Training D9: Modeling Wind and Solar

kind of a brief overview of wind and solar and such and then at the end I'm going to talk about the models because all the new who could probably say all all of the new wind Parker new generators you're putting in transient stability models they're all these things so I want I'm gonna get and talk mostly about the models at the end but just kind of briefly introduce what things are so we've talked about hydro we've talked about solar steam so here we're going to talk about the renewable plants I think people are finding this isn't something power world can help you with but emt piece that type studies are you guys doing those on wind farms here yet okay I know Earth has started doing them and the reason as funny as I I do presentations on this but when I introduce sub synchronous resonance I used to say and people figured that out in 1960 and we don't worry too much about that anymore they broke shafts on generators in the 60s and they figured out how coal plants could have this resonance between the electrical and mechanical as I said but we don't worry about that anymore but then her cot was having a lot of trouble probably about 5 to 10 years ago with wind turbines just tripping and what was going on is the inverter is it's just a piece of software and you tune these PA loops and everything to design your control system and they were creating resonance in that control system with the electric system and if they ran a transient stability simulation everything looked fine and it wasn't until they went after the turbines kept tripping they went in and started modeling all the emt-p stuff so that might be on your horizon someday how'd they get him to share the detailed models they're usually pretty tight about that I assume they just paid him lots of money all right so we you guys all know this wind is growing a lot this is the new installed wind each year and it's growing and this is true I mean if you look at offshore installations as well are starting to come up they're really though much of the world is adding this stuff so here's the growth in the u.s. wind cattell we made these slides in about 28 isn't as it stops that point I was like looking at these the colors are showing you like the first second third and fourth quarter and the fourth quarter is like all the big things because they're trying to get him in the ground before the tax credit might expire and they don't know for sure so I always imagined two trucks and two guys with the shovel going okay we started but but that's what's going on here so it is growing definitely growing here's cumulative Bend so this is the wind installed by state so I mean megawatts yeah you know if you look at this it's funny because you think of Texas as this oil land oil and gas but it has been wind they've invested a lot in transmission so that they can have all the wind farms out here and get them over to the cities so here's the only problem nothing you know what was it it smell I got four we gotta go through beautiful country Texas and they have one state commission to talk to yeah I'm sorry but they held all that wind back right they held it all back and then it was like almost all of it came on at one time yeah well they built the transmission then oh yeah it's my other customers at power world are our big customers are companies like you and little companies that buy and sell energy futures and credits in participating in markets so those customers actually really didn't like Texas after they built all that transmission because the prices all dropped and kind of leveled out and there was no I don't know yeah I don't know I'm gonna guess they're all type three four for sure okay I don't think that's right this is a slide you know states are driving this there are lots of state you guys know this Washington Oregon have rules Washington says you must close all your coal plants right is that happening yes yes it is a 25 yeah your gas plans also sticks and stones all right wind resources now part of part of probably what has who's funding your big project to merge all the grids who's that spp and this is probably part of the reason is because if you're gonna build a lot of wind and this is where it's at but this is not this is between the interconnects so you're gonna need to get the power of both directions if we did a huge amount of that there's Texas so again all the wind in Texas tends to be up in West Texas oh one of the funny things it takes a lot of the wind in Texas is in the eastern air connect because it's up here and all right when turbans they're getting really really big actually I think these are Tom slides and I like them because this is the Washington Monument the new turbines they're putting in that's how big they're they're bigger here's the Statue of Liberty so when they drive by you on the interstate and you're like that's like a quarter the length of a football field with a blade going by there very very big the advantage of that is they get up high enough when you're down low on the ground the wind speed varies a lot you get higher and higher up that's a little bit more consistent the other Avengers lock no birds no no no the bigger it gets the slower the turbine spins of the first key dodge it gotcha right this is not true it just breaks the bones as we see that they don't die to a week like tip speed is really fast okay bullet physics right yeah yeah no it makes sense and so it's killing birds but bags and bags of birds don't leave they do they got them up that's accountants all right I want to move on from the all right one thing to know about wind this is actually this is true in your car too but the amount of energy and wind goes up I think it says the cue of wind speed so it goes up higher and higher and at some point the blades gonna spin too fast so you're gonna basically feather the blades you're not catching so much man so they're designed to catch as much wind as possible these are some quiet sensible do they put mechanical resistance now the blades yeah I would assume the blades just rotated to kind of like not catch it you're like this no all right when firms are made up of collector systems you guys know this in a power in a power system tool like ours the tendon model this is one generator or maybe a couple and then have an equivalence of the collector system you guys don't have cows out by your wind turbines but West Texas probably yeah yeah we do we dugouts tons can do with most water okay all right another thing that's changing is a very good place to build wind turbines is out on the ocean because there's no birds that are welcome in us them and the wind is more consistent out there so as long as there aren't people with expensive homes here complaining you can put them off in the ocean that's that is starting to happen here's a picture of England and Denmark they have big wind installations out in the middle of the sea all right there are four general types of wind turbines type one wind turbines it's just an induction machine like the loads I talked about yesterday the problem with an induction machine is you have no voltage control whatsoever whatever the P output is you do a bunch of math and the Q is just a function of the voltage and keep so you have no ability to control voltage type two wind turbines add in a rotor resistance and give you some ability to control voltage it sounds like you guys have quite a lot of these they're old women against their 15 years or older so nobody's allowed to install these if either ones or twos anymore because they just don't provide enough control for that for the grid itself if they had small caps in inside to bend in the type twos even the ones I thought maybe the ones as well thanks correct they would just have external gaps too probably just to keep it so you have a power factor of waters alright Type three wind turbines this is where it's a little outside of my depth but basically they're taking a an induction motor and then taking the output of the of the stator voltage and feeding it back into the field and somehow they call this a doubly fed induction generator and it magically gives them the ability to control it which I'm guessing some of you in this room understand that better than I do but the point is this is a technology that that gives you the ability to control voltage but the actual generator itself is directly connected to the grid through an AC feedback so what you'll see in the models that I show in a little bit more in the element is that there actually is there are parts of it that are modeling the blade and modeling the generators swinging and so there's more feedback between the mechanical and the electrical system when you go to type for wind turbines at that point you have an inverter that's taking whatever electric signal you're creating on the wind farm running it through a full power electronic converter and creating an AC waveform so at that point whatever your source of energy is is completely isolated from the grid so the models actually get much simpler with the type for winter because there isn't it we're not going to model the stuff on the mechanical side at all and at that point a solar plant a battery you know any any device that's hooked up with a control converter like this in a transient stability - it's going to look the same all new wind farms are type 3 & 4 at this point another issue with wind turbines there are rules about that's actually I think these rules have been misinterpreted there are rules about when you are not allowed to trip that's what they're supposed to be and they're saying that if the voltage drops below this value for this amount of time as long as you're up here you're not allowed to do I think you guys have figured out I think some of the users and in our space we're treating as this as I have to trip if I'm on the other side of this curve so I think some people were modeling putting these curves in and just saying they must trim but they don't actually half the trip that's okay to keep operating and keep holding the grid up hang on until the fault clears them and still be there so these are these are rules about when you are not allowed to trip not about what you have to trip just try to get ahead this is just going through the okay this is a this right here is the wind speed I think is that right CP I was forgetting I always look at this and go this isn't in the model basically all of the models that we use in transit stability make the assumption that the wind turbine is catching the maximum amount of power as it can all the time these are curves related to as I change the pitch angle and as the wind speed varies how much power do I get so they think this is wind speed and this is how much power I can get and this is at what pitch angle so depending on the wind speed you end up varying the pitch angle to get the maximum amount of power and none of our models do that yeah is it as tippins's okay so it's related to wind speed it's it's a normalised wind speed depending on how long the blade is it's the ratio of a wind speed to blade length I'm kind of skipping over this because once you get to the transient stability models we're not we're not modeling any of this so what are these models look like when you're on a type for servant this is a type for wind turbine at this point we've got that converter so a lot of these turbine technologies from what I understand they may have a permanent magnet synchronous machine it could be anything the point is it's really up to the technology of the hardware folks and because it's going through a converter we just don't really care on the transmission side just creating energy somehow oh that we're gonna compare care about is the control system that gets wrapped around this and that's what we're all right solar panels this is an another big area of growth which kind of two different types of solar uses one is on people's roofs and the other is in a farm field next to a substation or nearby so it's like more do I have ten panels on my house or do I have a thousand panels out in the field somewhere and that is definitely growing a lot this is a picture of the where's where are good places to put solar panels is there a lot of it going up here north run more than you think atheria yeah it's the same vanilla that would be you're about the same as us in Illinois and that would be the answer for me too it's when the government pays for 30 to 40% of the installations but most of it when you talk to utilities it's tends to be Arizona and the Southern California folks I mean that's who's seeing it the most of it this is a picture of the whole world lots of places funny I remember talking to a customer in Abu Dhabi once about wind turbines and EXO if they're not winter it's about solar panels and the biggest problem for them was sand it's well and it covers up the two panels just like they have to go out and clean them off so it just depends where you add I mean the money it's you look at Germany Germany's were the biggest installations of their percentage-wise are and it's not even necessarily a particularly great place to put solar panels so this is a plot over 2008 to 2015 of how much solar generation there is so I mean you can see summer winter summer winter here but it is growing a lot that's still not a huge amount yeah I like this figure this is a plot from got got off a website of how much output there was from a solar panel I assumed this was like you know sunrise to sunset but the point is there are a lot of clouds going by so they're there up and down a lot so if you look at one solar panel it's gonna look very noisy the assumption is if you had thousands of them in a subdivision the clouds aren't covering all the panels all at once but gave me a good thing to point out about that image is from a written storage company oh yeah this is yeah come by our batteries is what this being said right there all right let's tributed pv i talked about yesterday of how we model that with the load yeah there it feels like there is the ability to have a lot more voltage control down at the house right now that's not being used at some point most of these panels just try to do no harm essentially but if you are starting to at least tell them hey you can't be absorbing lots of ours but they're being asked to provide both the trouble yet eventually he knows anything anywhere not that I'm I'm not aware of it is anybody aware of I don't even know exactly it'd be the communication system to like to talk to all the turbines it's like maybe Verizon could get in that business the other 5g Network exactly it's gonna solve all of our problems start supplying bars you could yeah everybody have a set point but them throughout the day that descent point if you were trying to hold it here as this just went up throughout the day you'd be absorbing bars and putting probably be futile to keep it fixed yeah this was again some of you love this more than I do but there was a fire in California and the blue cut fire and a big part of the problem with that is in the original wind or solar panels put in the manufacturers are being told you need to you know drop off line if the frequency drops to here or voltage drops the year so they installed all these panels that had all the same settings so then when an event happened and a whole bunch of generation got tripped and the frequency dropped then a huge amount of load all saw this frequency drop and all left at the same time when they didn't really need to and I think they ended up with an over frequency events in the end that familiar to people they had a big gen trip but then they lost way more load than the gen they had tripped in that was the problem it was that's where all of the momentary cessation stuff came from that the panels did not need to do that they could have stuck around but they were being told to trip so they tripped so the newer I Tripoli standards I think from 2014 have changed those rulers and basically er Tom's hang around as long as you can yep well I think they're worried about fighters on a rooftop you know they want to hit the voltage of frequencies going too high they don't want to start fires be my assumption well if if you know if your house blacks out and your solar that you know it's like you lose the grid okay and your solar keeps on working its back the high side it's like going into your house and saying hey I've got a generator let me plug it into an outlet so I can just distribute it around my house without a disconnect switch her back beading and there's somebody work on that distribution sort of a cutie okay so we've got to make sure these things rip off but if the dip is for 30 seconds the crews nothing would be out there that soon so you can stick around for a while yeah okay that makes it alright now I get to this stuff I feel like more of an expert which is how to model these things alright so what happened with these models say in about 2008 or so is it's new technology and all of the wind turbine models are coming in to software tools like ours as user written models because honestly people are just learning how to model these things and those tools and as you guys know thankfully and whack you're very resistant to that because it's kind of a nightmare for you to manage and figure out what's going wrong and there's some black box code running and you don't know what it's doing so in about 2010 or so the whack and vwg group got together and started creating what we call the first generation generic models so this up here was the first generation renewable a wind turbine model and we already had a whole software set built around machines exciters governors and stabilizers so we just tried to create a model that map to that so we have a machine I'll talk about in a second we have an exciter so your brain should be going you don't have an exciter on an electronic converter that makes no sense so in simulator we call it an exciter but it's really the electrical control for the wind a renewable plant we then had a governor which again it's not it's not there's no shaft on us or so it's not it's not it's not like a typical governor so governor really just means all the mechanical stuff and then the stabilizer on the type three was actually a pitch controller so was the thing that looked at the frequency in our would vary the pitch if frequency was going out alright the first generation the problem with these models was it was people like PSLs and me and esse and folks at every designing these models which is great we write good software but we don't know how the hardware works so we're just kind of guessing so having done this long enough I've accepted the fact that the only way to get a good model is to make a crappy model first so that's what we did we put up this type one model and then it had some ability to do control features it was out there for three or four years and all the hardware guys were like oh suck they don't work you know and they didn't work so what happened was we made the second generation models which worked much better so what was different was this is the type three mostly it's just extra control features so there's extra stuff up here that I'm going to walk through in a moment about they control the torque on a type three wind turbine and then looking at voltage bars and power on any wouldn't plant and we rename the models when we made the models in 2010 solar wasn't that big of a deal yet so we were naming the models WT and so for there was like a four year period where people would call how do i model a solar plant like oh you put in a wind turbine model which was confusing to people that was perfectly appropriate so we renamed the models into renewable energy generation converter renewable energy electrical control this stuff is only related to wind turbines so we included those down there and then this is a wind turbine torque controller but then the plant controller again is generic for all solar battery whatever I'm gonna walk through these models in just a second but type for wind turbines we're like this in the first generation and then we added the plant controller the Solar solar model you end up using these same pieces I think the folks here know this but when we made our renewable models again so the problem with having software people do stuff we don't know another artwork works very well we assumed that the solar panels were simpler so the re ECB model I'll show in a little bit is it's a simplified version of the RTE see a model where we stripped out about half the input parameters so that you don't have to deal with them and we were like oh solar panels don't need these we were wrong they need them so you shouldn't use this model you should use the a instead they're all the things related to the momentary cessation are important in the solar panel energy storage I'll show that in just a second so the software implementation of this is you ended up with three new types of models so inside simulator so when you're on a generator dialog and you go to the stability tab you have models machines exciters governor stabilizer and then you come under other and the other it's it's like just all the other models and so there's various types of other mamas there's aerodynamics there's over excitation there's a plant controller there's a P ref controller relays and and so forth so we just added in some new categories the plank controller and the P ref controller were new for the second generation yeah and you see them inside simulator when you're looking at the models and you go to other you get different folders with other models and yeah you guys are GE users so you're fine but I mean you'll see things where there's like an re PC a and then an re PC a one those are the exact same model just ones in GE and ones in pslf and Tom and I over the years have just kind of given up everything I always say every engineer knows that the tool that they use always gives them the right answer because otherwise they wouldn't be using it in there right so so we've just been lay all right we need to give them a model that's got the same name as what they're used to so even though they're identical models they have different names so that esse users see their correct Wallace and PSO pslf users see their correct all right this is a table and actually if you go into power world's website later and you go to power world.com slash web help that's our help documentation on the web if you scroll down over on the left to transient stability add-on and overview in the overview there's an entry here on renewable energy miles and it's got this table it's kind of telling you depending on the technology you have here's the models you should use and then all of these models have grown out of the wack community and there's actually links to kind of the papers from 2012 to 14 on that I actually there on our website I don't trust wek not to move it on their web websites so I just downloaded it we'll be a permanent part of our help so let me walk through these models so that you are familiar with them to see all the new generators you get are these models all right yesterday I talked about this low voltage thing and how this isn't really this is for software so the hardware people don't like it but it's just it's a artifact of how the software works all right calling this a model I always having worked with it now for ten years I laugh this is not really much of a model ok what what this thing is doing is so we've got this renewable model I'm about to talk about right there's a control system over here that's sending over a signal that says I want this much real and reactive current and that is all that this machine is getting and then coming out of it it's telling the network here is how much real and reactive current I'm giving you so this is supposed to model the power electronic converter so let's go look at that model and make fun of it all right the fact is that the power electronic stuff is going on in tens of microseconds it's like a thousand times faster than we're simulating so all our model does is it says I want this much reactive current and it says with a delay of twenty milliseconds it magically creates it it's just it's to delay blocks that does all this model is we're not modeling pulse width modulation and pulling on wave stuff that it's too fast we can't model it you have to do MTP stuff to do that so you just put in a delay block this is the same as a delay block you have an integrator with the feedback subtracting this is the equivalent to a delay block so it's just showing it's mostly drawn this way to make it easier to show there's a rate limit on it that says I can't go up faster that was certainly great so that is all this is doing there's then a little bit of a feedback that says when the voltage drops the amount of real power will drop that's what this lookup table is right here and then this right here is related to you guys have run studies I started to talk about this yesterday you see this all the time and about every probably at least every 18 months at a wek meeting we have to spend half an hour talking about this again because there's new people in the room but when you run I look at like the voltage at the terminal of a wind turbine a fault happens it goes down and you always get these spikes this spike here and that spike can go quite high alright that spike is occurring because I have a fault occur the control system is going to live in that fault and go oh I need to pull the voltage up so if you go look at the renewable model without electrical control will again in a second it's gonna push that up to the maximum desire reactive current the maximum possible reactive current and it's gonna be sitting right there and that's gonna go through a delay of 20 milliseconds or so which is about five time steps or so right that's fine that comes in here the voltage collapses this stuff does some magic to push the current down to zero because the voltage is zero that's all fine we go through the fault just fine then suddenly the fault recovers or clears and the voltage goes from something close to zero up to say 0.7 or 0.8 and now this signal here is still coming in as maximum current maximum reactive current but now the voltage is recovered and so it's going to push the voltage way high because it's asking for maximum current when the voltage is recovered and the problem with this model is weird we're not modeling every 10 microseconds and the power electronics that would go home and it's recovered I should back off now that's not in our model it's too fast so if we didn't do anything for about four or five time steps we might hang out it like one point six per unit voltage for something and it's just because it's not responding quickly and there's there's nothing to do except throw off your hands and use the MTP and PS CAD tools so yep just to kind of smooth out stuff I mean you could make it small or Bettina if you didn't have any time delay then then you'd create your own numerical yeah and there is some time delay in there yeah I mean I don't know of the control systems the converter but I'm assuming they're detecting faults and doing special stuff and they're detecting fault clearing and doing special stuff but I don't know that for sure if you had it as an instantaneous there's still a little bit of delay over in the renew even that even the Renewable even even in the electrical control they're still going to be a couple time steps where it responds right yeah so so to fix this problem again this is we have this thing that says high voltage reactive current management again why don't you really say is software fix so that we don't have these big voltage spikes and so what it says is if the voltage goes above some threshold it's gonna try to just in not insert it's going to suck extra bars out for a few time steps just to pull the voltage down to clip it to some value so if you look at a simulation when you you have the fault happen these where's voltage here is goes down like this and when the fault clears you get this spike right here you guys see it in pslf you see it all all the tools do Huyen from used to be with every totally talked to converter person once who said if you go look at an oscilloscope you'll see a spike - it's just shorter duration after a fault right it mean it makes sense that so anyway what we do is this is what I just tortured and simulator there's a parameter that will say what is the maximum voltage and so if you zoom in on that basically we're just clipping it at that voltage and we're doing that by just sucking bars out around the voltage down and it you know it happens over about 20 milliseconds 0.02 so I bring it up because I don't think this really matters that much but it is confusing to new people looking at it right you guys think it matters or you're just some smoking on it yeah it's a numerical problem with the kind of tools that rhythm so we're just trying to get into it all right another choice about a new model that we're working on is called re GCB and it's similar and then it has the to delay blocks upfront but then instead of taking these two currents and saying I want this current they're running them through an equation to create an equivalent voltage behind an impedance that's Emin the user tells us what gives us a source impedance and then we create a voltage instead in the algorithm and the hope is that this gets rid of these voltage spikes and we're testing this with every scalping us tested across software tools I'm a little pessimistic I think we're gonna get rid of the voltage spike which is probably good because people mostly look at voltage but if you look at current you're gonna see a spike there instead what people don't look at current so they won't notice it so we just traded one problem for another but I'm not sure if we just started playing with it the C model is another one we've coded but no one's using it yet all of these converters they're trying to really create a voltage that's a certain phase out of phase with the terminal voltage which means there has to be a phase lock loop trying to figure out what the phase is and usually that is very very fast and we don't include it this predates mine working in this but I think in like 2008 or so all the user written models people had included this phase lock loop that was really fast from the old guys who are not retired Tony was a huge pain and caused all kinds of problems but we're reintroducing it again so we'll see this will be another choice so these are beta models all right so let's get to the electrical control this is the electric control that you're going to see on solar and wind plants all right the first thing to do on this is to go it's not as tell yourself it's not as complicated as it looks the reason it's not is because there's a bunch of flax the Bluebell is there's a flag here a flag here a flag here and the flag down here so depending on how you set these integer Flags this model gets really simple so for instance if you were just doing I want it to stay at constant Q and just stay there people would come in here and they'd set this Q flag to zero and then all these P I loop here in the middle this complication just it's eliminated because you're going around it so and then you're just coming out the other way looking at the plant controller coming in from the green valleys so when someone gives you a model they have to make a choice as to am i controlling voltage current constant power factor you can do constant power factor over here where you're looking at P multiplying it by the initial power factor and then feeding in that I want to stay at constant power factor it's another thing you can do with this this model all these models but you're going to see in the power flow they're gonna tend to see something that looks like this where this is say 230 kv this is say 34 and a half about right through your distribution feeder networks and then this is like I don't know 460 volts or something so I guess I should say 0.46 or something down here so what you're doing is you're modeling the transform are at the point of interconnection remember that whole tree massive of wires going out to the hundreds of wind turbines that's modeled by one equivalent branch so somebody had to come up with that equivalent this is the transformer mounter if each wind turbine right here and what's going to go on with the re EC model it's going to be looking at its terminal so it's doing the local controller trying to control the constant power factor here or voltage here it's right there at each turbine that's what this is meant to represent and depending on how you set the flags in here did you get voltage or constant power factor and then up on this path this is more kind of what I think of is the emergency quick control so it's saying if the voltage suddenly drops with some gain start pushing out lots of reactive power right away that's what this stuff across the top is doing and then down along the bottom this is the real power so that it's got some reference power coming in and then again a delay in the control system before it feeds it off to the other side so right here I've got a here's one much power I want here's how much reactive power I want and then at this point you have there's a total current limit on this thing so that is modeled right here and then the user makes a choice of R and if I'm asking for too much current so I want to get rid of the reactive power or the real power first which ones more to be in this plant so you get to pick a flag that says get rid of real current or reactive burn so if I'm with my maximum current to say 1.2 per unit current and I'm asking for one per unit power and one per unit reactive power then that's going to be like 1.4 per unit total current so the question is do i back the P down or the Q down first who goes away first and that's this choice you have to make kids up as a user whoever gave you the planet so this model is really used for all the renewable plants anybody I mean I know it's ridiculous that hate and just showed you this two minutes you have musicians every 18 months at least for half an hour yeah I mean Modi's right if we get make that other type essence smaller go away quicker but it's that would then cause other problems so I mean I think what's going on is the converters have slow tens of milliseconds based control but they're also smart enough during sudden weird events like a fault or a fault clearing that there's special control going on and we just can't model it so we're this is our vain attempt of doing it and then we just kind of have to keep telling our users over and over again ignore that ignore that it's kind of like the hash marks that happen when the network equations don't solve you guys get used to going oh that was weird for a couple times steps in there but you ignore it the same thing just ignore this all right one thing with this model when when it was made we were building it for wind turbines so we're like you can't you can't absorb power in a wind turbine can't suck in power so we just hard-coded the minimum reactive power to zero and that was great right and so if somebody wanted to use this to model a battery then we're like oh crap we can't charge so our ECC it's literally the same as a except for the red stuff down here basically it says that the minimum current is the negative of the maximum so that it can charge so it can absorb power charge up the battery or flywheel or whatever I can say I lobbied hard for us not to include this stuff down here cuz I think it's silly it's it's modeling the state of charge of the battery so that you can say it starts at half of the capacity of the battery and it you know takes you know 30 your second server actually who installs a battery that only has enough storage for 30 seconds that's the whole point no one doesn't have a fleet so usually this type constant here will be like 3,600 for for an hour or something so then but it's cool you can keep track of the battery charge but you're never gonna hit the limits and a stability study I don't think but you could you can model it here if we only had yeah it had ten seconds there once it hits the min or max charge it basically makes it so you can't create the power or you can't absorb power depending on well okay if you only had a shorter period okay but even then it's like you as a user would probably go let's assume you run it for 10 seconds and then you run out of power so you just trip it that would work too okay okay well then but then you could use this for that and then the renewable environment but that's what this is for you would use this yeah they need any storage device power like okay there's also a D value one of the things that I make jokes about look up teammates in the a B and C versions of this there are four points I think I'm the BBL maybe it's five this is just a piecewise linear curve that says now as the voltage goes down the maximum current goes down or something like that there's like some you know I max for real power versus voltage so the user can put in a piecewise linear curve for that all right so they have about five points on here the hardware manufacturers are like we can't get by with less than ten so the D version that we're working on now is the red parts are new but the big part is that there's ten points here the problem for you as a user and me as a person who supports you is that's ten piecewise linear points on two curves that's forty input parameters for you to be puzzled by but then they're all right there so you can model this in greater precision this is modeling the interaction between the turbines on a type three wind turbine I mentioned that there's feedback between the electric and mechanical system so this model will lets you put in how much inertia is there on the actual machine itself that's on that's in the part of the blades and then how much inertia is there on the blades themselves and a model the blade system is one inertia and then you can have it kind of oscillate back and forth and see that if you include this model it's never clear to me how important it is to meet you you have any is this it's like so the mechanical people can see that it's oscillating it I think it's good the data the kids put in it yeah because I think in the end it doesn't really affect the transmission system that much so but it's in there anyway that's the pitch control this is for again for type three wind turbines where it's looking at the road the speed and the power it might be changing the angle of the pitch on a type three wind turbine to prevent it from going over speed and braking and then this model had some Max and min angle limits on these different parts of the block diagram and people always want more so the B version of this it's identical except we've got different limits all over the place this is just more input parameters for people to mess up this kind of there this is a torque controller that's on a type three wind turbine want to just finish up and get here to the renewable plant controller so everything I've been talking about so far so we had a re GC a we had electrical control there were some various torque control but there's some other models here and finally we have a model called re pca alright all of these models are looking local they're looking at the local control of the river this is the plant controller so this is the guy that's going to be looking out at the point of interconnection so this is the guy where you have to say I am regulating at the plant level now let's say it's regulating this bus and I am regulating the power flowing in here and the reactive power and also current matters in there too so that's part of the input parameters for this model do you have to give it a branch and a voltage bus which is one of the terminals of the branch where the plant is measuring voltage and power and then has to be sending with telemetry that data back to the all of the individual Internet's so this is what the re PCA is looking out there and then this guy it's got some as Tom set some phase compensation found a lead leg block it's got a PII loop it's got a PII loop down here in the feed so this is kind of the control system model for it and then everything else is what is it measuring so you've got these flags if I said if I wanted to regulate two constant there if I wouldn't if I wanted to try to keep the reactive power out here constant then I would come in here and I would set this breath flag to zero to send the signal down this way so I'm coming in here and now I go this way and I'm going to be measuring the reactive power on that branch comparing it to where it starts and running it through a dead band so if I move just a little maybe I don't respond there's an error min/max run it through a key I loop and it's going to be driving this value to zero trying to bring it back to where it started so that be for constant Q if I wanted to be regulating the voltage right here but I wanted to kind of regulate back behind it with that's like droop control where you're not regulating out into the system but back behind in particular I might do that if I had a second wind turbine that a second wind turbine at the same point of interconnection if both of these guys try to regulate that voltage together they're going to play with each other so if I set it up to regulate a little bit of an impedance behind it so that instead they're regulating here and here then they're not going to fight with each other so in order to model that I set this left leg to one I come this way and if I want to do reactive current droop then I set the V comm flag to zero and I go this way and so now I'm regulating that voltage Plus No - it's an error on the block diagram the pluses and minuses aren't there I'm pretty sure pitch to me voltage minus cube times AC so that as the reactive power goes up the voltage setpoint goes down so you're gonna be regulating kind of against a curve like this where as the voltage goes up I want the cue to go down and gives you a little bit of a dream control but I'll talk about that this afternoon even in the power flow so this is just all depends I want the wind turbines doing one of the things I don't understand maybe you guys do this is the control strategy of a wind turbine change like seasonally okay good I always worried and the part along the bottom this is for the real power control most things I've seen in any commercial case this freak flag is set to zero and there's no power control is that your experience of anybody doing that yet so eventually you could have even a wind turbine and it can't obviously you're not going to operate a wind turbine like ten percent lower then its maximum power like all the time just in case because that's very wasteful but certainly if the frequency goes high the wind turbine could respond and drop right now none of them do in your simulations if you wanted them to or if you're thinking about asking them to essentially you'd come in and set this freak flag to one and then you'd come down here and tune this is the droop going down or the droop going up so the droop going up you'd probably set to zero because you can't go up the power can't go up but the droop down you can set to something and then it would provide the governor type control with the winter so that's in the bottle although no one uses a breakdown alright I'm gonna type a whole presentation on that later last thing to bring up what if I had if I had this situation right here where I have two different wind farms that all kind of come in to the same substation it might make sense to just do drew control that's completely separate from one another that works fine what if I wanted to not do that though what if I wanted to regulate this power instead if I'm gonna do that I cannot have these do this independently because they will fight with each other so I need to put in a special new model called re PCB or you have one control objective over here but your outputs are going to more than one generator and in simulator we have re PCB where we let you go up to 50 generators and we have re PCB 100 because the customer banked because they had 60 they want to do but I don't think anybody's doing this yet but over time I think it will happen because you'll have a wind farm with type 3 wind turbines and then 15 years later they'll come in and add some new type for wind turbines that are all controlled together and then this model is going to be heated do you guys have any systems like this right now you do okay it's like could be an SDC also being controlled so do you use this model yet er no yeah okay all right I'm gonna hand it over to you well I think they have any other questions about wind turbines like Jones Canyon that'll be an example we have all the wheat fields and couple Springs and junipers Jones Canyon a theater type to went tripping so I think they're both ports you been

Channel: PowerWorldCorp

Views: 834

Keywords: software training, PowerWorld Simulator, power flow

Id: NjS04LMd9aU

Length: 63min 5sec (3785 seconds)

Published: Tue Mar 10 2020