 Welcome to the third class in this course. In the last class, we are talking about DG technology trends, especially related to power electronic applications and the content of power electronics in DG technology. We are talking about internal combustion engines, combined heat and power, then micro turbines that was where we left. Today, we will complete that particular discussion we had. We will look at fuel cells and energy storage elements and take it from there. If you look at these systems in general, both the sources such as what we discussed in the last class and also storage elements are important, especially in distributed generation, where both the load and the source can potentially be intermittent and not just that many sources do not have very high bandwidth to track the instantaneous power requirement by loads. So, the storage part of it is also important aspect of distributed generation. If you look at fuel cells, the fuel input for the cell stack is typically hydrogen and at times it could be other gases such as methane. There are variety of fuel cells depending on the material technology that goes in. You have solid oxide fuel cells that operate at high temperature. You have phosphoric acid, these fuel cells, then proton exchange membrane which operates at comparatively lower temperature. For example, people are looking at PEM variety for transportation type of application, where the temperature is not very high. Whereas, for stationary applications, having something hot can be more easily handled in a stationary system. So, people have looked at the higher temperatures to get the benefits of the systems such as solid oxide. If you look at the early applications of fuel cells, they have been used in space applications for generating power and there the cost was not that much of a constraint in applications such as space. But, when you bring it to the distributed generation or transportation type of application, then cost is extremely critical and as we discussed in a earlier class, the life time of power generation equipment has to be quite long if you want to compete with existing systems. So, people are developing fuel cells for stationary power application, for automotive and also fuel cells for mobile type of application where the amount of power that would be available on a fuel cell based mobile phone or a laptop would be much longer than what you could have from a battery based mobile computing platform. If you look at the fuel cell itself, the individual cells are low voltage and they need to be stacked to get the higher voltage. So, similar to what you would do in a battery, the fuel cell today is in a transportation application is not cost competitive with the IC engine technology, but people are working on issues related to fuel cells. If you look at the fuel cell, you can think of it as a evolution of a energy storage system. If you look at a primary cell, it is a electrochemical energy storage system that can be used once. So, when it is dead, you throw out the primary cell battery and you buy and replace it with a new cell. If you look at a secondary cell, it is a electrochemical storage device that can be recharged. So, your rechargeable battery is a secondary cell, but use once and throw is a primary cell. If you look at a fuel cell, it is a electrochemical energy converter with fuel as a feed rather than having to electrically recharge. If you think of hydrogen, it is not a primary energy source in the sense that hydrogen is not freely available as a element. So, you have to generate hydrogen out of something and you would have to do, you could do say electrolysis of water or you could reform natural gas to get hydrogen. If you look at the energy storage, the fundamental element of it, if you look at what is there in something like a capacitor, the electron is a unit of energy which carries the energy in a capacitor. If you think about a fuel cell, you could think of a proton being the carrier of the energy. It is a proton flowing through the membrane, through the cell which actually generates the net electrical power in the fuel cell stack. Also you can think of the hydrogen then to be the stored form of the energy carrier which is the proton. Also the hydrogen does not occur in the natural form. So, it has to be converted from other elements. So, these are things that become important. How efficient is each part of this conversion process to actually get your useful electricity or useful energy out of such a fuel cell. If you look at the common energy storage components today, electrical energy storage at the scale of distributed generation level, you have batteries. The most common application of a battery is a UPS system and you can have a variety of batteries depending on the chemistry. You have lead acid which has been around for a long time. You have the flooded lead acid cells. You have the maintenance, the batteries which need less maintenance like the valve regulated cells. You have tubular batteries etcetera. Those are all lead acid based chemistry. The initial electric vehicles had you looked at nickel metal hydride as an option because it has higher cycle life compared to the lead acid cells and people of late are focusing on the lithium ion technology. Other ways of storing energy at a distribution level could be flywheels. People have looked at flywheels, ultra capacitors. Again, if you have a system where you could produce hydrogen and consume hydrogen, you could think of the fuel cell as a energy storage element where the storage capacity depends on your storage capacity of hydrogen. People have also looked at other methods like compressed air etcetera and for all these systems you would require power electronics to interface the storage element with the AC litric grid. If you look at specifically the battery technology, there are many challenges in batteries. The lead acid is a mature technology today and cost wise it is quite competitive compared to the other challenging battery technologies. However, there are challenges of how to estimate the state of charge. For example, if you are using it in a vehicle application, you would like to know when your fuel gauge is empty or whether it is full. You would also like to know how much charge or how much discharging can be done from the battery. It depends on aspects like the state of charge, your temperature, what is the expected cycle life that you are targeting etcetera. So, that can be a challenge in a battery. If you have a typical application, you might not have single cells, but you might have a bank of batteries and cell balancing becomes an important issue because you do not want to be constrained by the weakest cell in the overall bank. You want to actually make the maximum use of your cells in the bank and so cell balancing becomes an important issue. If you look at cycle life, cycle life is a challenge in a battery especially if you are looking at lead acid type of batteries, you are looking at hundreds of cycles. In something like lithium ion, you might be able to get 1000 or few 1000 range, but the cycle life is something that is important in a battery application. Also voltage matching is important. If you look at typical 230 volts or 415 volt AC systems without a transformer, if you want to connect a battery to the AC grid, you will need a fairly large number of cells to be connected in series. So, the method to get the voltage matching between your AC and DC side is important concern and of course the inverter for generating the AC output. If you look at the emerging battery technology such as lithium ion, it is having a much higher initial cost compared to the lead acid cells that are available today. If you look at fly wheels today, fly wheels are machines that operate at fairly high speeds. If you look at the amount of energy stored in a flywheel, it is proportional to the square of its speed. So, you typically try to apply as large an RPM as possible. So, people use it in tens of thousands of RPM to hundreds of thousands of RPM. Recent papers and journal of mechanical engineering journals show that people have done prototype flywheels that go even up to a million RPM. So, the technology required to actually hold a mechanical structure without falling apart at those high RPM is actually a big challenge. So, the mechanical consideration is a big issue. So, if you look at the energy stored, just increasing the inertia of the flywheel will give you a linear dependence with the inertia. Whereas, if you look at the RPM, it is quadratic. So, pushing the RPM can actually have benefits. And if you look at the power electronics required for controlling a flywheel system, it is essentially a regenerative motor drive. So, when you are sending power out to the machine, you are actually driving the machine and when you want to take power out, you are actually regenerating and taking power from your stored energy in the inertia and putting it back into your electrical system. If you look at the ultra capacitor as a energy storage element, people are looking at ultra capacitor closely today, because its ability to charge and discharge is much higher than a battery. So, people are looking at possible combinations of ultra capacitors and batteries to be able to rapidly charge and discharge energy. So, compared to a regular capacitor, these ultra capacitors can have much higher value of capacitance. So, we are talking about devices that can have 100 ferrets capacitance. So, many ferrets compared to the micro ferrets range that you would typically deal with say polypropylene capacitors or may be even mille ferrets in electrolytic capacitors, but it is not typical to have ferrets of capacitance in an ordinary capacitor. There is also a lot of work going on in the material technology, the type of porous carbon to increase the surface area to get higher capacitance. People are looking at things like nanotubes to increase the value of the capacitance. So, there is a range of material and structural properties that people are looking at from the actual ultra capacitor, the device point of view. If you are looking at it from the power electronic converter, what you would essentially need is a bidirectional DC to DC converter. So, to charge the ultra capacitor, may be you would use a buck converter to send energy from your DC voltage to your ultra capacitor and may be you would use a boost converter to send the energy from your ultra capacitor back to your DC side. So, essentially you would typically use a power electronic system as a integral component of a ultra capacitor system. So, if you look at the distributed generation and storage devices, people call them distributed resources to indicate it is both generation and storage. The power electronics form integral part of many of these systems and understanding both the source element and the grid to which it is connected is important for a successful design of the overall distributed energy resource system. So, if you look at the overall distributed energy system, you might think about three elements, the DG source, you might have your power conditioning, you might have your grid, it may be a distribution grid. If you look at these three, these are three critical elements of your overall DG system. The source, there are a variety as we just discussed right now and in the last class. If you look at the other aspect of the DG system, you are looking at the power conditioning which may be a machine or a inverter and you are looking at the distribution grid. Overall, the distribution grid and the power conditioning system falls clearly in the electrical domain. So, we could look at these two aspects more closely and try to keep it technology neutral, respect you of the type of source, what aspects of power conditioning, what aspects of the grid are critical for DG technology. And the DG source, to go into the DG source, you will have to look at the materials, the construction, fabrication process. So, the focus of our course would be in the first part, we look at the grid and then in the second part, we look at the power conditioning in terms of the inverter. So, next we look at the grid aspects now. So, if you look at a typical distribution system, you are having a wide range of components. If you look at the typical system, you might start off from the substation. Before the substation, you have the transmission system, the sub transmission system etcetera. So, what comes into the substation is the high voltage aspects. And then you would have the transformers at the substation followed by the low voltage bus. And then you would have feeders going out from your substation to your actual consumption points. So, the substation would consist of say your transformer, you might have tap changers to ensure that your feeder voltage stays constant irrespective of the voltage, which might go up and down at the transmission level. You would have the substation bus, you might have breakers at individual feeders, you might have some line voltage regulators, sometimes the line voltage regulators might be located further down in the line. So, essentially these components form the main portions of the substation. If you look at the distribution line itself, the feeder, you have the circuit breakers at the substation point, you would have the actual line and you might have branches of the line or laterals branching out in a radial form, a radial feeder is quite common. You might have capacitors for compensation of reactive power, you would have individual loads fed through distribution transformers that you would see on the streets. So, you would have the loads which might be connected through the transformers will be protected with fuses, you would have the consumption point, the loads might be at the low voltage level 230 volts. So, you might have a transformer for may be a set of homes on the street and at the individual homes, you would have meters and fuses entering into the home or if it is a commercial establishment, you might have a transformer for the commercial establishment. So, you could have a variety of combinations of the distribution feeder system. The transformers can be of different configurations, you can have different grounding, the feeders could be overhead lines, it might be bunched conductors in cables, it might be underground cables, there might be different varieties of grounding depending on your grounding philosophy at the transformers. So, there can be variations on this nominal system. If you look at the typical configuration of this system, it is radial and you would define areas of protection for such a system. So, you might say you might have a zone for protecting, so you might have a zone for say protection of your substation transformer. So, you might have a zone for say the low voltage bus at the substation, you can have zones for your individual feeders, you might have zones for your consumption which might be at the loads towards the loads or it may be towards the laterals etcetera. So, you have areas of protection that are defined for your feeder and these areas of protection are overlapping, so that you do not leave any particular point vulnerable in the system. So, every part of the system is protected and you give priority for different zones for protection. So, zone closer up towards the stem of the tree would have higher priority, something closer to the end branch and the leaves would have lower priority. So, something which is at higher priority would have more sophisticated protection systems etcetera. So, for example, you might put a high end protection relay with circuit breakers etcetera at in a higher zone, whereas you might want to look at something more cost effective, may be just a fuse and some arresters for over voltage protection at the other end of the zone of protection. So, you have priority for the zones of protection and you also need coordinated protection. So, for example, if you have fault in a particular load, you would like the closest protective device to actually open and you want to minimize the balance, the disturbance that you the fault creates to the balance of system. So, for example, if fault is there in zone four of this particular system, you would like this particular maybe breaker or fuse to open and you do not want the remaining portions of the feeder or even the adjacent feeders to get disturbed by the fact that you have a fault somewhere there. Also, you would like to have backup protection in the sense that say for example, if one device fails, you would like to have if possible the closest upstream protection device operate to actually provide a backup, so that the fault does not spread further up into the system. So, these aspects are quite important for the distribution system. . So, if you then look at what are the protective devices that can be used in such a distribution system, you have relays and circuit breakers. So, relay would be you could think of a relay as a brain behind the operation of a protective switch and the switch is the breaker. So, essentially the relay would command the switch to be on or to be off. So, you might use sophisticated relays at the substation level. You might have surge arrestors distributed through the system to prevent voltage. You have fuses especially out on the feeder and also as backup components in many of the protection devices. So, the fuses are comparatively cost effective devices for protection. You have circuit breakers. So, here I have indicated two breakers at the substation transformer. The actual transformer configuration would depend on your individual substation. You would have breakers feeding the individual feeders that radiate out from the substation. So, you would be able to say operate or disconnect or protect on a individual feeder basis. You also have devices such as reclosers, reclosers and sectionalizers. Reclosers are essentially you can think of it as a breaker which has the ability to actually close again after opening. The reason why people would like to use reclosers is that most falls are actually temporary. So, once the fault is cleared, you can actually reenergize the line. So, people consider using reclosers rather than some line maintenance person later on coming and manually turning on the line. The recloser can actually able can be can actually clear a temporary fault. If the fault is continuing, eventually the recloser will lock out. These are devices that are meant to operate with reclosers and we will see later on in the course how sectionalizers and reclosers can be used together for the protection of the feeder. If you look at what exactly constitutes a fault, a fault could be a fault could be essentially something that is not working. So, what would you think of as a fault? Say it could be a short circuit somewhere, it could be a different scenarios. A general way of describing a fault is a partial or total local failure of insulation or continuity. So, essentially you could think about electrical systems as essentially consisting of parts which conduct electricity and parts which would prevent the conduction of electricity. So, those are insulating components. So, a component which is supposed to insulate, if it fails to insulate, you could have essentially a failure in the electrical system or if a component which is expected to conduct electricity fails to conduct, you would have essentially a fault in the system. So, it is a fairly broad definition and if you look at say for example, an over voltage. A voltage can happen for a variety of reasons. A over voltage can actually lead to insulation failure. So, even an over voltage can actually lead to a fault. So, you want to prevent over voltages in systems. So, definition like this is fairly broad rather than just looking at short circuits as the only possible fault. For example, if you look at over voltage that commonly we would think of that comes to our mind would be a lightning striking a line. The interesting thing about a lightning striking the line is that say if it strikes a section of the line then you would end up having over voltage in a small section of the line or the feeder. You would have surge arrestors, you would have over voltage protection components which are situated close by. You would have if it is a exposed line you would have insulators on the electric poles and they might arc over. So, often the over voltage caused by lightning is then followed by arcing or essentially short circuit. Once you have a short circuit then essentially what people would see is actually the voltage collapsing rather than over voltage. So, in the immediate neighborhood of the lightning you might see an over voltage, but further away in the system what you would see is actually under voltage because you now have lines that are arcing over and then upstream breaker would open. So, may be on a section of the feeder you would see over voltage, but on the adjacent feeders and in the further away on the feeder what you would actually see is a under voltage. So, you will have to look at a combination of what can happen when you look at fault event on a typical distribution feeder. If you look at the nature of what happens when you have a fault say you have arcing of insulators etcetera ceramic insulators on a line. If a arcing starts on a line essentially to stop the arcing you have to actually disconnect your deenergize the line by opening an upstream breaker. So, you can think of it as a stopping fuel supply to that particular fault point. Once the arc is cleared then the insulator can recover and actually go back to the normal mode which is one of the reason why many of the faults are actually temporary faults. So, you have a system where you are feeding energy you have a fault then you stop providing energy to that particular point then the fault clears away and the system recovers then you can actually reenergize the line. Also you can have situations where may be squirrels are coming on the line or tree branch is stretching the line. If the branch say for example, burns out then it would stop causing a fault. So, the many reasons why you can have say temporary faults. If you look at continuity of a conductor what could be the situations where you would have discontinuity of conductors. We just had the monsoon season you have heavy rains say a tree faults on the line it will snap the line. You can have say a vehicle going and hitting a pole you can have heavy wind or in northern regions you would have ice depositing on lines increasing the weight causing lines to snap. So, for a variety of reasons you would also have discontinuity of loss of continuity of conductor which can actually be considered a fault and anything which prevents normal operation is something which is undecidable and you would like to actually have a system where you are actually doing what is intended which is delivering power to the loads. So, if you look at the models of the component that go into a system you want to actually calculate what is a response of a typical system with faults. So, you want to actually model the system which means that you are looking at how to model the components that we just discussed which are transformers your line you might have protective devices. So, how should we actually model these components. So, the most common model that you would be you would have already studied in your undergrad classes is the T model of the transformer. You could have variations on this depending on the type of transformer could be y delta zigzag type of windings you can have different types of grounding of the transformer. So, there may be variations on that particular transformer and but the basic model essentially consists of a T section where you would have leakage inductance of the primary you would have the winding primary winding resistance you would have leakages of the secondary you could reflect back that back to the primary side you would have your magnetizing inductance you might have core loss resistance terms and the question is what could be a typical value that you could take for such a transformer. So, if you look at the power rating of a transformer say for an as an example may be if you are looking at a transformer at a sample power rating of say 4 m V a and say your turns ratio would correspond to your actual voltage say you are talking about a 11 k V slash say 415 volt device. So, if you look at your leakage inductance L L from your primary plus reflected secondary you might be talking in terms of say 4 to 8 percent. If you look at it in terms of what this particular value is on your high voltage side you are talking about say 0.4 to 1.1 milli henry on your high voltage side on your low voltage you are talking about say 0.6 to 1.6 micro henry of inductance if you if you look at the resistance of your winding primary plus reflected secondary you are talking about 1 to may be 3 percent depending on how much copper is used in the winding. So, you are talking about say 0.3 to 0.9 ohms on your high voltage side whereas, you are talking about some like 430 to 1300 micro ohms on the low voltage side. If you look at something like your magnetizing inductance you are talking of something much greater than 10 per unit and. So, here you are talking of greater than 140 milli henry on the low voltage on the high voltage side or you are talking about 200 micro henry's on the low voltage side. You are looking at the core loss term you are talking of something which is greater than 100 per unit. So, you are talking of something greater than 3 kilo ohms on your high voltage side or 4.3 ohms on the low voltage side. So, you can see that if you take a typical parameters of such a transformer if you look at the physical values of inductance resistance etcetera you do not know when someone says I have something which is 1 milli henry you might think it is high inductance whereas, 2 micro henry's you might think it is small. But, if you say it is 8 percent it is the same irrespective of whether you are talking about high voltage or low voltage. So, it makes a lot more sense to talk in percentages or per unit compared to the actual physical unit because then you would have to look at in with the actual physical unit what the power rating is what the voltage level is etcetera. Whereas, talking in terms of percentages or on a per unit basis gives you a more intuitive feel of what typically to expect when you are dealing with a component. Similarly, if you look at the winding resistance if someone says that the winding resistance is 0.9 ohms or 4.30 micro ohms you might think that 4.30 micro ohms is really small, but it depends on the voltage level. So, if you say that the resistance is 1 percent then you might say that might be a reasonable value if someone says the resistance is 10 percent you would immediately realize that may be it is not reasonable it is too large for the winding. So, again the value in percentage gives you a better gut feel of what to expect in the system rather than the physical units. You might actually make use of the physical units in your actual design procedure, but to convey it in a more general manner the percentage or the per unit is a good way of doing it. Again if you look at say the magnetizing inductance say for example, if someone says the magnetizing inductance is just 2 per unit I would you can immediately say that transformer draws a lot of magnetizing reactive wars because that would correspond to 50 percent of its rating would be its no load wars. Whereas, if it is 10 per unit it means that the no load current is 10 percent of your rating which may be reasonable. So, again just looking at this components for example, if you look at your core resistance if it is 10 per unit it means that immediately the core losses is 10 percent of your. So, your no load losses would be 10 percent which might be immediately considered as unacceptable. If you say it is 100 per unit it means that your core loss resistance is 1 per unit which means that your core loss resistance is 1 percent when you have 100 per unit which might be more reasonable. So, again the parameters defined on a percentage basis can give you a feel for what is considered acceptable or reasonable and what can be expected on a typical system. Because often you may not have access to actual data and you want to make estimates of what to expect based on estimated information. So, if you look at the next important element in our particular system it would be say the actual line. So, for the line you would have studied in the power system course there are different models, lumped models, distributed parameter models etcetera. The distribution line is fairly short compared to long transmission lines etcetera. So, lumped parameter models are actually can often be sufficient in your analysis. If you are actually using cables rather than overhead lines then may be capacitive effects also become important rather than just a modeling it as R L sections or R L components in your distribution line. Your resistance of your conductor depends on the formula for resistance of a conductor is proportional to length inversely proportional to the area of cross section. So, depending on your ampacity of the line you would decide your cross sectional area of the conductor and based on that you can get resistance per unit length of the line. It can be of the range of talking about 100 milli ohms to half ohm per kilometer again depending on the ampacity of the conductors in a range such as that. If you look at the X by R ratios, if you look at X by R ratios of a line you are talking about a distribution line where the impedance the reactance to resistance ratio is of the order of 1 distribution system is typically much more resistive than transmission system. So, if you look at the X by R ratios of a transmission line you might get numbers such as 3, but distribution lines tend to be much more resistive. Also if you look at the X 0 by X plus this is typically of. So, this is typically around 3 for these lines. If you look at what contributes to the inductance of a line it is essentially you are looking at how much flux is length by a loop. So, if you look at the loop for the conductors of your lines you are talking about loop areas which are smaller. Whereas, if you are looking at the conductors sitting on a line and the loop area when you are thinking about the return path through earth you are looking at a much larger loop for situations such as earth faults. So, you would have higher X naught compared to your X plus in a typical system. So, many of these things you can actually link it to the physical lines to determine what are reasonable range of parameters of a typical line. Also you will have to look at the length of the line if you are looking at typical urban feeders you are looking at shorter distances you are talking about may be 2 to 4 kilometers. Whereas, in rural feeders you will have much typically much longer length. So, you are talking about something of the could be 10 kilometers or much larger in really long rural feeders. So, if you look at the power quality that you would see in a urban system you would see lower voltage drops along the line in a urban scenario. Whereas, in often in a village you met encounter that there is a the lights are not blowing brightly it is quite dim it is because the physical line is actually is situated physically. So, far away and the impedance of that entire line is causing larger voltage drops when people are drawing power. If you look at a typical distribution system there are a variety of ways of grounding you have you can have solid grounding of the transformer. So, people talk about T n networks etcetera T stands for Tera where your neutral is solidly connected to ground you can have impedance grounding people sometimes talk about I T networks where your neutral is connected to Tera through impedance. So, people look at things like you might see terms like T n I T T etcetera. So, there are various ways of grounding you can have multiple grounding. So, in the systems where may be each distribution transformer pole you have explicit grounding for your feeder and such systems are common in other parts of the world like North America Japan etcetera. So, there are variety of grounding ways for the transformer and in the course we will actually look at the impact of these different grounding approaches on the system. So, if you look at the model for a typical line we this is what we just discussed and if you look at what happens during a fault such as insulation fault or some sort of short circuit fault you can have different varieties of fault depending on your fault impedance. So, the question is what do you what number do you use for typical values of say the fault impedance in a fault scenario. So, would it to have Z f to be 1 per unit that would not be reasonable because it means that your current that it is drawing is actually just rated current. Again you have to look at the section in which you are doing the calculation and normalize to that particular section, but if you are looking at say over current you might say your device is capable of a 20 percent overload or may be a 50 percent overload, but it is not often that you might put 2 twice the rating as your overload capability. So, your fault impedance might be 0.5 per unit or it can be 0 per unit 0 for a solid fault. So, you could typically also look at range of fault impedances depending on what you would consider the fault current level in that particular system. So, if you look at the type of faults that are occurring in typical systems you would have single line to ground faults. So, in fact 80 percent of all faults people have collected statistics and they are actually single line to ground type of faults. If you look at so the single line to ground is the most common variety if you look at your solid 3 phase faults they are much more rare I mean it is typically around 5 percent of faults might be 3 phase faults, but the main benefit of 3 phase faults in your calculations is that your calculations are simple and it is something that you could readily calculate and then based on what would be the ratios for the different varieties of fault type single line to ground line to line etcetera you could look at what variations can happen around the solid 3 phase type of fault. If you in the next class what we look at is we will also look at the DG models and based on that we could then go forward and actually look at some of the issues before we start looking at what the fault current levels are in the system and how to actually do coordination and the objective of looking at coordination in this class is from the perspective of not to actually implement coordination which can the coordination can be taught in an entire course our perspective is to look at how coordination gets affected when you add a distributed generation system into your network. Thank you.