Did Interconnectors Save the UK from a June Blackout? | Noémie Baud, Montel

Round of applause for Noemi, please. Sorry, it's more technical issue. Hi, everyone. My name is Noemi, and as promised, we're going to now, like, try to answer this one big question about, sorry, the first heat wave of this summer, which took NISU into, like, yeah, I guess, more dangerous territories, and the questions that we want to answer is, did the power interconnectors save the UK from a blackout back in June 2026? The short answer, I'd say, is partly. The interconnectors were actually responsible, in part, for the crisis, and made it worse for the UK, but also helped resolving it. We saw that the French interconnector flows changed by more than 5 gigawatts from importing to exporting on the 23rd of June. And that was during exactly the time we needed the power the most during the evening peak. That is actually a big part of why the frequency dropped on the day for about two hours. But we can also see that the way out of this crisis was actually mostly related to the interconnectors. We saw an emergency instructions coming from BRITNED, the Dutch interconnector, and another emergency assistance from RTE. So, and just for context, blackout is actually a bit of a strong word. I would say that the frequency never reached the statutory limits range that is between 49.5 and 50.5 hertz during the entire week of the GB, or European heat wave. But it did spend long periods of time outside of the normal operating range. So, here I'm just going to talk to you through the setup of this week, and in particular of these two first days of the heat wave. So, it was a heat wave, warmest June on record in the UK, and potentially in some other part of Europe, ending at 38 degrees by Friday. And it wasn't just in the UK. We saw five plus, like, yeah, we saw these extended high-pressure systems bringing the temperatures all over Europe 5 to 12 degrees above seasonal normal. So, the neighbours we're usually getting the power from, as discussed in the previous presentation, were also short power and wanted to keep the power for themselves. Furthermore, we saw the gas plants, the rating, and losing part of their output because of the heat and because of these, like, high temperatures. On top of this, it was peak maintenance season, so, as it usually is, summer is peak maintenance season, and there were, therefore, less spare units to begin with. And in the two rows at the bottom, I actually tried to highlight how every technology was actually impacted, or is, in general, impacted by heat. So, solar panels, so, solar panels, actually, above 25 degrees, start to lose output. We see that wind and gas, for exactly the same reason, also lose efficiency due to one factor, that hot air is actually thinner, and therefore, lead to less output power. And, anyway, like, even though, yeah, wind was low to start with, but even from the low level, we saw even less wind generated than expected. The CCGTs, also above 15 degrees, baseline, are the rating, and by the end of the week, by the Friday, on this week of June, we saw a decrease of up to 15% of the CCGT outputs due to high temperatures. Then, then, I'm moving on to the nuclear, and, like, thermal plants that actually cool with water, and river water in particular, warm, like, they faced a double bind, where warmer cooling water and reduced water supply, basically, reduce their output. Then, then, there is the grid hardware, that also suffers from the heat, like, one example here was, during this week in June, in France, the transformer overheated, and 70,000 homes lost power. And, and, all of these factors also compounds, where solar is suppressing the daytime prices, so, during the middle of the day, the gas plants are not really in the money to run commercially. Therefore, they need to recover, like, or run only in the evening, and recover the startup cost on fewer hours. So, overall, no one escapes the heat. Like, everyone is affected, every technology. Here, sorry. No, no. Next, sorry. I will call the flows, so, the interconnector flows fragile, and, for exactly the same reason as we've discussed before, interconnectors follow price. The power flows from the cheapest place to the most expensive market. And, here, there is a table where I've tried to highlight, basically, the darehead prices of the Tuesday, 23rd of June, in the evening peak at 19.30. And, you can see that the reference, that I call, like, the GB reference, is an average between the apex and the Nordpoll price. Was that 243 pounds? We see that Belgium, Denmark, and Netherlands were all above 500 pounds per megawatt hour. That makes up a spread that is greater than 300 pounds. Therefore, power was flowing out of GB on the three interconnectors. Ireland was also flowing out. The only two countries where we could see that the flow would flow in was France and Norway. And, the French spread was only 11 pounds per megawatt hour, which was not a lot. And, just as a quick reminder, Nisei, on the 20th of May, introduced a new rule that they could not trade more than 300 megs against their darehead position on each of the interconnectors, with an overall sum of 1.5 gigawatt across all interconnectors. That, actually, also limited Nisei's, like, how much Nisei could lean on the interconnectors when things actually tighten on the day. So, next slide here. I just want to show you, basically, explain a bit more the French GB spread, because that will be interesting later. So, here, the dashed lines are actually the darehead French and GB prices. And, you can see they're really close together. And, they're around 240 pounds for the evening pick. What we saw, like, on the two solid lines, we see the intraday prices 75 minutes before delivery for France and for GB. The blue one is GB, the green is, no, sorry, the green one is GB, the blue one is France. And, the green area at the bottom is actually the spread, so the difference between the green solid line and the blue solid line. So, here, what I'm trying to explain to you is that, on the day, we saw France spike to up to 1,200 pounds per megawatt hour in the evening pick, where, actually, GB stayed around 600 pounds, which is still very high, but France was higher. Therefore, we see that the spread, like, this negative spread reflects that the flu actually reverted and the power flew from GB to France because France was more expensive. This is what I will summarize here, the heart of the story, the result of this price gap. At the day ahead, GB was set to import 3.5 gig from France over the Tuesday evening pick. Then, the French price rallied up and the power flows, sorry, the power followed the prices. What I mean by that is that when the final schedule at 7 o'clock of the interconnectors was published, we saw that France was actually importing 1.5 gig from the UK. So, that is, overall, a difference of 5 gigawatts for GB. So, we were shorter, or NISO was shorter by 5 gigawatts from the day ahead to the intraday stage. And what I'm going to show you in the next slide is that, what we've mentioned previously, NISO tried to buy back some of this volume to reverse the interconnector, but they couldn't hold the flow back. So, here, what we see is NISO's intraday trades by cable, what we also know quite well in the industry as BSAT trades. This chart, which is, like, a replicate of an enapsis chart, is basically showing the volume that NISO bought on, like, in the intraday interconnector auctions on Tuesday evening to cover the shortfall in GB's operating margin. The trades were done quite early on, so by 2 p.m. at a price of up to 875 pounds per megawatt hour. That's three hours before the peak. And that's also what's important before the French price actually rallied up intraday. So, one of the key lines that NISO made in the preliminary report that was published in September, they did say that trading by other market participants in the opposite direction on the 10 connectors reduced the benefit of this exact trade that they were taking for operating margin. In other words, NISO bought the import, and as the price froze in France, the rest of the market sold the power in the other way, and the net result was an export position that we saw in the previous slide. Here, briefly, I want to show you the price evolution. That is just for GB, like, from there ahead to intraday. We say the dashed line is the deadhead position. I think it was, yeah, at 230 pounds per megawatt hour on IPEX. Not full, I didn't put, just for clarity, but it was roughly in the 20 pounds mark, like, more or less, it was close to IPEX. We saw with the green line, like, an increase on the half-hourly deadhead auction price, which is also at the deadhead, which was to be expected, but it only went to 256. So, so far, expensive there, but not that expensive. Intraday, we see the yellow line is the MIDP, so this is the short-term intraday trading average. That climbed to 560 pounds for the evening. And the purple line is the imbalance price. That went as high as 800 pounds. For reference, I've actually added this low-frequency shaded band. Between 6 and 8, this is where the very low-frequency event happened on the day. And we're going to go through it now. So here, on this graph, I'm actually showing you the frequency line. Shortly after 6, the frequency fell to 49.85 hertz, sorry, and stayed at or below this level for almost two hours, spending 48 minutes in total below the operational limit. That's a very long time. NISO took every action possible. Here, in total, they said they took 4,750 actions to try to get this frequency back to normal. But the fix actually came from the interconnectors. The orange arrow is showing the BritNED instruction. The Dutch, at 7.12, exactly, NISO sent an emergency instruction to BritNED asking for the flow to be cut. So they immediately gained 750 megawatts. We see the frequency coming back up. It reached just about 50 hertz, 50.03 hertz. And then it fell again at this lowest of the day. And that was the most concerning part. It went down to 49.657 at about 8 p.m. And then you see the purple arrow. The purple is the emergency assistance delivered by AIFA, which is the French interconnector. And they delivered, they regained 500 megs from it. At this point, that brought the frequency back to normal, healthy range. And the crisis was more or less over. But one thing to actually mention is that this help from the interconnectors, and in particular from RTE, was discretionary. We saw RTE on the day withdrawing twice the help, the emergency help to NISO. So were we close to a blackout? I think the answer is not quite. The frequency, the statutory frequency, was a limit, that is at 49.5, was actually never reached. And NISO still had options. They had more interconnector exports that they could have reduced. They could have taken further emergency actions. But it doesn't mean that the system was stressed. I think it was actually very stressed. And this is what we see here from this chart, where we can see that the British system margin by fuel on Tuesday collapsed for the evening to around 2.5 gigawatts between 7 and 9 p.m. For National Grid, or for NISO, there were very little left in the stack to deal with. There were only 600 megawatts of CCGT's margin, which was CBank 1, that was offered at 800 pounds per megawatt-hours. And the whole battery stack was actually discharged by the evening. And they were discharged or offered up at up to 1,000 pounds per megawatt-hours. These batteries really helped save the day and the early evening. But by the evening, they completely ran out of power and couldn't help when the system needed them the most. So, the findings from NISO's report are as foley. To be fair with the interconnectors, the interconnectors were not the only reason why GB struggled on the day, or on the 23rd and 24th. There were other factors as well. The demand was higher than forecast due to cooling demand. The gas plant derated, as mentioned before. There were network limits with exceeding constraint limits, sorry, by up to 800 megawatts in the southeast, which actually prevented some of the generation units of being used. And we also saw that response and reserve services were under-procured under-procured because the batteries were empty. The wind was under-delivering, even at a low level. It's still under-delivered. And there were unplanned outages on the CCGTs and the interconnectors. So, the interconnectors were one factor, but one factor among several. So, here's just a quick idea or a quick view on how can we actually improve the system security from here. And I've just put, like, a few big ideas to start the discussion. So, here I would say that, first, more battery storage. Batteries really carried the evening through Tuesdays, but the fleet was actually empty by the evening peak. Two, more visibility on demand. We have more and more embedded generation that sits behind the meters and that NISO cannot see or manage really well. That feeds into the demand forecast delta that we saw for this entire week. Three, demand response, just to help taking the load off the peak. Four, more or a different preparation or planning on outages. Outage season is in the summer. Is that still valid or is it something we need to rethink? Five, a more secure, reliable, and predictable interconnections. For example, with the use of NTC or ITL, which are, like, tools used by NISO to manage capacity on the interconnectors. And six, here, I'll just put some already tools. The tools we're currently using for these tight days, tight margin days, are winter tools. And for valid reasons, EMNs were, like, something that only happened in the winter in the past. But now we're seeing them happening in the summer and we need to think, how do we apply this learning on a hot summer day all across Europe with cooling demand? So, one last question I have is, how can we prevent European health from being withdrawn when we actually need it the most? And here is just, like, my, almost my last slide. I would, I wanted to put a verdict or, like, my verdict of this. Did the interconnector save the UK from a blackout? And I think it's, it's not a yes, it's not a no. It's, the interconnectors are actually a good safety net. They do help most of the time. But they're not a guarantee. And we've seen it here. The, in, when the price in France became higher than the GB price, we lost five gigawatts of power. And secondly, the second point I'm trying to make is that on Tuesday, it's not the market flow that saved the day, but the emergency tools that are linked to the interconnectors. And finally, the assistance that we've received from the TSOs is actually at the TSOs discretion. On the 23rd, RT withdrew it twice in one single day. So, interconnectors were part of the problem, but they were also, on this day, part of the solution. Thank you for listening. Thank you. Brilliant, Naomi. Excellent. A real fascinating deep dive into the events of the 23rd of June. It would be too easy to just blame the French, wouldn't it? Maybe a lazy, lazy accusation. But, no, unfortunately, we're running out of time, so if you have any questions to that excellent presentation, please approach Naomi in the break, because we have to be back here at five o'clock, unfortunately. So, please, one more round of applause for Naomi, and here's a little present for you. Thank you. APPLAUSE