that
are towards the middle of the feeder would be lower. So, you have simplicity and cost
reduction in a number of ways, so if you now connect say a distributed generation source
over here. Then you have this issue that now to clear a fault that you had on the DG system
you have the issue that you had in this previous case, where if you just open this breaker there
is now the potential this DG source would actually feed the fault.

And then you have now the issue of being able to coordinate multiple DG sources, it is not
just one potential source every customer on the feedback might potentially connect DG. And
you need to make sure that those sources get disconnected, when the protection philosophy of
the feeder has to actually kick in, so these are some of the concerns of not just simplicity. But,
also the cost impact of trying to actually connect more sources into a distribution system.

And, so you need to actually now bring in the DG, such that you can have the existing
distribution system. Because you cannot afford to put more protective devices or increase the
cost of the existing feeder, you have to actually operate the DG in such a manner that it
should not interfere with the existing protection philosophy of the feeder.

(Refer Slide Time: 24:43)

So, if you then look at the second aspect of what we will cover in this course, it is related to
the engineering economic aspects of connecting a distributed generation source. And if you
look at last year or two years back, they were people like to provide list of what are the grand
challenges facing humanity and one challenge was actually not just saying let us have solar
power happen, but to make it actually cost competitive.
So, the big challenge is not just building a solar inverter. There are actually 10's of 1000's of
solar inverters there, the challenge is actually to make it cheaper than coal, which is actually a
very cheaper source of power. So, if you can actually make solar energy lower cost, so
without actually thinking about what are the implications of your power electronics or the
source changes or the way you are doing interconnection, unless you are able to boil it down
to a cost you will not be able to address this issue; because the real challenges at the how to
actually rolled into cost.

So, can be it done? it can be done there are 1000's of people who have already done it and the
issue is, can it be done in a cost effective manner? And you have a number of tradeoffs in this
particular consideration of cost, if you look at a typical system, if you want to make some
component more efficient, you typically end up in a higher cost for it.

(Refer Slide Time: 26:50)

If you look at an example of say a simple wire, I mean if you have wire of some particular
cross sectional area of diameter say d and you will have the resistance R1 of a wire of
diameter d1. And, so just say R1 ohms and d1 is the cross sectional area of the wire and say
you want to reduce the loss in that particular wire and you decide to double the diameter. So,
if you double the diameter the area of cross section would go up by a factor of 4 so your
resistance would come down by a factor of 4.

So, you are at the handling the same level of current, your power loss would come down by a
factor of 4. But, if you look at the cost of the copper that is use in that wire, which is twice
the diameter, the volume has now gone up by a factor of 4, so you can see that your costs has
gone up your efficiency has gone up, but your cost is gone up. So, if you look at the issue of
cost versus losses you have a situation, where if you look at some parameter may be the
diameter if you are willing to pay a higher cost, then your losses can come down.

So, the question is the how much additional cost are you willing to tolerate to bring a loss
down to some particular level. At this particular point you can actually increase the cost by a
small amount and get a bigger benefit of loss reduction or efficiency improvement. Whereas,
over here you might not get match in terms of efficiency improvement, but your cost has is
going up quite rapidly. So, you have to actually make an engineering decision what is
appropriate, even at the simple case of selecting a wire.

And similar implications are there when you are selecting magnetic components, when you
are selecting capacitors, power semiconductors etc., where there is a tradeoff between cost
and the ratings of the components that you are selecting and this is one factor that you need to
keep in mind. So, just improving the efficiency forever will not take you in the right
direction, you have to ask yourself this is efficiency improvement being coming in at the
appropriate cost.

Not all situations of efficiency improvement is associated with increased cost, you might
have a situation where you get both an increase in efficiency and a reduction in cost
simultaneously. So, in that case it is a win-win situation and you have to go for it. Suppose
you have a power system, where you have power being processed in power converter one.
And then it is being processed in power converter two and if by some simplicity of design
you could actually merge it together and still retain the efficiency instead of processing it in
multiple stages.

Then you might be able to see whether you can actually get lower cost because you are
eliminating one big component and simultaneously you are reducing cost associated with a
big component. So, you have situations where you definitely have to choose a design based
on what would definitely give you benefit, but often in engineering you have situations like
this, where you are looking at a tradeoff between what additional amount you are willing to
pay and what is the benefit you are getting out of it.

Similarly, if you look at the issue of reliability, if you want to get improved reliability you
may have to end up paying a higher cost. For example, if you in the case of a example about
the wire that we just discussed, if you have a situation where, because of the increased losses
in the narrow wire, your wire temperature is high that higher temperature would lead to poor
reliability. So, the higher temperature poor reliability would mean that you would have higher
failure rate and shorter mean times between repair etc. So, if you want to improve reliability
you would then go and for a thicker wire, which means that you are paying a higher cost. So,
if you look at again the tradeoff that can potentially happen in such a situation.

(Refer Slide Time: 32:26)

You are looking at say, a tradeoff such as some component parameter, such as the diameter
and on the one hand if you are looking at increasing the diameter your cost is going up. But,

1
your mean time between failures say  would actually come down because you are having
a more reliable component. So you also have this tradeoff between reliability and cost. So, if
you are designing a power electronic component and you wanted it to last for so many years.

If you reduce the cost there is a chance that the component may fail before it reaches the
decided life time. And often in power generation systems you are talking about power plants
with 20 or 30 years of life, so can we meet the life of say for example, you build a dam, you
want it to last for 50 years or 100 years, if we are talking about a nuclear power plant, you are
talking about 40 to 60 years. So, if you were looking at a wind turbine people are looking at
wind turbines, which can last for 20 to 30 years.
So, if you look at solar panels, solar panels can actually last for a long time, the question is
will the electronics that go along with the panel, last for as long as the panel does. So, this
issue of tradeoff between reliability and cost also exists, there might be other considerations
too such as what is the weight or volume of a component. For example, if you want to make
something more compact or extremely light weight, you may have to end up paying more.

And applications such as say transportation is the place where there is lot of emphasis on
volume and weight. For example, if you had a solar converter and you wanted to install it in
your hostel and if the solar inverter was the size of almeriah, you would say no it cannot enter
my room and the solar inverter is, so big. Whereas if it is the size of may be a cell phone or a
laptop, then you do not mind having a panel outside your window because it is not taking
much space.

Similar I mean, so you want to have there is a benefit of reducing your volume and weight.
For example, when you talk of an application like, transportation say aircraft or a satellite,
size and weight is extremely important, so when you want to fly somewhere the aircraft
company that take you there will weigh your baggage. And see whether it is under 20 kilo
grams, if is smaller than 20 kilo grams then charge they will charge some extra for carrying a
heavier bag or if your bag is too big, they will not allow you to check it in.

So, they operate under tight constraints of size and volume and the implications of size and
volume say on a power converter sitting in an aircraft or a satellite is it will cause more fuel
consumption in the aircraft. So, there actually a cost impact of size and weight, which needs
to be considered when you are designing your engineering system and you can see that you
make a purchase decision based on say initial cost. But, if you forget about efficiency or
reliability or how what the size or weight is then you are not actually taking in the overall
picture.

So, you need to actually roll in the different aspects of design, including efficiency, reliability,
weight, transportation constraints, the reliability has implications on how often you do
servicing. So, many of these things have to be rolled in to a single metric and then you make
a comparison the design that you are doing is it heading in the right direction or not.

So, the metric that you would use might give more weightage to say weight and volume, may
be in an aircraft application it might give more weightage to efficiency and reliability and say
a application such as long duration industrial power systems. So, there can be different ways
of weighting things then but a methodology to actually bring this all together is important for
any engineering design.

(Refer Slide Time: 37:52)

The third part of the course is about the power electronic details will be looking at the
characteristics of power converters, the topologies, the requirement of filtering, the output of
the power converter you do not want your neighbors computer to shut down every time, your
solar panel on the roof generates power. So, you want a tight filtering and you want it to be
compatible with the things that are there around you.

So, we will look at filtering the way you model the switches what does its impact on a
common mode and a differential mode, bases. And will look at a things like how to select the
DC bus, often if you use a voltage resource converter you are talking about how to select the
DC bus, primarily the DC bus capacitors for such a power converter, how do you select the
semiconductors, this has an implication on not just the temperature of the semiconductor
junction.

But what is the cycle life of say a given package of the semiconductor, also with any power
converter, you it is not just the main power circuit components there are lot of ancillary
hardware, you have sensors, you have protection components in your power converter, you
have your controller, it might be small analog controllers or more sophisticated, digital
controllers, you have breakers and contactors that need to operate when you are running the
power converter, you need to handle surges, you might have some surge protectors, you need
to pre-charge your power circuit components. And you also need to provide control power to
your circuit boards that are there within your power converter. So, all these ancillary
components are important for proper operation of the power converter.

(Refer Slide Time: 40:06)

If you look at the control related aspects to some extent we will look at the switching and
average models, some of the important requirements are to be able to do things like phase
locking, how to controlled the current that is coming out of your distributed generation unit,
how do you control the DC bus, how do you control the real and reactive power coming out
of your power conditioning unit. So, these are issues that are control related which we will try
to address some extent.
(Refer Slide Time: 40:48)

If you look at power quality related issues, one of the first question is how do you measure
what is power quality. And will have some discussion about that and as important to make a
justification to add equipment, which is addressing power quality issue such as UPS. You also
have power quality is not just an issue from the customer side, so a customer side power
quality issue is that a power cut or an outage.

Whereas you can also have power quality from the provider side, where customer load is
drawing a lot of harmonics or a customer load is highly unbalanced. So, it is a power quality
issue which is impacting the provider; so power quality is not just one sided issue it, you have
to look at it from both sides from both the loads and the sources and ensure that overall the
system is in good shape. You also have issues like flicker, which comes up when you have a
periodic power injection in to a node, you have to deal with sags, swells, surges in a real grid.
And you also have to look at what are the ranges of frequencies that you need to operate and
how to operate within that particular range appropriately.
(Refer Slide Time: 42:27)

If you look at the issue of efficiency of a the distributed generation versus the centralized
generation, you can have a variety range of efficiency over here depending on the type of
plant, whether it is a just a boiler running a turbine or it has got different stages which can
help improve efficiency or whether it is a natural gas power plant. Where you can have
improved efficiency, you also have losses in transmission systems, losses in the distribution
system.

And then finally, at the consumption point you are talking about overall power from the fuel
to the actual energy that is delivered, you are talking about something in the range of 20 to 40
percent. If you have a generator source, which is sitting too close to your point of
consumption, you can have if you are just providing the electrical energy output to your load,
then you could have similar levels of efficiency. But, in addition to the electrical energy that
is being provided to the load, if you can make use of the waste heat in the load for some
purpose.

Then you can actually have better overall fuel to end use efficiency and such systems are
typically called co-generation systems. And the comparison of efficiency between the
centralized and the distributed is not that straight forward for example, if you have a
centralized system, because it is a very large power source. You can actually make use of
generators, with high performance components with a better cooling etc to actually improve
the efficiency at the large centralized generator making use of the economies of scale.
Whereas you may not be able to increase the cost of this generator to that extent, so you have
some advantages, because of higher performance of system and thermal management, in the
centralized method of generation. But, if you are looking at from a distance point of view,
you have definitely advantages for the distributed generation type of sources. And making use
of the waste heat at the point of consumption with will actually give you a big boost in terms
of increasing your overall efficiency.

In fact, Professor Amulya Reddy who set up the center for sustainable studies in IISc, he had
an acronym he used to call it defenders, which is Development Focused End user Oriented
and Service Directed Energy Sources. And such sources fit in more with the distributed
generation parade and the idea is that say if you are providing, say lighting solutions to a rural
household there is definitely development, that is possible more work that such a person can
do and provides a direct service. So, there are advantages to the distributed generation
paradigm, even when some of the initial cost might up to be higher. And over all the fuel to
load efficiency is an important aspect to be kept in mind when you are looking at or
comparing these systems.

(Refer Slide Time: 46:22)

Another aspect is about reliability, when you are look comparing the centralized approach
versus distributed approach, the centralized power system covers a much wider area. So, it is
a geographically more exposed to problem such as environmental or lightening, rain,
pollution etc, so you can have faults, because of a number of reasons such as this or your
distribution systems, might have, tress that are close by which cause periodic faults, you
might have squirrels and birds coming and sitting in and you have situations, you would have
poor loads on your distribution systems.

For example, on a distribution feeder you might have may be a hundred distribution
transformers and from each distribution transformer you might have many tens of loads. And
each consumption load might be household or commercial establishment and each household
you might have 10 or many tens of loads, so if you are looking at the actual load connection
to the physical distribution system it is a very large number. So, even if your load has a
failure of once in 5 years, if you have many 1000 of loads these faults are occurring all the
time.

And you need to be able to deal with that especially when it is spread around in a large
geographical manner. And, so your transmission and distribution system traditionally is more
exposed to the problems, but on the flip side you can make use of higher performance
materials and components on a larger scale you can have more sophisticated, control and
protection packages in the centralized system, where it might be justified based on economies
of scale. Whereas in DG systems, it is typically co-located with the load, which means that it
is less exposed to external disturbances.

And, so you can tune your reliability to match your load requirement, rather than make it all
very high or all very low you can actually try to match the reliability requirement of the
source with the load. And often DG people use it as a backup system, when you do not have
power from the grid, so overall if you look at from the efficiency perspective and from the
reliability perspective. What I believe is important, is not that one method of addressing
generation will solve all problems, it is a way in which you need to design your system.

Such that both centralized and distributed generation have to co-exist, so it is not that the
centralized system would go away and you would have only distributed sources or you would
never have distributed sources and everything is solved in a centralized manner you need to
have engineering architecture, where both systems can actually co-exist.
(Refer Slide Time: 49:53)

So, we looked at it from an efficiency and reliability perspective, if you look at the typical
upfront cost perspective. Scale actually favors centralized power plant, where you have larger
ratings of the centralized power plants, whereas the smaller ratings of the DG unit. So, you
have benefits of efficiency economies of scale up to some extent, once you go to very
gigantic sizes((Refer Time: 50:26)) then again you have problems, but when you talking
about typical DG and centralized systems, you are talking about benefits of efficiency of
scale for the centralized systems.

If you look at renewable DG systems like, roof top solar or combined heat and power
systems, they or of a lower rating, typically may be for a household or commercial rating. So,
you cannot use the level of sophistication or the advanced materials, you may not be able to
use hydrogen cooling for example, to cool the conductors of a residential synchronous
machine, where as you could actually easily use that; it is typically used in a large centralized
generator.

So, whether it is again, as we mentioned the DG does not have to be always in a renewable
did not always be a centralized power, I mean roof top solar it might be a small diesel
generator that you have a in your commercial establishment, which you use as a generation
source, either as a backup or it might be use for some additional functionality depending on
the type of interconnection or power conditioning device.
So, again if you look at the economics of scale issue for each distributed generation system,
you need to actually duplicate control and protection. And, so you have some impact over
there, but it is becoming less important now a days, as you have a sophisticated digital
controllers, which can handle not control of the DG unit, but also many aspects of the
protection can be integrated in to a single processor. So, you can actually have quite
sophisticated protection and control methodologies one at lower power scales with the newer
digital controllers.

If you look at efficiency from the component efficiency point of view from the machine
efficiency point of view, you have advantages from the centralized power method of
generation. Whereas the overall efficiency, especially if you are making use of waste heat you
have advantage for distributed type of solution, again when you look at the reliability from
the use of high performance materials or cooling, you have advantages to the centralized
power plant, but reliability is not just a having high up time.

But, if you look at reliability in terms of redundancy you have say for example, if one large
generator in a coal plant goes down, you lose a large amount of generation. Whereas, if one
turbine out of 250 turbines come down in a wind farm, you might use 2 mega watts out of
250 mega watts, so the redundancy actually gives you benefit in some of these methods.

(Refer Slide Time: 53:46)

So, if you look at the societal needs for distributed generation, there are many issues, which
makes energy a pressing topic it is an issue from the source side. Where you have concerns of
resource depletion, you have concerns from the sink side after you make use of the energy
how do you deal with things like carbon dioxide, which is essentially at the same climate
change issues. You also have issues of power quality- is there energy at all in the first place, I
mean this is a way big societal cost for of poor power quality or lack of electricity being
available is actually a big cost.

(Refer Slide Time: 54:31)

So, over all for this course we do not have a standard text book, we will be looking at papers
from journals, we make use of data sheets or technical reports from companies, you can make
use of your power electronics or your power system, text book that you use in your under
grad or in your courses, that you have done in your first and second semester. The first couple
of books are one by Professor David MacKay.

And the second by Amory Lovins; they are actually more general reading you do not need to
have much technical background to actually read these things; may be you can ask your
parents or relatives to also read and they may be able to comfortably read those books. So,
this smallest profitable is similar in line to small is beautiful paradigm; E. F Schumacher had
a book, small is beautiful.

So, this is the overall introduction for the course, in the next class we will look at specific
distributed generation technologies and how power electronics has an important role in those
technologies.
Thank you.

You might also like