Recording and presentations: The keys behind the Aug 9th blackout in UK and how to avoid it going forward


15 October 2019

Media Partners:

Automatic transcription (It may contain errors)

Hello, everybody who’s joined this webinar on the keys behind the August 9 blackouts in the UK and hopefully how to avoid it going forward.

We’ll wait a couple of minutes before we actually start the presentation so that other people can join. So we’ll just pause for a couple of minutes.

Okay, let’s start the webinar. My name is Jonathan Walters. I’m the moderator of this webinar on the August 9 blackout in in the UK and how to avoid it going forward and I think will draw out their significance for other countries have what happened in that blackout. We have two great speakers

We have Weijie Mac from Aurora energy research, project leader. And Simon Williamson, commercial manager from Kiwi Power.

In a moment, I’ll ask Weijie to do the first presentation each of the presenters will present for for about 12 minutes or so. And then we’ll have discussion time after that. You’ll see in the tool pound screen, there’s a chat box where we would appreciate if you would introduce yourself. Now just so that we get a sense of where the webinar participants are coming from. And then there’s a q amp a box where you can ask questions as the presenters are speaking or as the questions occurred. You can also ask them later but, but right from the beginning, you can be sending questions through the

q amp a box so that the panelists and myself seeing them and when it comes to the discussion time, I will pose the questions from the q amp a box to each of the panelists. Okay, so let’s, let’s start the presentations. Now we’ll start with where she from a lower energy research who will who will present on the August 9 blackout.

Great Good morning.

Hi, good morning everyone. So for those who might not be familiar with over energy research, we are basically analytics and consulting firm based in the UK but we operate on throughout Europe and in Australia as well focused on tackling the questions or the pertinent questions of the energy transition. So what I’d like to do today to the cost of this 12 minutes or so, is to basically highlight

What has happened during the knife of August blackout, but more importantly, focused on some of the key reasons and the implications of this going into the future.


so what has happened on the ninth of August as some of you might already be aware, we had somewhat of a blackout in the UK the sentence on this was an occurrence overs, sorry, this was the kind of outcome of several events happening. So we have kind of taken this out from our national police report itself. And what you can see here is that slightly before 5pm, on there was a circuit thought. And after that are basically there was a wind farm, Hornsey that was owned by Austin that lost power, and that was shortly followed up by RW his little Bedford guess, plug this well. Now, as you can see on secondary, five primary and secondary response, kind of services kicked in, and then actually partly mitigated. The Frequency fall, but that was a point to them, but shortly after additional plans kind of went off the system from RW as well. So, essentially what has happened at roughly on 1653 hours was basically we had on demand disconnection. So, that was when a 1.1 million households or customers in the UK lost power on the implications of that was that you had hospitals losing power train stations losing power, so even though are the actual blackout or other the power was restored by about 5:37pm so slightly less than an hour the overall disruptions the real way services and hospitals etc last four hours and there is a question there which might have to be answered which was why did the black op sorry Why did the backup generators and hospitals and on train stations etc, failed? Now, um, there was seven reasons and this was taken basically from national groups report itself on of why this breakup happened. And I’ll go through each of these quite interesting quite quickly. So the first of that is essentially a lightning, several lightning strikes that happened. So this itself shows the number of lightning occurrences on between three to 4pm. And between four to 5pm. You can see that between four to 5pm there was quite a few more lightning strikes that happened. But once again on national grid did point out that this was nothing that was unusual. So nine instructs happen all the time. On some my question, why would the lightning strike result in a blackout? Well, it’s questionable and they happen all the time and a student in general, but in this case, one of the Lightning’s actually hit the circuit. And what happened there was basically our decline of systems in place of this power generators and there was some embedded smaller generators as well kicked in and to protect the power system itself on the generators decided to basically or rather the automatic system comes in this generators kicked in and shut down the plant of intent. Now on the second question here, which is rather The second reason here, which is actually more interesting, and it’s a fundamental shift in the system, it’s basically the system is operating on lower and lower inertia. So what I show here is essentially on the kind of historic generation in Great Britain, and I’ve split this out between scar synchronous generation. So cc GT is coal plants, nuclear plants, I also test the orange line. And you can see that synchronous generation has been kind of falling over time, at the expense of rising non synchronous generation. So the increase in renewables penetration to solar, to Venus cetera has resulted in more and more non synchronous generation on the system. And quite interestingly, on this ninth of August day itself, it was actually a day where regeneration was really kind of high level. And there was lots of reports and Klay off I think that one from Twitter were national agree that The kind of boasted that we did the system was running on 67% of low carbon generation. So, it was a particular day where non synchronous generation was very high on something that is kind of much higher than normal in the year. But overall, in line with the trend that we have more and more renewables coming on the system, synchronous generation is falling and non synchronous generation is increasing. Um, it should be noted as well here that when variability in itself was not the cause of the blackout on but it could have kind of exacerbated the blackout because it limited the amount of synchronous generation on the system. So, cc GTS, etc. If there was very few co plans on the system, operating etc. There wasn’t as many kind of spinning turbines that could mitigate the foreign frequency,

default risk and kind of line with what I’ve just said. And this is quite an interesting graph here on our panel system inertia and the kind of deviate And frequencies in the system. So what I have here on the top chart, I’ve taken this out from the Australian energy market operator, but what they did was basically they looked at the 50 hertz systems globally. And they looked at how are the frequency performance has been in their systems. So each of these dots shows a particular point on in January 2018 as a sample period, and it shows that basically are the systems have actually very well even though they are 50 foot systems, the systems have varied quite a bit in how much they have deviated from 50 hertz. Now, one of the reasons here could be the amount of wind and solar generation. So, that’s what I’ve plotted at the bottom of this chart research which shows that arm in countries where wind and solar generation are higher, you could see a slightly higher deviation in frequencies around 50 hertz. Now there is also a secondary point here that has to be made and our kind of job icon this in the later Which is on if you look at the UK and Australia and even Ireland these are basically Island models. So unlike Singapore etc, where they are individual countries, the UK, Australia etc are quite interconnected in Australia sense there are several states that are connected to the men, the Australian market. So there is a secondary question here of arm is being interconnected to other countries are resulting in higher frequency deviations. And a related question to that would then be can you then rely on your neighbors to provide security or supply when you need them to do so? And I’ll answer the question in the latest slide.

And lastly,

it’s a question of whether there was sufficient ancillary services. So in sufficient black, I’m sorry, backup services that were in place to kind of cut the loss empower. Our what I’ve taken up here is once again, this is from national report that shows two instances where our The largest plant in fit loss. So national grits procures ancillary services according to the kind of largest plan in fit loss at any point in time. Now, in both of this instances on the largest plant in fit loss was about 1000 megawatts. And they actually if you look at them or frequency response holding the head, it was roughly the same amount. What’s interesting here is that on on the first of July when the one gigawatt on MIMO, interconnected to Belgium trip that was one gigawatt interconnectors so there was sufficient frequency response in place to kind of cut that loss in frequency. What happened on the 19th of August, however, was that there were multiple plants that tripped. So while National Grid has procure enough capacities to cope with a single incident, a single plan in the last largest planning for the last, there wasn’t enough of frequency response in place to cook with multiple plans going off the system at the same time. So a question that would then naturally be Do we procure for a single plant? Or do we have to procure for multiple plants as we go into the future? And once again, I’ll kind of tackle this question in a later slide as well. So now that we have gone through the kind of key reasons of why this kind of backup has occurred, Allison reasons that national grid has put forth on what’s kind of more important in this discussion is actually, how do we proceed going into the future? So the I think there’s several key questions that one has to answer when we’re looking into the future. The first of which is are we procuring sufficient services? So frequency response or reserve services to ensure system security? And should this evolve as we have an additional penetration of renewables? So, obviously, with the 2050 targets that various countries have, I think that’s build up that renewables penetration would increase globally. The question is, what do we do with power systems to ensure that there is sufficient hi of backup generation when we have more and more intermittent generation coming onto the system. The second question that has to be answered is our well if we conclude that we need more and more arm capacities to be in place to backup this intermittent on generation, what technologies are best place to do this? And the third question that I would like to answer to do at least give some thought on today is our if we were an island so like UK, connected with the rest of EU or Ireland or even Australia connected with other states in Australia, arm, can we rely on the other party of the interconnection to provide for security or supply? So this three questions I would focus on in the next few minutes. And there is probably a fourth question here, which is our plan settings to sensitive so one of the reasons why clients were coming offline as I’ve mentioned earlier was basically there was some Rothkopf settings etc. So rate of change of frequency set etc that kicked into ensure that plants are listed plant operator believed would keep the plant on payoff. Reduce damages to the plant when the voltage changes on the question here is where the regulators have a role to play. It might be quite interesting to note that in Australia during the 2016 blackout there in South Australia, regulators are now looking back at the incident and trying to sue when operators because the system frequency or rather their settings were too sensitive. So there is a question here where the regulator should have a role in how sensitive a plan setting is of gender himself. GB the regulator in GB actually recently looked into our weather they had to increase the kind of broke off settings for all smaller embedded generators as well below five megawatts. So it’s something that regulators are looking at. It’s a question of whether they should be looking at it. And is there a role for regulators to play to ensure that plants have a higher tolerance to changes in Africa? So focusing on the first three questions here, first of which are, which is basically are we procuring sufficient services. So what I’ve shown here is our three things I’ve shown kind of our Aurora forecast of renewables penetration going into the future. That’s the first kind of graph here I’ve shown the system initial evolution as renewables penetration increase. And I have also shown our forecasts of what’s required in terms of response and reserve requirements to firmer this on renewables on system inertia decreases. So overall, what you can see here is that as we get towards this kind of 2015 world and UK has legislated kind of net zero in 2015 arm overall we expect renewables penetration to increase quite substantially at the expense of baseload generation, which is basically synchronous on CGG. Of course, the drop off the system, what that means for system inertia is the overall system inertia will be falling quite quickly. So if you look at where we are today in terms of GVA is about 417 gb as on by 2030. We expect this to fall to 267 GVS. What this means is that arm you need a lot more response and reserve requirements. So, lots of terminology there. But basically, it’s the same as any prime primary and secondary response markets in different countries that would be in place to ensure that they have sufficient backup requirements in place. Now, one important point to note here and in relevance to the black talk itself, was that our national grid claimed that this was an extremely rare event. We did some very simple statistical analysis that said, Well, if you have two independent trips of two power plants in every two minute window on once a day, how often would that happen? Well, once in every four years, that’s kind of the probability that we kind of

Kind of the probability that we think it would happen on that’s obviously a very rough statistical measure. So maybe it’s not as extreme as what they think it would be, or as infrequent as what they think it would be. On the second part that is quite interesting here is how much does it cost to actually firm up this services? So if we have procure two gigawatts example, instead of one gigawatt of frequency response, arm, we worked out that that actually only costs on households about two pounds more than annum. And that’s not a lot of money at all compared to an average household Bill 400 to 500 pounds. So maybe the answer here is not that expensive to kind of double your requirements. The second question here and I’m sure Seymour will touch on this a lot more given his experts on batteries is on what technologies are best place to provide for power in this conference, unforeseen circumstances. So I’ve put up an example here of the final frequency response but dynamic from frequency response, essentially one of the primary response markets in GB. And I’ve showed here two graphs, one of it on the left, and let’s focus on this first is on how much frequency response can each technology provide when they are actually running. So you can see that actually across the suite of technologies that I’ve shown here, are kind of better it is kind of providing output within 10 seconds of 30 seconds on all of these technologies are actually quite suited to provide for frequency response. So it’s not just a battery that can do it and OCGTOCCGT can do it as well. The key question or other the key nuances that this is what they can do if they were already running. Now in that knife of August blackout, what happened was this happened in a summer in GB so maybe CCGTS or co plans for not running which means that in order to kind of cook that the sudden spike in frequency of both changes, you require batteries or something that’s a lot faster response to coming and our live sim on to touch on this on this later slides. But there’s just something to bear in mind, which is if we are in the future where someone is basically filled up with lots of stolen when there may be cc GTS, which are not already running can provide for frequency response. And that’s actually something that is that you’re seeing on the right hand side of this. So the right hand chart essentially just shows from a historical perspective, what technologies have dominated the FR market options in GB, and you can see a clear trend that batteries and DSSDSF in this case honestly, but here has has been increasingly dominating this option. So the green and blue chunks here in this column charts, has basically taken over a lot of our guests and hydro plants etc. And the last question that I would like to answer here is can we rely on our neighbors to provide for security or supply of It’s a very tough question to answer. But hopefully this give one slice of the data that could potentially start the discussion. So what I’ve shown here is essentially on the kind of IFTA source interconnected to France historical flows versus on GB demand at any point in time. So what I saw the each point of this notes on what the interconnect, the flow was given different points of GBT mind. And so if it’s on the positive side of this on your exporting, and sorry, if you’re on the positive side of this GPS importing from France, and if you’re on the negative side of this, you’re exporting to France. And what you can see here I’ve taken up two key events that has happened in GB, so the beast from the east, which happened in winter 20 1718. And when the French nuclear outages happens in winter, 2016 and 17. And what you can see here is that when this two incidences occurred, GB was basically exporting significantly to France. Which means that, um, it’s hard to really say whether you could rely on your neighbors when a significant event happens and your peak demand is at the highest. So when peak demand is so when demand is high in GB, as you can see underwritten yellow dots were basically exporting power to France because the price differential will higher. Um, it’s like I said, it’s not easy to really determine whether this means that you can’t rely on your neighbors because you would need both of the countries to riches price cat, which has never happened in the history of gv yet, but it’s just something to think about. So that’s the slides I have for now. And I’m looking forward to the discussion shortly.

Great, thank you. Thank you very much for that presentation is very rich in data and rich in the conceptual Insights. Before I assignment to make his presentation, can I just remind everybody, you can send in questions from The q&a box, thank you to those who’ve already sent them in, but please, you know, continue, continue sending so that the panelists can respond to them when the presentations are finished. I think one question I’m going to ask both panelists just to give them a sort of warning us get. I’m the kind of person who likes to really simplify complicated things. So, Mike, my question is, is going to be isn’t it the case that there just weren’t enough batteries on the system in the in the UK to provide the kind of response that was needed, and that there’s something wrong with the, the incentives whether it’s the the planners hadn’t adjusted enough to how much intermittent renewables are on this on the system that the market rules will not support? sufficient to call forth the necessary investment in batteries or not with the right kind of lead time and so on. So that’s my question is there is there something wrong in the market or, or the planning system or both. That meant that there just weren’t enough batteries on the system to cope with the increasing penetration rate of intermittent renewables, which is a really important question for just about every country because everywhere is trying to take advantage of the fact that at a project level, intermittent renewables, wind and solar are becoming very, very cheap. So everybody wants to use them for purely economic reasons, as well as in some cases environmental reasons. But you reach a you reach a limit unless you’re doing the storage, particularly the fast response storage from battery He said at the same time, so you have to make sure that you have the market rules and the regulations in place that, that make that happen. And it seems like in the UK, something wasn’t done something went wrong in terms of getting enough batteries on the, on the system. So that even though there’s actually only a small amount of, you know, megawatts of generation capacity that that went offline, I mean, to modestly sized power plants in a, in a very large system really shouldn’t cause the whole system to, to black out. So that’s my question what what, what would what would bring more batteries onto the, onto the system? So let me ask Simon to introduce himself and to make this presentation.

Yeah, absolutely. Thanks very much, Jonathan. And I’m a very wide ranging but a very worth question I think absolutely to be asking, not just for GB not just for for Europe, but absolutely everywhere, as you correctly identified. And I’ll get into some of the details around where we think that the the market rules and regulations, for instance, are at the present time during during my presentation, but in brief, I think that we have what is still a somewhat rare event that occurred, where National Grid performant all of its duties, you know, correctly in accordance with the term rules. I think where we’re at this is a position where maybe those rules aren’t quite right. Maybe those metrics are the metrics that we’re looking at. aren’t quite right as well. Unfortunately, I think that there is there’s work on going and all of these directions What would you like me to jump into my presentation Johnson this summer or shoot over to Mike for for response to Your questions.

Now why don’t you jump into your presentation and then we’ll take responses all together.

Wonderful. Hopefully everyone can can see my slides and thanks very much for for dialing in just a very brief word by myself and by QENC Williamson, commercial manager for energy storage here, a key part which is principally known as as an aggregator. So provider really have a route to market to all sorts of distributed energy assets, including energy storage and batteries. So we rely on our technology which were very technology focused company to enable to provide the hardware and the software to optimize assets in the markets such as we have them today in GB. And so I prepared a few a few slides and with the help of my colleagues on the role of batteries on August 9, so during blackout so this Be more of a more of a focus on energy storage, particularly maybe a deeper discussion on some of the aspects of the inertial question that was raised by by Mike O’Meara. Okay, so I don’t propose to go through the events. Again, I think max assessment was was very good. I’ll just pick out a few, a few particular points of interest for me. So the first is that we see at the point, the, the Fourth Arrow dine in the tree is his little Bart Barford ST trip. So that is the steam turbine of a combined cycle. gas turbines will cc DT and little Barford How has the particularity of having one steam turbine for two gas turbines. So what I find interesting is that we have this the steam turbine trip, and then later this is probably about five minutes later, for Five minutes later, we see that little Barford, GTYA and GTYMV trip subsequent. So those are the gas turbines that feed into the state. And this is all really to say that we’ve moved from a system where we have several generation providing almost passively and not really being rewarded, or remunerated explicitly for the provision of inertia. There’s been confronted with new conditions, new operational conditions, and therefore, we see that actually, they’re facing up to some slightly unforeseen sort of issues and probably never expecting that first steam turbine to go down. And then, you know, having to deal with the cascading trips basically through to the gas turbine. That being said, this is obviously a very, pretty remarkable event that actually happened. Just for those as well. I noticed there was a question that through the q&a function, we had the load shedding questions. So frequency load shedding. That’s the point at the very bottom of this blue trees. So we have frequency breaching 48.8 hertz. And that triggered really when, when you KPMG national grid, we’re starting to make decisions about where to cut, cut load from the system in order to re stabilize the frequency. So that’s, again a relatively rare event for the GV system. But one that Jonathan was quite right to indicate that to modestly sized generators going down should probably not have led to that particular situation. So how did batteries start to help basically in this in this situation, the real I suppose the vast majority of batteries in GB currently are providing at least to some degree frequency response. So pretty NC responses is a broad term to use for essentially providing additional megawatts or taking in additional additional load in order to help with the subsequently with the frequency. So, this is a frequency is a quite dynamic sort of property of synchronous grid. And this is managed in a quite active way. And what we see there is in the dynamic block at the top that’s where batteries have been participating. This is overwhelming the batteries in this instance in the provision of


excuse me, sorry, I think I lost the sign there briefly.

So, batteries have been holding capacity essentially to to deliver megawatts whenever the frequency deviates, substantially upper So, What we saw is that they’re getting units in bouncing mechanism 36 units and non BM and 10 units and enhanced frequency response, providing dynamic frequency response. So reacting really within very, very small frequency fluctuations to any opposite direction essentially to support the frequency of the grid. We also have static assets again, it’s going to be provided by batteries as well but from other from other technologies to which are triggered at a slightly slightly lower frequency to the dead by under 50 hertz is actually substantially larger for that and have slightly lesser sort of conditions essentially associated with with the provision of those additional megawatts when we also have some some demand units, in fact that we’re able to participate very actively management to that frequency. And what we can see whenever we look at these tools, really is that the low frequency holding crust and static primary secondary was more than what the electricity system operator and National Grid was expecting as the largest energy loss so that seems to be appropriate. What we see here specifically for some of QE powers assets that we have in our management, these are a couple of battery assets. The first Canon Insights wheels for magnet 4.8 megawatt our system, co located with some Peary’s wind power and anaerobic digestion is we can see on the left hand side of this tree, as we see its operating, we have the grid frequency in blue, and we have the power output from the battery in green, so it’s it’s mirroring broadly the frequency that we have measured that happens at the side. And on the left hand side this is a very typical sort of dynamic frequency response. profile. And suddenly, then we have the massive drop in frequency. And what we can see is that the battery responded very, very quickly to its maximum our level of four megawatts, holding that for as long as was required for the frequency to start to come back into a more normal sort of debt around around 50 hertz. And then on the right hand side of this chart as well, we have following the 15 minute period where frequency really was quite substantially lower than intended. We have to rectify that by going over frequency for a short time. And again, dynamic frequency response providers such as the cannon battery, were able to provide that over frequency response by absorbing energy from the system so batteries very good books providing over frequency and under frequency services. Likewise, we have This is the place the battery places semiconductors in Plymouth. Behind the meter asset as well, we see again the blue trees in frequency, orange trace of the part of the roof from that battery. So very, very quick response batteries are very, very good at doing all of these very fast actions that we used to use currently to provide for instance, the likes of frequency responsible hands frequency response, but also are useful even in balancing the system just in much more normal frequency conditions. So Mike also mentioned of the rate of change the frequency of Rocco visits. Typically, you know, on here this chart is showing just how quickly the frequency was changing at any particular time. So what we see here is a really substantial drop. Time to avoid minus point two five hurts per second. Now this is well outside the range and target range from National Grid. So, so then wants to really maintain the rate of change frequency, plus or minus point, one to five parts per second. And the consequences of that

was that it caused the automatic triggering of relates in the distribution network. Know, what we heard was the these two generators together. Yes, it’s quite a substantial drop in generation. But what really exacerbated This problem was the disconnection of distributed energy resources. So this is smaller thermal generators, but lots of lots of solar for instance, is connected the distribution level, lots of wind to the we’re not particularly in this part of this part of England anymore. So I think we also need to be cognizant of this really is that these rid of Change the frequency restrictions can exacerbate the problems caused by actually the lower ownership sort of system that we’re seeing currently in GB. And in other markets. So essentially, we should be very, very aware of, of not just allowing new by setting our debt boundaries and frequency correctly, but also around the rate of change about frequency and this is where inertia really comes in, is when managing rate of change of frequency should be mentioned as well that national grid are looking at, at increasing that tolerance, essentially a rate of change of frequency to one parts per second, which is obviously eight times more than than what it currently is, which would have prevented that disputed generation from disconnecting and help to remediate the situation. But that’s a three year program. So we probably can’t expect to see the results of that for a little bit.

So we talked about rate of change of frequency and about inertia really being the driver for controlling rate of change of frequency. So again, just to reiterate, sorry to hammer this point, but previously in the old world of thermal generation, large point units, we sell coal and gas, typically providing all of the inertia required by the system. So this is really just the spinning mass of these turbo alternator sets that way many, many hundreds of tons. Essentially, just their reluctance to slow down or speed up particularly much Newton’s laws, and made sure that the frequency was usually roughly where it should be. And that’s a passive type of type of approach. So what we’re seeing with with new non synchronous sources of energy, like when like soda like always The interconnection to GB is DC interconnected and it’s non synchronous. That is a bit of a change in the rules of the game. So, what can batteries do to alleviate this batteries don’t have a big trouble Ultimate Reset, but they can provide synthetic inertia. So synthetic inertia really is just about matching the capabilities of response capabilities so closely of what we’re calling analog inertia here from spinning mass. That in effect, you can you can avoid the requirement for that. And there’s some research here at the bottom from from Queens University in Belfast where I’m

referring to the in this instance this is the ICM so that all Ireland TriCity market

but that basically states that by allowing synthetic inertia to participate on a level or an equal footing to sort of synchronous Marcia We can actually see of the consumer, plenty of money. This is 19 million euros on unwind here. And so really by by having the really, really fast response times the batteries are capable of delivering, you can actually provide some of the initial requirements that could be could be demanded by the system. Another question is about, you know, we had a question already actually in the q&a about what what the sort of model is for for battery storage system GB. And I responded by saying that it was roughly in trading between are optimizing between merchant markets like the balancing mechanism here, and provision of ancillary services, and that’s to maximize the rep and maximize the revenue and therefore justify the business case for the building of a battery. And so what we’re seeing here is what happened on that same day in the bouncing mechanism. This being the shortest term Market to real time delivery of part to the system. And we see that this is actually a fairly unremarkable day in the balancing mechanism. So that’s not where the only to state that actually when we talk about alternative market arrangements that would enable the further access for batteries, then either we need to see some sharpening of these these market signals. So in a purely merchant sort of world, or we should really be looking at actually, should we be setting up additional ancillary services markets, should we be procuring different consulting services? Should we be measuring the right things and procuring the right things? So what does national grid so need to do to avoid a repeat of like the box blackout? There are a few ideas that we’ve noted down here on our side. So one is to procure a more fast acting frequency response. I mean, that’s, that’s an obvious solution, I think to most. The second is to increase system inertia by whatever mechanism that might be the third reducing sensitivity or response to the rate of change the frequency. And the fourth is to review and market mechanisms that we just discussed. So and point one, you know, we’ve seen the frequency response was was delivering, broadly speaking, actually, as it should, we had over a gigawatt of procured capacity. And we had over a gigawatt of response from those capacity providers for the purposes of frequency response. That was insufficient in this in this instance, today, a couple of different factors. But as max already mentioned, you know, even doubling the amount procured for frequency response is a relatively modest increase in terms of consumer bills. Obviously, we need to get the balance right because you can always say more or more That is a relatively simple way of doing, it does raise some further issues around reduce budget generation where we need to leave enough headroom actually for frequency response providers to provide that frequency response. And that could incur some some secondary costs as well. But you know, that’s, that’s an obvious way to try and avoid the sorts of issues again, second one is to procure more inertia. So we can do that a couple of ways. One is to dump off non synchronous generation and replace it with synchronous generation. That to me feels inconsistent, wholly inconsistent with the 2025 targets for essentially a decarbonised grid. We also see we can provide synthetic inertia through smarter technology, very fast responding technology such as batteries. We have different technology types as well that are available but batteries are Pritam. They’ve already done that. And they’re already actively doing that, in fact, today in the system. So I think that increasing the system inertia, perhaps even through explicit remuneration of inertial services is quite a quite an interesting idea in terms of avoiding this, this repeat of obviously, third point around the sensitivity of response to rate of change of frequency. I’m delighted that

national grid is looking at this

new increasing some tolerances, right, you know, frequency, not just frequency, but the rate of change of frequency, we have a system that includes increasing the minds of distributed generation of all types. We’re getting into more decentralized system. And indeed, we don’t really want to get into the sort of arms race of having to always increase transmission capacity, for instance, at great cost, in order to enable for further further transition. So I think that in this newly decentralized system that we’re living in ensuring that those generators are not automatically disconnected for instance is a really good protection mechanism indeed, is very useful and reviewing the market mechanisms have referred to international markets you know, we have ancillary services there are procured by National Grid car we have a certain suite of those from enhanced frequency response frequency response, short term operating reserve, we have our spinning Treasury

but we have

no real explicit inertial sort of market bear. We do have one for instance in a nice I’m again making reference to earlier which is it increased or an even more isolated 50 hertz system will be interconnected through GV. There is a an explicit ancillary service for synchronous initial response and opening that up to all providers of inertia, whether synthetic or not, could be an interesting idea in terms of really reinforcing actually not just the business case for it for, for the provision, the building of batteries, for instance, the charge which are required for this 2025 transition, but also in terms of actually providing a level playing field across all generators. And to think about a initiators requirement of the system. How do we grow by providing that

as efficiently as possible, and as is Greenway as possible.

That’s the end of my slides here soon. So, thank you very much for your attention, and very happy to obviously answer any questions that come through the Punic.

Great, thank you, Simon. That’s, that was a really helpful presentation. We’ve had a number of questions come through the q&a, but let me encourage people to send in more questions as we as we speak. Let me start with my question, which actually Two or three of the questions that have come in are pretty similar to, to my question. So I’m going to change my question slightly take it slightly a step backwards. So if if both of you, Simon and wedgie if you were both, you know, the policy maker in the in the UK, whoever that is, and you had to choose like one policy change to make or one policy to emphasize, and you had to choose one because if you try too many things at once, then it all gets a bit lost in the in the shuffle a bit lost in the noise. You want to go for one and you just want to kind of push it through,

you know, other likes of Brexit or something, but

you’ve got one thing to choose, but what’s your one thing that would prevent or at least substantially reduce the risk of something like this happening again, in In the UK?

Yes. Um, so So from my perspective,

I think, well, it’s definitely increasing either the tolerance of our plan settings or procuring more capacities to stop the frequency deviations. And so just to kind of bring this back this very briefly on to your initial question, because you asked, Do we have enough batteries in GB at the moment? And I think the question here at this from the notion of a blackout is not whether we have enough batteries, but whether we have enough phone capacities in place to prevent a blackout. Now, batteries is one example of a technology that can solve it. But at the same time, you could well be procuring for cc GT, so peaking technologies, etc, that could have solve the same problem at that point in time. So the question here should not be on whether we have enough batteries or whether we have enough manpower in place to prevent a blackout. And I think related to that, it’s a second question of if we believe that battery is the answer. This and maybe it is the answer because it’s a lot faster responding, or because it kind of ties in with the wider decarbonisation objective at this at this point in time, then the question is how do we go around on kind of procuring batteries and whether there’s sufficient financial incentive in place for batteries, which I might be for later question, but I think that’s my thought on this.

Yeah, I would echo. In fact, most of the points that Mike made, I’d say that what we can do today, or what National Grid can do today is to procure more, they said procurement volumes for frequency response monthly, and indeed, they can even start doing weekly with the facts or acting frequency response auction trials. So I think the the immediate step that I would be taking is to procure more frequency response. Unfortunately, I think that what we saw in the immediate aftermath was actually a direct copy. Tracking with thermal generators to stay online and provide that personally, I think that’s totally inconsistent with the stated objectives of of the government. And indeed, you know, just the absolute imperative to decarbonize our our energy or electricity system. So that’s the first the first step. I think that absolutely that this this question about tolerances around frequency on our own rate of training frequency, and should be should be addressed. I think that there we saw with the the rate of change of frequency question that actually contributed at what was it what is intended as a positive sort of protection mechanism for distributed generation actually contributed significantly to exacerbating a problem that was existing on grid. I think the distributed generators are really capable of providing quite a lot of the requirements of the grid and, you know, allowing them to participate And a more level wing is always important to us. But Furthermore, they’re also this is good technology that sitting at a lower voltage level, there’s no real reason to automatically have to disconnect. So I think that it will be first and foremost, for tomorrow. And secondly.

Okay, thank you. So a kind of related question that I’m seeing coming up in the q&a box is should it become mandatory for those who provide intermittent renewables windows solo or whatever, to to also provide battery services? I mean, this this question may not just it may not be a UK specific question. I think it’s arguable in the UK. But when you look at countries that have less complex, less sophisticated wholesale electricity market mechanisms or have auto really have a wholesale electricity market, then, you know, it’s one way to deal with the with a lack of a market or perhaps a lack of sophistication in power system planning to just make it mandatory to say, if you’re going to provide something into milsom, then you need to firm it up with a with a battery. You may Of course, then kind of get a cumulative over investment in in batteries. But it’s better perhaps to get it wrong in that direction than to get it wrong in the other direction. What do you think?

Maybe I’ll take this one first. Too much. I think that making it mandatory is is maybe unnecessary in the GV context. I think we’re getting quite a lot of interest through from renewables developers in particular You’re actually looking at coupling, particularly solar at the moment with with storage. Indeed, I was at a conference a few weeks ago about solar and storage, specifically together. So I think that that in the GB context is maybe coming naturally, to protect against price cannibalization, and all sorts of other all sorts of other risks essentially associated with with renewables development. More broadly, though, I think that we’re in a world really where we’re tending towards more much and exposure to renewable providers, for instance, you might be interested in building a battery as well, even if it’s a relatively small battery, to limit their imbalance exposure. Because that can be a quite penalizing sort of element in the business case for for renewable development as well. And so, my my answer would be that probably it’s not Yet necessary, or I don’t think it’s necessary in the GB context to to impose that. But it could be well worth considering and different in different jurisdictions, particularly in a more decentralized sort of system. So if we’re if we’re looking at connecting up large parts of sub Saharan Africa and large parts of India, for instance, via solar micro grids, then I think that micro grid with solar battery would be a fairly appealing sort of prospect, whereas when with a battery unit on the correct management system can be to be really, really strong.

Yeah, that’s an interesting response. I mean, certainly emphasizing the decentralized aspects. That may also be an argument for doing this kind of thing and systems that are more centralized than, than the UK in the sense that if you Let’s say in a fairly typical developing country power, power sector power system, where you have a single buyer at the wholesale level, not a not a competitive market, and that single buyer perhaps needs some kind of rules of thumb in order to plan capacity and plan, including planning storage, then one way for them to do it is would perhaps, and sort of almost thinking aloud but would perhaps, be to say, if you if you build an intermittent generator, then build a battery with it or if it’s an intermittent general generator above some particular minimum capacity wouldn’t do it for really small ones. Then build a battery with it so that we have enough storage capacity or enough battery capacity in particular, on the on the system. That we the central planner, leave it to you to think through how much capacity we need or leave it to your decisions to make that happen. We don’t have to we don’t have to figure it out because we don’t we don’t know the overall need well enough something along those lines because we don’t have a wholesale a competitive wholesale electricity market, you know, as you have in the in the UK. You need to come up with some very different solutions. veggie Do you want to respond on this?

I think that’s there’s two parts actually even when we talk about imbalance payments and stuff so Simon Simon Simon alluded to that. But one of the key things to note here is arm when we talk about renewables paying for the imbalance, that’s one which is you have balanced with the system and you pay for it but there’s a why the one which is how much the great must spend to ensure that there is sufficient backup services for poles and wires in place to kind of firm up your power for you. So I think that’s a wider question there of whether you have to have big storage assets with renewables, etc, to lower the cost of the wider society of having intermittency in the system. And part of that comes from the fact that, in today’s world, lots of renewables assets are subsidized in GDP. So if you’re subsidized, you’re not bearing the full cost, even of the kind of intermittency that you’re closing to the system, because essentially, whatever megawatt hour you generate, you get paid for it. And you don’t really care about the wider impacts of it. But if you were in a subsidy free world, which is what GB is transitioning towards, on, Simon Simon has alluded to, you see a lot more business case where you want to stick storage with your renewables. So because of this wider cost, etc, does that bearing an impact on to your business case. So in GB itself, Nestle being the company Develop the first on subsidy free solar assets, has actually said that without the storage system with the solar one, they would never have gone ahead with that investment. So it’s about the portfolio impacts that are in place where they are probably a multitude of futures that could happen. And having a solid pipeline and a battery actually helps to kind of leverage that risk a little bit. So it does help with your business case. And even if you don’t think about the wider impacts in the system, but answering your question directly, I think there is indeed a role to be played where renewables should bear the cost of intermittency.

Now that’s a that’s an interesting responses. Interesting. perspective. I mean, what what, what I think I’m seeing happening and this is more broadly than specifically in the, in the UK, really is that many, many countries are rapidly transitioning to subsidy free renewables and at least for the kind of, I think what you can now call conventional venture Your host the PV and the and the wind, they’re moving towards a subsidy free system. there they’re moving to it because the the sort of level eyes cost of the project level not talking about the system, but the level is cost at the project level is declining so rapidly that that people want to invest in it and sis and systems wanted, meaning that you are coming more rapidly to a high penetration rate of, of intermittent renewables then than many people expected until extremely recently and any people still don’t quite expect it. But it’s getting so much cheaper, so fast that solving this problem of flexibility solutions and particularly of fast response storage is becoming more urgent, then that most of us expected it would have been Many of us saw Okay, eventually that’s going to be a big problem, but we didn’t realize eventually was was just around the corner for a lot of countries. So this this question of, you know, how do you structure it incentives to, to investors or to operators? How do you structure incentives so as to call forth enough batteries or similar fast response, flexibility solutions? How do you call them forth soon enough to avoid events like you like you had on August 9 in the UK and that you could have in another in other countries as the penetration rate of intermittent renewables rises. I mean, what I perceive happening is that these things are evolving so fast, because the cost of renewables is declining so fast, that policymakers makers and planners and even individual investors are kind of running to to catch up, which is why this kind of discussion is so important and why, why I think that the August 9 event in the UK may prove to be a really kind of watershed event in people’s thinking about how to manage power systems and particularly how to manage high rates of variable renewables on those on those power systems. We’re coming to a close but let’s just see if there’s a question,

nestling there and they in the q amp A.

So here’s a question that because a few times

if you wanted to emphasize more, the kind of demand response sickness side of this managing demand responsiveness as compared to managing the sort of storage side of this or the interconnector side or some supply side


what what would you emphasize in in the UK context? What would you do more of on the demand responsiveness side in order to deal with this kind of situation?

Whichever of you wants to take that first

I’ll jump in. Yeah, so QE parles, who does a lot of demand response. So commercial and industrial clients in particular, I suppose that a really important way to expand that and I think that this dovetails really well demand response dovetails actually really well with flexibility in general, it can provide substantial flexibility across electricity also in arguably for heat from instance you can use distributed networks for instance, and actually manage the demand of emotional theaters for instance, to manage that. I think there’s some really, really interesting thinking currently in academic circles. That needs to move from some of those academic circles, I think into into reality. So talking about power specifically, say the time of us tariffs and smart metering and it’s a, it feels like we’re buying this from again, but really sharpening the price signals again, if we want to enable wider participation and actually actively managing one’s own consumption of electricity. Those will be really, really key enabling sort of factors for for that and QE car and are amongst those who are probably best situated to go and deliver some sort of offering to those consumers who are interested in and actively managing their their demand.

So from my end, I

think there’s two parts to this. One of that is how do we encourage more demand side response? And I think it goes back to the fact that in GB we have options for this we have from frequency response options but demands a response can participate in it. But batches have outbid the demand side response in this auction. So maybe the answer is you have to procure more. But more importantly, on the demand side response question. I think it’s about what happened in GB when the blackout happened was that the frequency hit something like 48.8 hurts national agree involuntarily turned off 1.1 million customers. Now if you had demand side response, I think the question here would then be perhaps you could actually turn down customers who are actually voluntarily turning down when there is a need to turn down and I think that’s actually the crux of the answer here, which is, can we move away from a world where customers who value their power more stays online and possible vendor the power less or willing to accept the payment to turn down, actually are the ones that are affected by a power cut. And I think that’s a wider question that should be answered.

good way of putting it in a system where

hospitals were cut off and, and the rail the railways with, you know, hundreds of thousands of passengers were cut off, I think one can pretty safely assume that they were not voluntary, voluntary cut offs. And if there are if there are other customers that could have, you know, happily born that for for an incentive, then then then they should have done. I think we’ve kind of run out of time. In fact, we’ve gone a bit over time. I think this is a topic we could probably talk about for four hours, but we don’t we don’t have those hours. So let me thank Simon and DG for for very thought provoking and informative Presentations thank everybody for participating and sending in your, your questions. I think this is some this is a topic that a TA will probably want to return to in future webinars in some way. Because it’s a it’s a very live topic for anybody who’s who’s thinking or working on the design of power systems and empower markets to enable countries to reach much higher levels of renewables, whether they’re doing it for decarbonisation reasons or doing it for economic reasons, which is increasingly the case. You really need to think through the incentive mechanisms and the and the system plans to make sure that you can, you can do it in a, you know, in a reliable and least cost way. Thank you very much to everybody. Bye

Share This