 Welcome to class 17 on topics in power electronics and distributed generation. In the last class, we were discussing about the need for faster switching when you connect a DG to the grid. Then, we discussed methods for having semiconductor base switches. We talked about a CR based transfer schemes and then, this is an example of a static circuit breaker. What we saw is that with semiconductor based switch, you can have faster switching and the number of cycles can be more compared to a electromechanical switch. However, the power loss will be more and the electrical isolation capability of a semiconductor based switch is not as high as a physical open air gap or an open gap contact. So, electrically it is considered less rugged and it is more expensive. Today, what we will do is, we looked at the physical switch. Today, we will look at the smarts that go behind the operation of a switch, which would essentially be a relay or something that commands the switch to open or close. For DG application, this would typically be a relay which commands the switch to either connect or to disconnect. The operation of the switch would be controlled by external device. If you look at relays, relays have been traditionally used as protective devices for protection of a variety of power system equipment. You are protecting for a variety of faults, different conditions. In a traditional power system, you have been using relays for generator protection, transmission protection at the substation level for distribution systems. In a distribution system, you would have protection at the substation level. At the substation level, you could have large loads where the cost of the load may be quite significant. So, you might have special relays for load protection like large machines, transformers, what large equipment and certainly a valuable device. If you connect a large distributed generator source, you would then think of protecting it with a relay. So, you would have the distributed generation or the DG interconnection protection. So, if you look at the attributes of a protective device, you are looking at protective device that operates at high speed. So, it is capable of making decisions rapidly. You would also like to have selectivity in the sense that if you have a relay which is making a decision of whether to trip or whether not to trip, it should trip where it is necessary. For example, if you have a fault in the zone of protection, it has to trip where as if the fault is out of the zone, it should not trip. So, you should also not trip for under voltage when you are having an over current. So, it should be tripping for the right reason. So, it has to be selective in making the decision of whether to trip or not. And typically when you want decisions at high speed, you might be more prone to noise, you might be more prone to making the wrong decision. So, tradeoff curve might be that if your speed is very fast, you might be having poor selectivity or if you are having more time to make the right decision, then your speed is naturally less. So, a tradeoff might be along a direction such as this and your ideal relay is should be capable of operating at high speed and high selectivity. So, ideal relay would be something which is high speed and high selective, highly selective, but the tradeoff in your algorithms would be between your speed and selectivity. Another tradeoff would be in terms of your reliability, you would like your relay to be extremely reliable, it should be fully protected against damage, it should be fully protected against malfunction, poor operation and to add protective systems, to add redundant circuits, etcetera, you would incur more cost. So, many times when you want to have very high reliability, your cost would be high. So, if you want to have very high reliability, your cost would be high and if you try to eliminate the cost a lot, you might end up with poor reliability. So, you might have something very low cost and not reliable at all. So, your tradeoff curve might be along a line like this and in this situation your ideal relay should be low cost, but highly reliable. So, your ideal situation would be somewhere over here. So, having an actual relay is a tradeoff between multiple requirements, you may not be able to have everything being met at once, you might end up having a more expensive relay, but it is meeting your requirement, you try to make it cheaper, you may have to trade it off with certain other aspects. If you look at what you need to protect in a DG application, where you are bringing in a DG to your system. So, in DG applications, so one thing you would definitely want is to whatever loads are there in your facility should be protected. The second equipment that you would like to be protected is a DG itself, the distributed generator source. So, these are things within the facility, but you also so that you could have equipment which say you could operate your DG and potentially damage maybe your neighbor's equipment, because of things like out of phase reclosing. So, and you want to protect the distribution and system itself, because in a situation of unintentional islanding, we saw that the operation of the DG can potentially damage your actual distribution feeder itself. So, your protection is not just behind your point of common coupling, it is also upstream of the point of common coupling and when you are incorporating a source, you need to actually ensure that the protection is being adequately accomplished. And we saw when we are discussing protection, we need to identify zones of protection, you would have overlapping protection between zones and potentially backup protection in case something feels is there adequate backup. And you once you have protective relays, you need to think about what are its settings, how do you what sort of faults are you trying to protect against. So, those aspects need to be decided. If you look at the evolution of relays, what has been around within the system for a while, the earliest relay would be your electromechanical type of relay, where you have say currents applying torques on discs and then you would have springs and damping of the discs to determine whether your relay would trip. So, you would have individual electromechanical devices with moving parts and each such device would accomplish some functionality. So, you might have over current relays being one mechanical package, you might have imbalance relays being another package, reverse power flow relay being another package. So, you would have electromechanical packages, one for each type of relay functionality. And if you look at what happened next, people then realize that the electromechanical relays follow some particular dynamical function. So, it is trying to accomplish some differential equation and if you have some equation that can be modeled, you could also implement it within analog domain with things like operational amplifier circuits. So, the next after the electromechanical relay were there came analog relays followed by digital and multifunctional. So, if you look at these three, these are actually solid state compared to the electromechanical relay which is having moving parts within the relay package. Then after people started looking at analog circuits for relays, then by the 80s, the late 80s you had microprocessors and people started having microprocessor braze relays. So, whatever was being written as a equation in a op amp could now be implemented in a processor. So, you had digital processors doing the calculation and that was the start of a relay where you could actually program the settings. You could have programmable relays and by off late the processors have become more and more powerful. So, you do not need multiple processors each for each one relay functionality. You could have many functions being accomplished within the single processor. So, you now have multifunctional relays and these are programmable relays. They are also called IEDs in intelligent electronic devices where you could actually do a lot of sophisticated programming and implement lot of sophisticated functions. If you look at the relay types, ANSI has provided a list of relay numbers. For example, 52 would be an ANSI number for AC circuit breaker. So, this 52 is an AC circuit breaker. You would have instantaneous overcurrent relays that would correspond to ANSI number of 50. 51 would correspond to time overcurrent. You could have suffix for these numbers. For example, 51G would mean a ground time overcurrent or 51N would mean a neutral time overcurrent. 51V would mean a voltage restrained time overcurrent. So, you could have different combinations such as that. Then, you can have O voltage. You can have instantaneous or time over voltage. You can have frequency relays over frequency, under frequency. You can have relays which are not just for opening. You could have relays which guide closing of breakers. For example, if you have two sources and if you are able to have its amplitude to be aligned, its phase to be aligned, its frequency to be the same. Then, you could close a breaker without causing significant transience and that is what a synchronization check relay would do. You could have synchronizers which explicitly give the command to change your frequency setting or your voltage amplitude setting. Say, suppose it is to a machine. It might give commands to a governor or a exciter to change the operating point. So, you could have synchronizers. You could have say for example, under voltage relays again instantaneous or time. You could also have say suffix like R which means that say if you have under voltage where the voltage is so low that you consider the voltage to be 0 which means that you have a dead bus and you need to black start a dead bus. You might close a switch to energize a dead bus. So, you would have relays for function such as that. We also saw that you could have relays which operate as a function of power level. So, you calculate the power level. If it goes to high, you might be at over power condition. Also, you could see that if the power goes negative. For example, we saw an example of anti islanding function where we look at the sign of the power whether it is becoming too low. If it is going negative, then you could potentially detect an unintentional island. So, that would be a reverse power relay. You could have power quality based relay functions like 47 is phase voltage in balance. You could then the result of a phase voltage in balance would be a negative sequence, negative sequence over current. So, for example, in a machine load, it might be sensitive to overheating of negative due to negative sequence current coming in. So, you would have relays for a variety of functions. This is just a sample. There is a large list. You can get the full list in say Wikipedia. They have a list of ANSI device numbers available. So, the question is now with this how to make use of it for operating protective device. And we will look at an example where you have a simple example of a three phase circuit breaker. So, here you have say a AC circuit breaker 52. You have incoming AC bus, three phase AC bus and you want to protect some line cable or load downstream. So, you have a protective bus downstream of the circuit breaker and say you want to implement a three phase breaker with over current protection and also have a neutral over current protection. So, you might have a neutral wire for which you want to prevent over current in the neutral wire. So, this is an example of this. So, for relays you would then need to make a decision which means that you need to sense what is happening. So, here you have three CTs sensing the current that is coming through the line and then you are applying it to over current relays. So, this is 51 is a over current functionality. So, you have 51 over current for phase A, for phase B and for phase C and your sum of the currents in phase A plus B plus C would be your neutral current and then you can then look at whether there is an over current on the neutral and based on this information you have a logic which would then decide on whether to initiate a trip of the circuit breaker. So, you would have a circuit breaker where you can initiate a trip action. Similarly, you could also initiate an action to close the breaker which could say in many cases for smaller breakers you would just manually operate the breaker to close it. You could also have say circuit breakers with motorize which means that you could then give a signal to actually close the breaker rather than manually go in and close the breaker. So, the logic can be often it is expressed as ladder logic or it could just be the plain combinational logic to look at what should be the condition under which you operate the switch. So, in this example you might say a tripping of the breaker for phase A over current which is accomplished by the 51 A functionality or phase B over current. So, closing in a simple case the closing may be a manual operation and then you might say at what current level are you going to trip. So, you might have a group setting for the phase over current levels you might have a different setting for your neutral over current level. So, once you have a digital programmable platform you can have a lot of flexibility on how you want something as simple as a circuit breaker can actually implement the characteristic. And we saw in the in when we are discussing about circuit breaker protection you could have now multiple say inverse time characteristics. So, to have short long instantaneous. So, essentially you are now grouping together multiple such functionalities to implement such functions. If you look at the actual computation of whether there is over current that is happening or some action that needs to be taken in typical power protection applications you are looking at sampling rates of 8 to 16 times your normal 50 hertz rate. So, it is not as fast as what you would in a power electronic application where your switching frequency may be 10 kilo hertz 5 kilo hertz 20 kilo hertz etcetera. In a relay you are actually sampling at a at a intermediate rate to make a decision on whether to actually operate the relay or not. And often and as we discussed many times the decisions are based on at RMS time frame rather than the time frame of microseconds what you would typically associate with the semiconductor devices in a power electronic application. So, in a typical system what you would also have is you might have a DG system where you might have a control system and a protection system. And what you need to ensure in many applications is that these control systems and protection systems need to be independent in what you mean is that if you have a control failure your protection should not fail. So, when things are going bad you should be able to safely shut down you should not depend on your control being right to protect yourself. And your protection system should be able to safely shut down your overall system in case of a control failure. And if you look at a typical high power DG interconnection you would have separate protection systems for the DG you would have separate for the interconnection you would have separate protection for for you would have separate control for your DG equipment. So, for example, this is an example where you are connecting DG to a facility and you are coming through your mains your point of coupling might be just immediately downstream of your transformer. And you have the main breaker coming into the facility you might say split it off into non critical loads which might you might be able to shut off in case of failure of or poor power quality on the grid. Then you have a interconnection between now your DG and your your mains and then you would also say for example, open the breaker 50 to 3 to be able to provide a power to your critical loads in case your grid is say for example, shut down. You would also like to have protection for your DG generator itself and in case when 50 to 3 is open your interconnection is open you want this breaker to also protect your critical bus. So, this would be your critical bus you might have one critical load or multiple critical loads connected to your bus critical bus and you need to ensure that you have your DG and load protection being implemented again in appropriate manner. And that again should be implement independent of DG control in the DG control you are trying to ensure the right power level, the right voltage, the right frequency etcetera for the control of the DG. So, you might have say governors if it is a power electronic converter you might have PWM control, you might have current loops DC bus voltage control. So, those issues would be independent ideally should be independent of the interconnection and the generator and the load bus protection. So, then you could ask what are the objectives of these protective systems. So, first if you look at say this generator and the critical load bus protection what would its objective be. So, you would like to have the standard over current protection for the DG and the critical bus. You may want to ensure power level and balance protection when those corresponding parameters are out of range. Then you might also have say resynchronization say for example, you might have a situation where the DG for some reason it might be in service it might be disconnected and your power is being fed through the grid to your loads. So, once you want to reconnect this back you might be able to operate your circuit breaker to resynchronize this machine back to this bus. So, when your frequency voltage amplitude phase is right you connect back and resume operation back with the DG. So, that could be one of the requirements. So, after a shutdown resynchronization might be important and the important aspect might be if you are if for some reason say your system has gone down due to a blackout and if your breaker say 52 3 is open then if you are able to start the DG after starting the DG you might be able to close 50 to 4 to give a black start for your critical loads. So, you might have critical load does dead bus restoration. So, many such functionalities would be the governed by that particular interconnecting device then you could also say ask what would be the objectives of operating this interconnection breaker. So, if you look at the objectives of that. So, first is again the standard over current to protect against over current for loads transformer and even upstream towards the feeder. So, you want that particular breaker to ensure that now you have two sources. So, you could have the grid causing over current damage downstream into the facility or potentially the DG causing over current damage upstream out of the facility. You might also want to prevent over voltage on the feeder. So, we saw that in situations such as unbalance falls you can have the DG providing causing voltage on phases of the feeder. So, you would want your interconnection protection to actually prevent that. So, another important aspect is to prevent unintentional islands. So, another requirement would be for your for this interconnection breaker to open rapidly when you sense poor power quality on the grid. So, if your DG is working and you have sense that the grid power quality has become poor then if you rapidly open the interconnection then your critical loads will not see the poor power quality that is coming from the grid. So, outage is one possibility you can have other power quality requirements such as voltage amplitude say just sag as well as frequency unbalance potentially harmonics etcetera and you want to protect your critical loads from facing poor power quality from the source. You can also use the interconnection breaker for resynchronization. So, if you have for example, situation where there is a outage on the grid and your 52 3 has opened and your DG is now providing power to your critical load. Now, the grid has come back and you want to see when your grid voltage and the critical bus voltage matches in terms of amplitude frequency and phase then you can actually close this particular breaker and resynchronize to the grid. So, resynchronization would also be required at the interconnection protection also you might have requirements for say reenergizing a dead bus. So, for example, you might have a situation where the whole system is dead and may be your DG is also disconnected. And if for some reason this particular breaker is open you need to ensure that you are monitoring the voltages and ensuring that it is a dead bus. If everything is dead you can reenergize your critical loads as long as you are ensuring that you are not reenergizing a connected DG you are not reenergizing a dead DG. So, you do not want to start up a large machine or inverter when it is still connected and when it is not supposed to function you want to prevent that, but you want to actually reenergize a dead bus. So, dead bus restarting with the appropriate logic would also be required also you might have some additional functionalities depending on the type of sensing that can be done you might be able to monitor the amount of power that is now flowing out you might be able to look at the real and reactive power that is going out into the PCC. So, you might have some additional functionalities which could provide additional services to the facility by looking at not just the variables at the interconnection, but additional variables at different locations on the system. So, the next thing that you can ask is now if you want to do all this functionality what all information would you need to measure in terms of what all sensing elements would be required for such protective operation. And typically in a power system application for sensing voltages you use potential transformers PTs and for sensing currents you would use CTs and you would then be able to measure your voltage on a normalized basis in acceptable range you might have 110 volts secondary Pt or you might have 5 amp CT where now the actual information that is coming into the relay would be on a normalized scale rather than on a wider actual physical range. And so you need to have the adequate number of PTs and CTs at different locations. So, for example over here you might have voltage measurements at the high voltage side which means that you would need something that measures the this at the distribution voltage level. Whereas, over here if you are measuring the voltage it would be at the consumption voltage level. So, depending on where you are applying you would need different types of PTs and CTs and often in a power system application you would also differentiate between whether CTs protection grade or relay grade or whether it is metering grade. So, relay grade CT would be capable of measuring large over currents its accuracy may not be as high, but you are capable of measuring large fault current levels. Whereas, a metering CT would be having a much higher accuracy because billing is tied to it, but it is not capable of having large over currents. So, it is it will not be used for a measuring say fault current levels. So, the details of PTs CTs are acceptable or can be obtained from manufacturer websites of the these sensing devices. So, in this particular case you would have say for example you might be making use of voltage sensing. So, you might be sensing your voltage on your high voltage side to see whether you are having say neutral shifts or neutral shifts because of say unbalance faults etcetera. When you have something like a delta y transformer it does not pass zero sequence across. So, you might have to sense your voltage appropriately to see whether something has happened on the high voltage side. You might have secondary side voltage measurement you might also have secondary side current measurement. So, with the voltage measurement you will be able to look at power quality whether your voltage is in range frequency is in range whether there is distortions etcetera. Once you have voltage and power you can then measure power whether there is power is flowing in or back out into the system with the current you can now make use of that sense current for over current protection. Then if you are having voltage that is now measured on your incoming voltage plus also on your critical bus then you can make use of these two voltage measurements to do synchronization of your interconnection. Similarly, if you have voltages being measured at your critical bus and voltage being measured on your DG output that can be used to synchronize and see whether you can actually close your DG or your load protection breaker. So, you could also then say for example, measure your power level if you have voltage and current you could then see whether your DG is giving a over current situation to for say protection of faults on the load bus. You could also see whether DG is actually supposed to output power due to a fault is a power coming back into the DG. So, you could ask a variety of such questions. So, sensing is an important aspect for the overall protection of your system. So, here also you might have interconnection between your generator protection and your interconnection protection. So, for example, when you want to reenergize a dead bus you may want to know whether the status of this breaker is open and you do not want to energize a DG when it is a dead DG when it is still connected. So, you might have some status information going back and forth between your protective elements, but you would typically implement your protection such that even if for example, some part of the system is non-functional you would always end up in a safe mode rather than in a mode where you can potentially have damage. Then so the next thing that you could say is in such a system what could be the potential zones of protection. So, when we talked about the distribution system we are talking about say you could have a substation protection where you might have relays which protect the substation transformers and the equipment at the substation. You could have feeder protection where you are trying to protect the components whatever is there on the feeder. So, if you have now a DG that is connected in a manner such as this then you would say what is your interconnection protection doing what is its zone. So, interconnection protection might say protect downstream of the breaker to prevent over current from your grid flowing into your load. So, you might have the zone of the interconnection protection, but your interconnection protection is also trying to protect your distribution system and also your neighboring equipment. So, your interconnection protection has to protect a fairly large zone. So, when the DG is present and when your grid is present if say for example your 52 1 opens you need to ensure that your DG is still able to detect such a situation and open your interconnection in response to such a situation. And if the DG is absent and the grid is present you want to make sure make use of the interconnection protection to protect up to the next protective device which would be in this case 52 4 or 52 6. So, you could actually determine what would be the interconnection protection. So, this would be for example the substation protection zone. So, what is shown over here is the feeder protection what is shown in pink is the interconnection protection. And you want to ensure that these devices are able to accomplish the required level of protection. So, if you look at the other CBs over here we had this particular CB would protect say the zone downstream over here this circuit breaker might protect the zone downstream over here. So, when you have your overall system you want to ensure that every part of the system is protected you do not have zones that are left out unprotected. You also want to have now in a situation such as this that you might have a situation where you now have may be just the DG or may be just the grid or combination when both DG and grid are operating and in all those combined situations you do not have any possibility of some zone being left unprotected in all these cases. So, the next thing is we will see now that you have identified what could be the zones of protection and what could be considered interconnection and say critical load protection. Then you could consider what would be could potentially be some of the relaying functionality that could be used to accomplish some of the protection. And here what we have we are looking at is say an example of a DG that is interconnected through a delta y transformer. And you might have a facility where you have non-critical loads which can be turned off and critical loads which needs support in terms of power quality plus you might be ensuring that you are trying to compensate and prevent say peak loading over here you might be trying to pump power. So, that your demand on your facility does not exceed some particular power level because many times utility will charge if your demand is higher. So, you might have a variety of such requirements and then you would ask what sort of relaying would be required to accomplish such objective. So, we will discuss this in the next class will. So, we have seen that it is not just the switching functionality, but also the protection that goes behind the smarts that go behind such functionality that is important. So, if you look at the overall protection diagram it might look complicated, but if you look at it as one function by one by one it is actually a relatively simple and straightforward to actually see what type of protection would be required, what would be required to trip and what would be required to close a circuit breaker. And even though it looks like you are having a large number of such functions in a modern multifunctional protective relay all these functionalities will be implemented in a single package. So, in terms of hardware it does not increase amount of hardware that you would need to have to actually implement all these functionality of course you need the sensors without the sensed information you cannot make a decision, but once the sensor information is available then it is just the smarts or the algorithms that you implement to actually see whether this your objectives which we discuss can be accomplished. Thank you.