25 - Keeping the lights on (Faye Relton and Neil Morgans - Ancillary Services Development Manager and Balancing Services Optimisation Manager @ National Grid ESO)

[UPBEAT MUSIC]

Hello, guys.

Welcome to the podcast. This is a very special podcast indeed. We've been waiting for this for quite a while.

I'm sat here with two big dogs from National Grid ESO.

You guys are celebrities. And I'm sure all of our listeners are going to be dead keen on what we're going to talk about today. We're going to talk mainly about frequency response. But we're also going to talk about other stuff. I just want a huge warm welcome and a red carpet to you, to Modo. And thanks for coming in to the Modo offices.

Thanks for having us.

Nice to be here.

So you guys--

I think the first thing to do is for you guys to introduce yourselves. So Faye, do you want to go first?

Yep. Faye Relton, Ancillary Services Development Manager, which is a lovely, lovely name.

I essentially lead on frequency response reform. Been doing that for about two years and loving it. Loving it, still.

QUENTIN DRAPER-SCRIMSHIRE: Awesome. We're going to delve into that in a second.

Neil Morgans. I'm the Balance and Services Optimisation Manager. And I've been doing that for around six months now. So our team run the procurement activities, the head for our response and reserve services frequency--

so DR, DM, DC, Store, FFR, et cetera. So all the exciting stuff about alternative costs and elastic buy curves and overholding, and those sorts of things.

So am I right in thinking that you're more of an auction person, and Faye, you're more of a, sort of, general designy management--

how do you guys talk about it between yourselves?

Oh.

Yeah so--

We do it all.

We've worked together in a couple of different roles throughout our time at National Grid. I think I'm sort of, typically, more on the technical side--

the analytical side--

and face more customers to focus on and delivery of--

Yeah, yeah, yeah.

But we--

I think between us on response, we tend to pick up what needs to be picked up. So we both worked on the procurement side recently because we needed to make some changes, improvements. But then, yeah, Neil's team will do, like, the day to day.

And then I'll have a look at what we've committed in our business plan. What we have to do? When, if we said, we'll do it? What are people telling us outside of ESO?

What are ESO telling us that we need to do? And how do you make that work--

how to fit together?

How long have you guys been at National Grid? There's a lot of, like, long-term--

people seem to go to National Grid and stay there, right? How long have you guys been doing this?

So for me, it's approaching 10 years. I started in the GSA--

so gas side of things--

as an analyst forecasting demand for gas as part of the new incentive that they bought in at the beginning of RIIO.

Then, moved on to the EMR team, where we worked together. And I was part of the first capacity market auction, the auction team for that. So that was pretty exciting at the time, at least I thought so.

EMR being Electricity Market Reform, right?

NEIL MORGANS: That's right. I'm going to--

[INTERPOSING VOICES]

I'm want to just jump in each time.

NEIL MORGANS: Yeah, sorry. I was trying to be mindful of that. But it's kind of straggling there.

[LAUGHTER]

From there, it sort of a fairly very pass. So I went into our legal separation program.

We became legally separate from the electricity transmission owner in April 2019. So I was leading on the separation of our IT systems that we shared with the electricity transmission owner, making sure their data was separated we had the right commercial arrangements in place--

services between the group. Then, led on the delivery ESO data portal, which I know you guys use and have liked.

Shout out to the ESO data portal. We love it.

[FAYE LAUGHS]

NEIL MORGANS: And then from there, sort of dived into the world of frequency. So I was the quantitative analysis manager during the technical design phase of DR and DM. So our team looked at stability in that--

dynamic regulation, dynamic moderation. Looked at the stability analysis, performance monitoring rules, those sorts of things, and then very recently--

again, this year--

moved to the Balance and Service Optimization, which is a new team focused on our procurement activities.

OK, cool. And Faye, how long have you been--

oh, quick question. So 10 years, do you get a commemorative tie?

I don't get a tie. There's something, and I think you get to choose what it is. But I don't know more than that, I'm afraid.

QUENTIN DRAPER-SCRIMSHIRE: OK, cool.

A tie would be good.

Yeah. You might be able to choose a tie.

But would it be, like, a bright orange tie?

That would be my preference.

[LAUGHS]

QUENTIN DRAPER-SCRIMSHIRE: The orange thing is kind of new, right? I like the orange thing, but--

FAYE RELTON: Yeah, the orange is like 29--

I miss the old days.

NEIL MORGANS: That was part of the legal separation--

so a new brand, a new identity.

QUENTIN DRAPER-SCRIMSHIRE: By new brand, you mean same brand but orange.

I wasn't pleased about the orange. The blue--

and I'm wearing blue to represent the old school, I suppose. But yeah.

Yeah.

I wore orange since, like, 2019.

We did talk about wearing orange. But I think I've only got a high-vis vest, and it wouldn't go with Faye's hair.

No. I didn't go, no.

And Faye, how long have you been--

FAYE RELTON: I'm 12 years.

Wow, OK.

I don't think--

I never think it's that long. And then when I say it, it surprises me.

So I joined on the grad scheme, 2010.

I had no idea what I wanted to do. But I heard it was a really good company, so I joined. I was in IT. I was a business analyst on the grad scheme for a bit, and then gas emergency response and repair.

That was fun--

you know, gas distribution.

And then I did some processing stuff, where we mapped out, and we looked at where we get the value and where you can save the most money and make everything better for the end consumer, essentially. And that was in gas operations.

Back into IS, and I did an IT project in the National Control Center for Gas. So brought in a new telephony system, which that was really exciting. I quite like being on the operational side of things.

And then I was over to--

I thought, oh, I need to do some electricity now. So then I went over to Future Energy Scenarios, FES.

Yes.

FAYE RELTON: Yeah, love the FES.

Love FES.

FAYE RELTON: Love the FES. So I led on the document for probably a couple of years. And that was a real--

that was like deep water for me because I had to understand the scenarios--

loads of stakeholder views--

to manage changing the scenarios each year. So that was really fun. And then I think I went to EMR from there. So then I was in the capacity market in the CM.

And I was a customer team lead. So a lot of my work--

I love working with people. And even though I get a little bit upset when people don't like what we're doing, I like to talk to people and find out how we can make things better. So at the CM there's a lot of, how can we make things better. So I worked on, oh, pre qualification--

that lovely process. And that was--

I think I was in the CM for about four or five years.

What years was it? Was this, like, TA turn-down time, or is this later?

FAYE RELTON: No, so I joined 2016.

So that was when--

so I was brought in because they said, right, the CM is going to get massive. We're expecting three times, was it? Because you were there in 2016. I can't remember.

The CM's never going to take off. No one's going to take part.

FAYE RELTON: But I remember someone saying--

like the senior manager at the time said, right, we need we meet someone in who can improve the process and make the pre-qualification process better, internally and externally. We were expecting about 1,000 applications that year. And I think previously we'd had like--

I'm going to make up a number--

I think it was around 200.

Yeah, I don't know the number of applications. But I think we were delivering at pace. There's a large volume of changes with developing systems and refining processes.

Yeah. There was a lot of change that year.

So that was fun.

It was really difficult, and there was a lot to process. And there was, like, a manual assessment process. And that kind of built year on year, so that was good.

And then I got this role. So it's my first managerial role. I love it. It's great. It's really difficult. I can't keep everyone happy.

And who's your team? Because I know some of our watchers and listeners will know some of the names, right?

FAYE RELTON: Yeah, you'll know some, hopefully. So Sherry Lee, she does a lot of the analysis for new response markets. She's looking at reserve now, as well. And she's looking at the new auction that we're developing--

the new auction platform.

There's a couple of my team who got promotions, and they've left me recently, which is excellent for them. But I do miss them. So there's Charlie Watts, and she worked on DC. And she delivered DM and DR.

And then there's Kashia Anderson, who's--

she did DC high-delivery. And she worked on DM and DR. And she's working on this year's release, which we're sexily calling Release One which is the first set of changes to all the three new services.

So yeah, I think those are probably the names. There are loads of other people who I'll probably get in trouble for not calling out. But those are kind of the key names that people know.

Other favorites are available.

FAYE RELTON: There's loads of favorites. There's loads of great people.

So can I ask you guys what's--

big question--

what's going on with frequency response? So everything's changed, right? We used to have FFR. We still do have FFR, of course.

And National Grid chose--

probably supported by industry is the way to say it--

to make some changes. And now we've got the new services coming in. But on the whole, what's going on with frequency response?

What's going on? That's a big--

That is a [INAUDIBLE]

question.

to start at like the wide end of the funnel before we go into the details, right? Because we've got people who listening who are maybe international, or--

what's the general direction of travel? What we aiming for?

In terms of the suite of services, in particular, what's the--

or--

QUENTIN DRAPER-SCRIMSHIRE: So like, we need frequency response to manage good frequency, right? We assume that bit of knowledge. And then we had a load of old services, and we've now got some new services. And what--

where are we in the process of introducing new services? And what have we learned along the way?

Why don't I do that one?

So I did a bit of a looking back at what we've been saying. 2017, there's a System Needs And Product Strategy--

SNAPS--

2017 document. And in there, we talk about how we need faster response products. So we started talking about that in 2017.

And then--

and this is before my time, so I was fully into the CM at this point--

but then in 2018 we started talking about a roadmap for the future of frequency. And we split it out into Response and Reserve.

So in 2018, we started talking to stakeholders about we need to speed up response. We're going to start developing a service. Tell us what you think. And I think dynamic containment--

at the time, it was called Project Marvel, which we changed.

so I think Project Marvel started developing in 2018, so we knew we needed to secure our largest loss. We knew it had to be quicker than what FFR does. So then we started to then think--

Secure your largest loss--

that means, like, if the biggest generator trips from the system. Is that what you mean?

So it's your generation or your demand loss. So currently, you're looking at, like, interconnected demand loss and then your largest generation loss for DC low.

Just to come in on--

I don't want to overcomplicate things but there's the concept of rock-off losses, as well as the rate of change of frequency. So a big instantaneous loss, depending on the system conditions at the time, could also lead to a loss of distribution--

distributed assets that have sensitive relays on the equipment. And therefore, we need to respond not just to the largest loss, but any consequential losses or knock-on effects that might happen through that.

FAYE RELTON: Good point. Very good. No, no, good.

[LAUGHS]

So yeah. We knew--

that's what we knew. And we didn't know at the time what that looked like because we have to develop a service that people can participate in, right? So we knew--

I think we knew a lot about DC because we knew the largest loss need and the rock-off loss.

And then we--

then the team, that existed at the time, wrote a document talking about a roadmap. So we need to phase out of our legacy services. And we need to start buying newer, faster services that are more simple, standardized, they make more sense, and they're accessible to more people. And then that's when we--

And not reliant on bits of paper.

FAYE RELTON: Wouldn't that be lovely? Yeah, like Excel and Manual--

there's loads of stuff that we wanted to do internally, and we knew that there were better things we could do to increase participation and competition. So we then started testing hypotheses around, OK, if we moved procurement closer to real time, what does that look

Does that make it easier for people to participate? Does that reduce costs? So then we said, well, let's do a trial. So we moved--

this was the first exciting move from Excel-based manual process tenders to let's get an auction platform.

So I think it was around between 2017 to 2019, we started we were working on this trial with EPEX, the auction platform provider, and we launched the weekly auction trial. So we moved FFR, Firm Frequency Response, to a weekly--

a week ahead auction, in addition to the monthly tenders, and we were testing a few things.

And we'll go into this I guess later on, but things like we had an algorithm. And within the algorithm, we said, well, let's prioritize really simple blocks, so that's linked to the merit order constraints that we could talk about in a bit. Let's make it really simple, and let's see what technology turns up and what the prices look like.

So then, that auction trial, we've got a lovely, big report online that I still dip into now. Actually, I was looking at yesterday to look at what we've learned and what we want to do in the future. And then at the same time, we started talking about these are the new response services. So you've got modeling going on, which is the department where Neil works, and looking at what the system needs are.

And OK, so what do we absolutely need, and what can't we change? And then my team will go, OK, what can we change? What can we tweak? And then we'd speak to providers and wider industry stakeholders and say, how do you see yourself participating?

What's mad about this? Right? And National Grid gets--

National Grid has got a tough job, and there is no playbook. This is the funny thing. It's basically a lot of experimentation and learning and doing iteration and feedback and experimentation, but you have to be completely transparent about it.

You're beholden--

rightly so, because of your position in the market. You have to be fair and equitable, and it is tricky. And so I just want to make the point here that listening to you explain all this stuff, when basically nobody's done it before, we can't look at another TSO.

There's bits. There's little bits. You can't look at another TSO and say, they've done this and nailed it before. They've nailed systems. They've nailed service design, auction design, all of it, and I just wanted to step in and say, yeah, tough job.

It's really tough, and I was saying to my team yesterday--

and I think I bore them with it--

I see us as the trailblazers in service and service development design and response, because we are doing everything brand new. So market reform for response, how do you design a service? What documents do you need? Like it sounds probably quite boring, but you've got to get the basics right.

At what point do you start telling industry, this is what we think the service looks like?

Because there's loads of questions we can't answer.

And we didn't get it right the first time, and I think we're still learning as to how we do this properly. And by properly, be as transparent as we can, give all the information behind why we're doing things at the right time. And I think we've, with the first--

and we're constantly trialing.

We use the phrase learn by doing, and I think people do tend to get frustrated with that at times, because they want to know the answer. And they want some firm ground to stand on, so they can start developing their systems. And then what service can they provide, and we don't always have that strong answer.

We might say, yeah, it could be this, but actually, we then change it. Because, I don't know, there's loads of different reasons why it might change, and it could be down to modeling, and I'm going to look at Neil again. Or it could be down to market design.

So I guess I was going to somewhere in that narrative say that we should probably mention the FRC, how the frequency risk and control reports. Which was a step change in our frequency risk management that's something that's new over the last couple of years and essentially looks at the balance of the risks that we take on the system, and how they compare against the costs, and to find the balance of what's the right value for money for the consumer.

So in layman's terms, that means we're not going to spend a billion pounds a day on frequency response, because it's not worth the risk when we could spend a lot less. So you have to balance the amount of stuff and the type of stuff you buy with the risk of the lights going off, because we haven't got an unlimited pool of cash. Is that what you mean?

Yeah, so looking at how far are we willing the frequency to go from 50 hertz and how quickly you bring it back and what are the sorts of scenarios and risks that might lead to that. For example, I think the last iteration of the frequency risk and control report looks at simultaneous events. So how do you define a simultaneous event? What are the sorts of events that we're worried about? What's the likelihood of those happening?

If we were to buy a response to control them, how much do we have to spend? And then based on that, looking at the scenarios and the risks and the cost, make a decision in terms of what's the appropriate level of value for money, and then consulting on that, agreeing then that and that becoming really the guidance that dictates how much we buy of x or y, and how we secure the system, and what risks that we manage, which do we have to live with.

OK, cool.

I want to ask some specific questions about the new frequency response services.

But first, I want to ask about one of the things we've done a piece on recently was MFR.

MFR gets brought up all the time. And I think it'll be really interesting to hear from one of you guys, what is MFR? And why do we have it? From the horse's mouth, if you like.

Yeah, so that's me. I think.

Because it's a lurky old--

there's a whole mist around this thing.

Yeah, so Monetary Frequency Response is a CUSC obligation. And so for generations--

CUSC being?

So grid connected assets, obligation to provide, monetary frequency response.

And we procure that intraday.

Yeah, so procure and PSH, very similar to FFR. I think there's slight differences in terms of the ramp rates slightly different costs associated with it. We need repositioning and holding availability utilization.

Some similarities, some differences--

I suppose different requirements to enter it, and therefore, different participants typically. But it is often our alternative cost, which we use to set what we pay for firm frequency response in some cases, DC. Because DC can compensate for elements of it. So is an important factor in consideration of how we procure alternative systems, because that is typically our reserve, if we haven't got sufficient response in other markets.

So would do you like to buy all of your frequency in advance--

or your frequency response in advance? Or will you always want to save some for mandatory frequency response?

So it depends really in terms of the value, proposition, and what's the right balance. I think there needs to be some intraday procurement. Because a day ahead stage, things can change in terms of inertia, loss sizes, levels, demand, forecasting, et cetera. So there's always going to need to refine to some extent. So I think the long term ambition is that the new response services would probably move to intraday and that we procure them through at an evolution of the current mechanisms such that we can do that, but we've started it at a day ahead basis and to build that up and with the likelihood that it will evolve in the future.

Which makes sense. Everything is going close to real time, and so it should, right?

Yeah, I mean Clean Energy Package is why we had to start all this really. I mean, we wanted to test the day ahead procurement anyway, but the Clean Energy Package dictates that we buy--

I think, it's at least 3% a day ahead. So when we launched dynamic containment, we had to be--

we had to do it in line with the Clean Energy Package. But we've talked about in the just a bit of a drop in on one of our documents--

so the market's roadmap to 2025, that went out March this year.

We'll link this in the comments somewhere.

Definitely, yeah. Definitely worth a read. So in there, we talk about what Neil just mentioned about intraday. So we'd like to see new response intraday, which should reduce how much we buy for our MFR. When we do that?

QUENTIN DRAPER-SCRIMSHIRE: Yeah.

I don't know yet.

But it is the principle's there, which is that--

FAYE RELTON: We know where we're going to.

Yeah.

Yeah. It's what's the right time and what are the trigger points for when we move to intraday.

OK, I want to talk about the new suite of services, which have by and large been a huge success and been a huge success for batteries too. We've seen so dynamic containment, dynamic modulation, and dynamic regulation--

these new three services, National Grid is buying a lot of DC, which is the post vault fast one.

And then they're not buying so much DMDR at the moment. So in some days, it's 0. And so I wonder why is that?

Why does the National Grid buy hundreds of megawatts of each? Now, why not just go pedal to the metal and buy it all? Because--

well, after you.

So there's a few things to unpack there. I guess so, there's the size of the market for start. So we've got I think about 1,400 megawatts pre-qualified in DC.

All the assets that are participating in the DR and DM at the moment are also DC assets. Our volume is a requirement is often higher than the volume we see participate in DC's area.

I think in recent months it's been max around a gig, certain [INAUDIBLE]

for periods. So DC at the moment, we value more because we're using that to offset primary, secondary high, so MFR or a short falls in the FFR.

Plus we also need DC for the speed of response. So that DC will help us mitigate the rock of losses that I mentioned earlier. So we need that fast-acting service, so that we can broaden our risk portfolio, as we've done through the FRCR and therefore, save money essentially by containing the frequency at a wider band than we used to say.

DC is valued higher at the moment.

We're not offsetting any services with the R&DM.

And the reason that we're doing that is because when we launched it, it's a [INAUDIBLE]

product. It's new. It needs to be implemented in our control room.

We need to see how it grows, how it performs, how assets react, and there needs to be some assurance and data built up to enable us to increase the volumes that we procure and then replace the legacy services or offset things like MFR and FFR before we do so. So some of these things, we've done already. And so the assurance work we've done internally to enable us to do that. Other things are coming up through the consultation and will hopefully improve the visibility of the assets--

so metering, also functionality, such as disarming non-BM units.

So if we are seeing issues with the behavior of assets, then we can act accordingly to mitigate any risks. And then there are the things, such as the frequency measurement specification or--

Standards.

Standards--

[LAUGHS]

--to be implemented in the next response consultation, which we have established a criteria internally that these things need to be in place, so that we've got the right assurances that we can lift the volumes, we can start offsetting other services, and then decrease the amounts that we pay in the legacy services, so that these are our future markets and build them up. But then, that needs to be balanced with liquidity, and assets into the market, and managing that pipeline.

So we kind of need more DMDR assets to show up to get the ball rolling before we can go out with the old and then with the new?

FAYE RELTON: Yeah.

But everyone's in DC, because the price is so high and you don't have to do much until you do. Right? Because the energy throughput requirement is so--

Yeah, so it's a balancing act. So equally, there's a relationship with FFR, so assets need to move across from FFR to DR and DM, DC. And we need to make sure that we've got the right assets in the right place providing the right service to manage the system at the right time. So that's what we're working through at the moment. And we've begun the transition of reducing the FFR requirements and as a consequence seeing increased volumes in DR, DC, and DM.

So as we work through those milestones and the market grows, then that's how we intend to transition out of. And I should say, probably a point of clarification, because there's a lot we talk about internally. I think when we talk about FFR, usually we're talking about dynamic. But there's--

Oh, yeah, yeah.

NEIL MORGANS: --two parts of FFR, which I'm sure that the other one you're less interested in.

Static is. [INAUDIBLE]

static, non-dynamic.

NEIL MORGANS: No, it's called static.

Oh, [INAUDIBLE].

NEIL MORGANS: Yeah.

I think also it's worth with dynamic FFR, we can't just turn it off from a provider perspective either. So we've been speaking to providers about--

current FFR providers about when can you move to the new suite. And it takes time.

Yeah. Especially, with the pre-qualification of the new suite. Pretty tricky.

FAYE RELTON: Well, is it?

Yeah, well, well, it is. It is.

FAYE RELTON: Got any single market platform, it should be a bit quicker, should a little bit smoother? [LAUGHS]

I mean, physically to do it, right? Because of the metering and the data flow.

And Testing.

FAYE RELTON: Yeah.

That's tricky.

And I think we--

especially during COVID, independent technical experts were quite difficult to get hold of was the feedback I was getting at the time. So I think everything had slowed down.

So yeah, I think what I'm conscious of as well is we've got all of the requirements that Neil's talking about. And we've also got from the external perspective, how quick can you move over? Why are you not moving over yet?

Is it because you like the steady stream and the mentally tender in that guarantee? There are some providers who don't yet feel comfortable about moving to day ahead. So we can't just turn it off. We need to get it done. We want to move to our new services.

That's why they're there to be used. But it's going to take time. And we want to do it, so that we're bringing people along with us at the same time.

Can I ask you, what does the control room think of the new services? Because--

FAYE RELTON: They love DC.

They do?

FAYE RELTON: Yeah.

Cause they can--

can just go the beach.

FAYE RELTON: The first day that we launched DC--

No disrespect for control room. We love you, sorry.

I mean, if they could go to the beach, they'd be very happy.

I did a night shift a few months ago, and I tell you what--

it's a difficult job. You've got to really know your stuff. And he's literally second by second. So it's very exciting.

But, yeah, it looks like a tough job.

But when we launched DC, I think 1st of October 2020, I think we got 89 megawatts. I think we had to use DC on the first day.

And I can't tell you the other times, but I think it was quite soon we were using it quite frequently. So the feedback we've had is that it's great that DC is there.

But I think, as Neil was saying about DM and DR, we're not offsetting yet. So I don't think they're using it in anger.

We've got the volume cap of 100 megawatts. We need to do this sensibly. They've got their DC tool.

Just explain the volume cap for a second.

So we'll buy up to 100 megawatts. And I'm just going to check with Neil, so I'm going to do a look to the side. 100 megawatts of high, 100 megawatts of low for each service.

Yeah, so typically, I require it's 80 megawatts at the moment. And then we've got an over holding part of the curve. So we might come onto that later in terms of the changes that we made to the buy curve. So it's usually 80 or 0, depending on our forecasted DC requirements and our latest view of the liquidity in the market.

But when we were designing the product and getting it through all the internal governance, because of the controlled manner we need to release it in. We're not just going to suddenly start buying the max, which we've said in our markets roadmap, 300 megawatts per service. We wouldn't just start buying that straightaway.

So if we limit the volume, we know that we're doing this release in a safe way. And we can still monitor the performance and see how the service is being provided, whilst ensuring people in the control room are comfortable.

Yeah, especially with DR, because it's prefaults around 50 hertz.

So my previous role in the quantitative analysis team, a lot modeling around the potential of oscillations, and if providers don't respond as we expect them to how could that influence the frequency and what's the consequence of that. So it needs to be a stage process to have the right assurance in place. And for us to work through that in the right considered manner.

Here comes my opinion.

I still miss the state of charge management bands. And I remember there was a little modeling done. National Grid was worried about herd behavior, right? Because everyone managed to take charge of around 50 Hertz.

If everyone uses that time to manage static charge at the same time, then you're going to get this herd response, and you're going to flip frequency either way. I felt there were a few ways to get around that.

Yeah.

One of them was like you do a random number generator and you have a delay where people are to respond to a delay or something. But I still feel like we're leaving something on the table with state of charge management, which means that we're not getting the most out of assets. We could eke more out of them. I think that's a real shame, personally.

But that's just me moaning. And I don't moan very often about National Grid. But I get the rationale, but I feel like we could have gone further down that rabbit hole and figured something out, which meant that the batteries don't have to manage state of charge out of service in a way that they do.

Yeah, I don't think the door is closed to improvements in state of charge. And we've got people working on state of charge actively. And if you have ideas, then we'd be happy to hear them and consider those as potential improvements we can make in the future. So the performance monitoring is quite complicated and a fairly dry subject to get people engaged on and interested in until you see the consequence of that sometimes and the results.

I mean massive. So can I ask you about the claw, or the clawback, or whatever? What happened there? Because I think it's really interesting. I'm sure lots of lessons were learned. And it'd be interesting to know, yeah, what happened.

Can I start with the trailblazing part, because I think we had a lot to learn on performance managing

I'm

not blaming it.

No, no, no, not at all.

QUENTIN DRAPER-SCRIMSHIRE: I'm just interested, yeah.

There was a lot to learn.

QUENTIN DRAPER-SCRIMSHIRE: Yeah.

So I think that's why with the first year, we didn't apply the penalties. And I hope that we were able to explain that sufficiently that there's a lot for us to develop and learn internally. But we are applying them now.

And Neil's more focused on the data itself. So I will look to you on that bit.

Yeah, so it's not--

It wasn't something I was part of directly. But as I understand it, there were a number of issues in the data that weren't necessarily poor performance. There were issues with switching between the services.

So for background, batteries that provide DC--

oh, sorry. We're agnostic--

whatever. We don't say batteries, we say--

FAYE RELTON: Technology agnostic providers.

Basically, batteries that provide DC, they have to give performance data to National Grid to show they've done what they've committed to do and they're getting paid to do.

NEIL MORGANS: Yes, that's right.

And in that data, there was some things that you guys weren't expecting. And then I'm going to leave to you now.

Yeah, so huge volumes of data. So it's 20 hertz data that's coming in and needs to be processed. New experiences for providers, the system that was set up provided some automated feedback, I think in terms of errors that it saw, but weren't always understood or reacted to--

and just little things, like I guess, timestamps being slightly off--

or, as I said, a transition between one service to another and what that looks like in terms of how you respond to frequency.

Clocks changing, things like that.

Things you need to work through and refine. And I think because of that, what went out came as a surprise to many. And it was appropriate to work through those and to make the right decisions to encourage participation and to grow the market.

And as I said, all the evidence that we've seen in terms of how the services performed when we've needed it have been really positive. So it was really a lot of process issues or refinements around how do we make this work better for everyone. And to an extent, you need to work through having some actual data and doing that and running the process to be able to learn from that.

That's the thing. You guys ask for so much data when you design a service. And it's like, oh, we've got this data, what do we do with it, right?

FAYE RELTON: Yeah, I think I remember when we launched DC. I had to learn about performance management really quickly. And I didn't understand why we needed so much data immediately. But then when you realize what the service means and the reasons behind why we need it, yeah, you then got to learn how to manage the data.

And I think we were both learning industry and ESO on that.

Yeah, for sure. So some batteries didn't technically didn't perform, right? [INAUDIBLE]

in some cases, it might have been at performance, some of it might have been a technicality. So what happened in the end?

That I think it was--

so we're not part of the settlements process.

QUENTIN DRAPER-SCRIMSHIRE: Yeah.

So, as I said, I think it's true that the reasons for the penalties understood them and make decisions on that basis as to whether the appropriate or not. But I couldn't tell you in terms of provider in settlements and what [INAUDIBLE]..

There were a lot of phone calls. There was a lot of what do you provide, how does it work.

Yeah.

And there was the decision to not apply penalties between two certain dates. And then we've been working--

well, the settlements team especially have been working very closely with providers with their account managers to work through like are you providing the right data, where are the real problems. So I think the time it takes to switch between services is still a challenge.

If you get about a bit off for acceptance in the BM and get turned up or turn down. And there's a natural ramp rate to that, because your inverters can't respond. So you don't do a straight line ramp. And then you've got to do DC around that ramp.

And you're only doing meter--

you're doing meeting at 20 hertz. I can see--

As a maths problem, you can see how everybody could be doing the right thing and it looked like the wrong thing.

FAYE RELTON: Yeah.

And I think that's with FFR, as well as MFR.

QUENTIN DRAPER-SCRIMSHIRE: Yes.

Yeah, yeah.

So I don't know the details of what's happening with that in this consultation. But I think we're looking at that for release one.

QUENTIN DRAPER-SCRIMSHIRE: Oh, nice.

I will confirm that.

talk about release one for a second?

FAYE RELTON: I'd love to.

It needs--

it needs a way sexier title.

FAYE RELTON: Yeah.

What's the vibe with release one?

I was just contemplating whether I need to say anything else on the performance monitoring before we moved on to that. Sorry.

QUENTIN DRAPER-SCRIMSHIRE: We can put a disclaimer in the show notes.

[LAUGHS]

No, I've just got to say, I'm really glad you guys have been--

I'm so glad we're having an open, honest conversation. It's brilliant. I'm really, really enjoying this.

Yeah, so what I was going to say is that's above and beyond building systems to manage these huge quantities of data when you experience these unexpected errors--

so providers haven't submitted data they were supposed to or they thought they had, and they hadn't, or the timestamp was wrong. Working around that and getting the data into the systems, finding a mechanism to get that data retrospectively, to understand it, coordinated it, to feed it--

it takes a lot of work.

And there's been a lot of hard work by lots of people in the settlements team and the quantitative analysis team to go through the issues with providers to make sure that they're understood, and explained, and to really strengthen understanding of how performance monitoring works and to make sure that we're in a good position to expect the right sorts of data from all the providers in a timely fashion, and so that we can avoid any issues that we saw like that in the future. Sorry.

Absolutely.

Well, I think that helps with release one. It's quite a nice segue. Because we're not--

because we're doing everything brand new. Nothing's perfect yet. And I'd never go out there and say it's perfect. I'm really proud of what we've achieved.

It's been really hard. But we've got three new services out there, and we're using them, which is excellent.

And release one, I mean, what could you call it? We've toyed with different names. One of my team wants to--

maybe we could use trees. And then I said, well, no, it's got to make sense in terms of frequency response. And then we started talking about, oh, could you call it hertz release?

I don't know, and then it just got a bit silly. And then we could go out to industry and say, what do you want us to call the release. But in the meantime--

Let's give to the--

who's the tree person? Let's just give it to the tree person.

No, because then we get feedback that we weren't--

it wouldn't make sense. Why are you talking about trees? So it just I just made the call. Let's call it release one.

And then in the future, if we can change it and it's more exciting, we'll do that.

OK, cool. Strong support for a new name for release two.

FAYE RELTON: Yeah, OK.

Unless you solve all the problems with release one.

FAYE RELTON: [INAUDIBLE]

postcard, as always.

Yeah, yeah.

Everyone will know by now, I always welcome feedback. So if you've got suggestions, send them in, Faye.Relton@nationalgrideso.com. So release one, so we've got a consultation. And this is a huge part of what my team does. EBGLs, it's the energy balancing guidelines.

And within that, there's an article 18. So you have to map all of the changes to this article. And you have your contractual terms, the service terms. And they all go out to industry and people have to review them.

Some context here, EGBL, that's European guidelines, isn't it? That's kind of linked to Project [INAUDIBLE]

and some other things people have heard about. [INAUDIBLE]

stuff.

It's one of few European guidelines that we've had to keep and we abide by. It makes sense. It's a useful tool, because it means that we as the ESO have an obligation to formally consult industry, which is good. The timelines can be restrictive.

But it's kind of--

I've got to the point where it is what it is. So you've got one month industry consultation and then after, you have two months to review. We just plan it in now.

And there's a lot of sensible stuff in there, right? Move stuff to like the day ahead or real time, you got to buy--

Well, that's the Clean Energy Package that says to do that. And yeah, they're sensible guidelines for us to work by. I quite like having boundaries.

So if there's certain things we have to do, we'll do it. And then if there's fun things we can do as well, we'll do those. But it's all about prioritization, of course.

But nothing about naming from trees in there?

Which is why that's not going to take priority right now. But if we're told it should do--

Which tree would have been if--

I think we got to beech. And then I was like, well, if someone misspells it to B-E-A-C-H. And then it'll be like why were you going for a holiday theme.

QUENTIN DRAPER-SCRIMSHIRE: Needs to be like an elm or something.

Because it's the summer.

Where you can't get it wrong.

Well, and then I went Harry Potter, because of the Elm Wand.

QUENTIN DRAPER-SCRIMSHIRE: Yeah.

It just got too much. And then I think that was all in 5 minutes.

So this release is like National Grid ESO saying, right, we did this thing. We released it. We've learned a lot of stuff. We're going to talk to industry and figure out how we make it better, right?

Then we're going to put it together in this release. What are we going to make better?

I have high hopes. So my view of this year was that it would be a bit slower. We could chill out a little bit. It would be less intense, because the past two years have just been mad. It's not going to be like that, because it's always busy and there's always surprises that come our way.

But the idea for this release is that we standardize the documents and merge them. So I think there was a point where we had 56 live documents out. And my team were just saying this is ridiculous. Providers were saying, well, which version am I looking at? So the first thing is we'll consult annually.

And it's feedback, we've been working with forums, such as the ADE. And they've been saying to us for a long time, slow down, tell us the structure that you're going to follow, so we can predict it. So we're going to do that. So we'll have an annual cycle.

National Grid [INAUDIBLE]

told to slow down by industry.

FAYE RELTON: Well--

You heard it here first. I love it.

I also get told to speed up so.

I'm trying to do both. But it was quite difficult. But I think the slow down bit is the talk to us. And this is industry as a whole, talk to us.

Let us know what you're doing, so we can understand it. Try not to surprise us, which makes total sense. So annual cycles. So that's [INAUDIBLE].

Where is it like Christmas, it comes out or is it summer?

So this is the first time, we're doing it. And probably, it's going to change a little bit next year. So the consultation we're launching on the 28th of September, that's what I'm holding us to and I really want to stick to that date.

Next year, the dates will change ever so slightly.

And I think we move--

I think it'll be a little bit earlier in the summer that we launch the consultation, because I suppose formally it's autumn by the time we've launched it this year.

We need to align with the C-16 consultation, which a different team run and that's another annual cycle. So ideally, we run the changes, we see--

What--

so sorry. What's C-16 in a nut shell?

C-16 covers a lot of stuff, so frequency response is just one part of this consultation.

So in theory, we get all the changes for response from the EBGL consultation, and then we feed those changes into C-16.

It's not a surprise. And it's an easier, yeah, we already know about that. That's what's changing.

So that's what we're trying to stick to--

annual cycle, ahead of the C-16.

We can have a product backlog, which we're working on, so people can see what we're working on, what's prioritized.

In terms of what goes in each cycle, it depends what IT changes need to be made, how much it's going to cost, how many people need to do it. And really importantly, if providers are saying to us, for example, performance monitoring is really frustrating, you've not got it right, you need to change this, we'd look into it. And if it's having a huge impact--

cost provider experience perspective, that would go in.

So this year, we've got--

I think you've already gone through a lot of the stuff, some IT changes that we're making. We don't expect the IT changes to be hugely impacting on providers. It's more about visibility for the control room around metering.

Frequency measurement standard, that is something we don't know what the is going to be yet. So we are going out to speak to industry. We're doing some roadshows, which I feel it's quite '90s roadshow.

QUENTIN DRAPER-SCRIMSHIRE: When we got the email here, we were absolutely buzzing.

Oh, were you?

Yeah, yeah.

I was a bit embarrassed. But then it was also another day of the trees, where I was like to call it roadshow.

Because doing one up here, one down in London, right? And you also doing the webinar one.

FAYE RELTON: And we're doing Scotland.

And we're doing Scotland? We don't want to make the Scotland one. We're going to be there. The other three--

NEIL MORGANS: Are you signed up? The first one is next Tuesday, isn't it?

FAYE RELTON: Yeah, we're in London next Tuesday.

I think it's great. You're doing it face to face.

FAYE RELTON: I can't wait.

Yeah.

Yeah, I mean, when we met Alex on the stairs. I've just met him on camera before, so it was really nice to actually see someone face to face. I'm really excited about seeing our providers face to face. Because most people I have actually never met in person, unless I met them through the capacity market.

Anyway, so we're in London next week. And it's going to be generally quite informal. We're having a big room open. People can come talk to us about performance monitoring, state of energy--

QUENTIN DRAPER-SCRIMSHIRE: Yes.

Future roadmap.

QUENTIN DRAPER-SCRIMSHIRE: Yes.

IT changes, APIs, single markets platform, what we're doing about our future, auction platform. It's very early stages, but we'll chat to you about it. Is there anything we're doing wrong that you don't like, would you like us to communicate better, differently, more frequently, less frequently.

Alternative costs, buy curves.

FAYE RELTON: Absolutely, all of that. So there's going to be a lot. And we just want to give the opportunity to have conversation. But we're also offering formal 1-to-1's, because some people will want to talk to us privately about some commercial stuff, which is also what we're really excited about as well.

QUENTIN DRAPER-SCRIMSHIRE: Cool.

So we're doing London next Tuesday. And then, I think it's Warwick and then Scotland. And I should know the dates. And I don't. I'll get in trouble with Kashia for it.

This is going to go after those things. So if you're listening to this, sorry, it's already happened.

Yeah, I hope you came to London ones, but you might not miss all of them, or will they?

Yeah, I think one's mid August.

QUENTIN DRAPER-SCRIMSHIRE: So, yeah, you can get a flight Scotland.

Quickly, quickly come to Scotland.

Yeah, so can I--

it's like when you know Jools Holland on New Year's Eve. And it's obviously not New Year's Eve, the hootenanny. It feels like that moment.

FAYE RELTON: But more exciting.

Way more exciting. Yeah.

Hopefully, there's no Tom Jones's this year. So what are the hot potatoes. You mentioned a few there, but what are you guys most excited about?

Excited or nervous?

QUENTIN DRAPER-SCRIMSHIRE: Yeah, love it.

Frequency measurement standard, I'm a bit nervous about. Because we don't know--

we don't know whether it's going to be--

because we're still building it now. And we need to speak to providers in these roadshows. So we need to find out is it just a small change that's going to be OK and manageable, and we can launch this change on the 1st of April next year or is it going to be something that we'll need to apply a grace period?

It's a dependency on us. So we're talking about stepping away from dynamic FFR. We need to get this standard in place for the new services--

Prior to April 1st, when we made the change to the buy order, we had a linear buy orders is such a fairly static price. We started using DC in a slightly different way. So before, whilst we were using it to manage the [INAUDIBLE]

that I talked about earlier, we were also using it to offset primary, secondary high above something that's called our minimum dynamic, so response that we hold in case of a large loss or the consequential rock off. So that changed how we priced it essentially, because it's got a different alternative cost to what it did before.

And then we also saw issues in terms of the use of hockey stick bids, which hopefully I don't have to explain in great detail. I have done a webinar presentation on the ESO data portal on that, which people can find and watch if they're interested in going to detail. But it's actually small bids that were intended, because of the particular auction to try and set the clearing price by speculating what our price curve would be and trying to sort of bid within that to set the clearing price higher than it might otherwise be if everyone's bidding at the margins, of course.

As background here, what we're talking about is in dynamic containment--

it's mainly dynamic containment, isn't it--

DM/DR? But in that service, some megawatts are more valuable to National Grid than other megawatts, right? So if National Grid doesn't have any DC, the first few hundred megawatts are really important.

And if National Grid's already got a gigawatt of DC, then the next 100 megawatts isn't so important to them, because they've already got a gigawatt. So they're going to pay less incrementally as they hold more. Is that correct?

That's right. And there are different alternative actions as you sort of move down, so you'd price it against those.

So that was the motivation behind the elasticity, I suppose, and the change to the price cap. And then over holding the hockey stick bids that we've seen because the way the current EPEX system is set up with the [INAUDIBLE]

algorithm. And it's not possible to overhaul. So we sort of paradoxically rejected bids. PRBs.

PRBs, guys. Once this discussed when paradoxically rejected bids started appearing in National Grid ESO webinars.

NEIL MORGANS: Yeah.

FAYE RELTON: This was new for us.

If you ever need tree name for something, it's this.

[LAUGHS]

So there's definitely go PRBs, because I think I can say that again.

Yeah.

So sorry. Just be clear, that's bids the National Grid rejected in DC, which don't make sense,

The algorithm rejects.

Hence it's--

sorry, the algorithm.

NEIL MORGANS: Yes.

So it's paradoxically, it's like, it was in the money. It should have been--

I say should have. It's odd that it wasn't accepted, and there's a reason in the algorithm, rather than the rules.

Yeah, so we're trying to fulfill our requirement. We step in through the bid stack. There's a large bid that doesn't fit within our requirements, therefore that's stepped over until it finds a bid of the size that it could take. And then fits that in, that's the higher price and therefore raises the clearing price of the auction. And we have, as a consequence, a PRB that was rejected because of the volume or lack of containability to be able to fit within that curve.

So by introducing an overhauled section of the curve and a stepped section at the end, which has got increments that will essentially mean that we can evaluate more options in terms of clearing the auctions. So rather than having to take the smaller units, that higher price to fulfill that requirements, there's a higher chance that a lower priced unit would be able to fit in some section of that curve, and therefore provide a more efficient auction outcome.

And I'm going to try and do this like--

I try and do things like a five-year-old. And I'm going to try it here. So what National Grid here had was they would buy--

let's say 100 megawatts. They would buy 100 megawatts of this DC service, and they couldn't over hold. So they couldn't buy more than that, which meant they were skipping some cheaper ones, because they would have pushed you above the 100 megawatt. And what you did is change the algorithm, so you could over hold a bit and that meant that hopefully, you'd have a bit more rationality in the auction.

Is that right?

It's not really changing the algorithm.

QUENTIN DRAPER-SCRIMSHIRE: Yeah.

It's--

well, almost. It's not changing the algorithm. It's changing the curve. So--

QUENTIN DRAPER-SCRIMSHIRE: Changing the curves.

We previously would have come down vertically. It now steps down, and those steps can accommodate a sort of broader range of sized bids and therefore will clear more efficiently.

So as a provider, you want to hit the steepest part of the curve then [INAUDIBLE]

just as it drops off the hockey

Right, yeah. If you're hockey sticking and it's not liquid, then you're looking to hit that. So that--

well, it depends on the volume in the auction, but yeah, typically, you're trying to estimate as high as you can go. And that's what we're trying to mitigate, which, it's still problematic when we don't have enough liquidity in the auction. But when the auction is liquid, then the [INAUDIBLE]

section should reduce PRBs and mean that we're clearing more efficiently.

I guess--

so it was--

it's part of the national grid is constantly is trailblazing. And there was an inefficiency in the market.

FAYE RELTON: Got you saying it now.

Yeah. Do I have to pay you like a quid every time for this?

FAYE RELTON: Yes, please.

You should pay me. What's going on here? I'm so under your thumb.

So yeah, so there was an inefficient kind of set up in the market, and it's become more efficient. That actually means that prices aren't--

well, apart from the fact that there's loads of other impacts on prices, but given the same say under the new market arrangement since the 1st of April, prices were likely to have been less.

But then, of course, we've got at times over 100 pounds of megawatt an hour prices right now. So what do you guys--

I know you don't talk about prices, but you do--

in this new world, you do have price caps, which are really interesting. So what's the thinking behind price caps?

I'm not--

I don't want you to comment on whether they're high or low. I know you can't answer that question. But how did national grid come to

the-- or National Grid ESO come to the decision that you were going to set price caps? I think that's a really interesting discussion. I would love to be in a room when that was decided.

So it goes back to--

QUENTIN DRAPER-SCRIMSHIRE: It's changed the market.

Yeah, so it goes back to alternative action. So what are we buying here, and where can we get it elsewhere? Are there trades or Bohr's or other actions that we can take which would cost less?

And therefore, if we use that to set our price curve and we secure the volume within that, then we're comparing that to what the alternative is and saving consumers money by procuring it cheaper than we'd have elsewhere. So that's what it's driven by. And there are a lot of interactions between the different markets and different exchange rates, and some are substitutes for some in certain conditions and not in others. So it's a fairly complex web of related considerations that you have to make in terms of the alternative costs. But that's essentially what sets what our price cap is is how much do you think we have to pay for it elsewhere and therefore what are we willing to pay for it in this market if we can secure it here instead.

Makes sense.

And I want to talk about Day Ahead Store oddly for a second because Day Ahead Store happened first, didn't it? And we learned a lot in that auction, in that market, that's been applied in DC and DR and DM. I wonder whether you guys were involved in that and how you learn stuff and what lessons you brought forward because I think that's really interesting.

So I wasn't involved in the move to Day Ahead Store, and I imagine Faye wouldn't have been either because--

I installed the project.

Oh, well!

I was asked to do DC and Store at the same time, and I quickly realized I couldn't do both. And it was a very--

I was there at the vague start of the project where you have to try and work out what to do and into some of the clean energy package. And then I went on to DC. So I was there at the beginning working out what systems we need to use and how we need to make it work. And that's where the decision then came to go to move to EPEX for response because we already had the weekly auction trial.

Yes, yeah.

FAYE RELTON: And we knew that we could use the EPEX platform as an interim solution, which--

and we could go pay as clear, which, clean energy package compliance as well. So I was key--

the decision making round, that was key. So we had some quick learnings there. But then I went on to DC and [INAUDIBLE]

Store. And that was my involvement until we then saw what was happening with Store and we were looking at, OK, how can we apply this to response and become more efficient?

I personally think Store doesn't get enough airtime. I love Store. It's so interesting, particularly now it's a day ahead market because you have--

I mean, sometimes Killingholme will set the price really high. You can get swings from like 20 p to like 20 quid. It's very interesting, and sometimes even higher than that.

NEIL MORGANS: Yeah, so--

At Store!

Our team runs the procurement process for Store as well. So I've recently become quite close to it. I think in terms of learnings and what we got out there, there are clearly some similarities, I suppose, in terms of the response from markets.

But then we also run it on different platforms. So Store's through the Salesforce platform, and there are some complexities that they don't suffer from in terms of the interaction between bids and assets to an extent to the other market. So it's simpler in some respects. And then therefore, it was, I think, more straightforward to build into some things like overhauling, underhauling, into the algorithm directly rather than the changes that we made in the response market because that's more complicated to work through the different welfare calculations of how bids are linked together and what the optimal outcome is.

I must interject at this point because I know if Sherry was here, she would say, yes, but Response and Store are very different products. So it's not just the systems. They are two very different services.

Yes. Yeah, yeah.

And you can't directly compare. So I just had to get that in because it's been--

QUENTIN DRAPER-SCRIMSHIRE: I don't--

I wasn't trying to directly compare--

NEIL MORGANS: Sherry's here in spirit.

NEIL MORGANS: You're listening to Sherry.

FAYE RELTON: I've been channeling Sherry because I know that she'd say that.

NEIL MORGANS: Yeah. Good point, Sherry.

And people often ask, why don't you use Salesforce for both, or why don't you use EPEX for both? And as with most organizations, you've got to have a look at legacy systems versus new investment, what makes sense to do it quickly. We were using EPEX for response anyway. It was quite a nice, quick move to do to bring the new services onto that as an interim solution.

It didn't make as much sense for Store, and it would have been a huge decision--

going to be shopping for another provider. We could build this.

FAYE RELTON: We--

We should be submitting these too--

NEIL MORGANS: --too late, I think.

FAYE RELTON: Yeah, well--

[INAUDIBLE]

FAYE RELTON: We're not announcing here. They are.

QUENTIN DRAPER-SCRIMSHIRE: Oh. I don't know who they are. So I couldn't announce anyway.

So I think--

QUENTIN DRAPER-SCRIMSHIRE: Oh, good luck to them, anyway.

We're towards the back end of the tender process now,

So now it's about, like, what are we going to do? what does the project plan look like? What can we deliver?

What services are going on there first? What services are going on there, which ones aren't. Loads of questions.

What's co-optimization? Great question. Loads of different answers.

QUENTIN DRAPER-SCRIMSHIRE: Can't tell you.

Well, I like stacking.

QUENTIN DRAPER-SCRIMSHIRE: Yeah.

And that could be part of. It so yeah, there's lots going on at the moment for that.

I want to ask you about two more things. I know we're going to run out of time.

But I want to talk about the BM, and I also want to talk about National Grid. I want to do the first bit, the kudos bit.

National Grid got pretty good at making data available, and I just want to say that on behalf of us, who use a lot of the data, and also a lot of our customers. Now, things have got a lot better a lot faster.

Is that because of Neil's data portal?

QUENTIN DRAPER-SCRIMSHIRE: It could be. Yeah, a big part of that is Neil's data portal. So yeah, thank you. You might want to do a quick plug.

What can you do on the data portal now that you couldn't do before?

NEIL MORGANS: What can't you do on the data portal? So--

It's a great portal nothing on it.

NEIL MORGANS: So yeah, I think--

big plug. Improves our discoverability, understanding we have data dictionaries. We've also got an API. You can curate that through SQL, which is very powerful and build that into the various Python scripts, et cetera.

You can subscribe to updates. You can do various queries on the metadata. So we followed some of the EDF, the Energy Data Task Force, guidance around things like doubling core standards. And I'm probably getting far too much into this, the detail of what--

NEIL MORGANS: --data portal is. But I'm very passionate about our data, and so are all of our colleagues, and there is a real motivation to share as much as we can and to get that out there and--

QUENTIN DRAPER-SCRIMSHIRE: Awesome.

NEIL MORGANS: That's been facilitated by the data portal, and, yeah, I think we've got close to 100 data sets on there now all accessible through the API. So it's gone from strength to strength.

Well, long live the data portal.

NEIL MORGANS: Yes.

And now it's a big topic, but I'm going to cover it. BM, right?

FAYE RELTON: Yeah.

The elephant in the room is--

so of course, we publish the data on all the assets in UK, battery assets and what they're doing. And yeah, the elephant in the room is there's very little bit BM expertise in batteries.

They're pretty much--

I mean, most of the time, they're either doing stuff in wholesale markets or mainly frequency response. And so what's the view on that? I mean, do we want to get more batteries into the BM and getting activated or accepted in the BM? I'm not going to go down the skip rope route right now.

That's a hot potato and a very complicated question.

NEIL MORGANS: [INAUDIBLE]

yeah.

Yeah, so what's the vision? Do the control room want to be dispatching more batteries? Is this good?

Would you rather you had batteries in frequency response than the BM? Or do you want a bit of everything? Where are we headed here?

The control room want to be dispatching assets that are suitable to meet the needs of the system. So I don't think they've got a preference in terms of this is the type of asset we want participating in the beam and different asset types of different attributes that are valued in different scenarios and manage different system risks. But I don't think we've got a sort of line on what we think batteries should be doing in the BM.

Yeah, I was going to go back to my technology agnostic because I don't think there's a specific view on batteries. I think everybody wants simplicity. Ideally, if everything was in one place and you could dispatch it and you could see everything was going on, that would be ideal, I think.

But it's so complicated. We've got legacy systems. We've got an operational environment 24/7 second by second.

So you can't turn anything off. So how do you make improvements and changes? There's a lot there's a lot to think about, a lot to do. But I don't know anybody who wouldn't want things to be simple.

QUENTIN DRAPER-SCRIMSHIRE: Yeah.

What the answer is, I don't know. I know that we've got a transformation program looking at the BM. There's been a lot of change on that recently. I don't know if you've got a little bit more info on it.

Well, there's a lot of engagement and a web page set up. I think they've done a lot of external webinars recently, and they've recorded a lot of the output. So I won't pretend that I know all the messaging behind that, but it's certainly available for those who are interested to go see what the future of our--

FAYE RELTON: [INAUDIBLE]

I have been before but not--

Because I think the balancing--

I think there's a big IT program looking at the balancing at the BM, and there's--

a lot of people went to Wokingham, a lot of external providers.

Oh, well, I've been before but not at this particular--

FAYE RELTON: I missed that.

--for a secret tour.

There's probably going to be more. I think there's a lot of--

it's a huge program.

want to see more batteries getting dispatched in the BM, This is just--

I can see the battery.

I just want to see it happen.

Anyway, I know it's just another plug for the [INAUDIBLE]. There's the dispatch transparency data set, which I think goes some way to explaining--

I know you mentioned scripts earlier, but I think we make decisions not just based on price merits or overall costs. So there are different reasons why you dispatch assets, and that could be constraint management.

There's also--

there's a lot of different ways to calculate skip rate. And I think one of the things we need to do as an industry is define it better because, yeah, we've done a lot of work into this, and it takes sitting around a table having a real think about how you can look through the data. So yeah.

Can I just drop in the operational transparency forum on a Wednesday, 11:00?

QUENTIN DRAPER-SCRIMSHIRE: Wednesday, 11:00.

--a lot of work that goes into that forum and data visibility, transparency, costs all of that. All of the really big, important stuff is being discussed there. And I'd recommend if you don't dial in already.

This is a webinar, right? Not a web--

FAYE RELTON: Yeah.

A call.

FAYE RELTON: Yeah.

No one says webinar anymore.

FAYE RELTON: --is, really. [INAUDIBLE]

We call it webinar. I don't know if that's outdated or not, but I think--

FAYE RELTON: It's a call. It's a webinar. It's online.

But where you can ask questions, and if we can answer them in the space of an hour, we will do.

Yeah.

And there's lots of expertise both on the call and behind the scenes who are lined up to do their best to be transparent and to provide feedback on the areas of interest.

I think it's been quite a successful one. It's a lot of hard work. But yeah, if anyone doesn't listen to that, which I'm sure

Was that the one that was started during COVID?

FAYE RELTON: Yeah.

Where it's like ODFM started, and it was like--

I think because--

QUENTIN DRAPER-SCRIMSHIRE: [INAUDIBLE]

--like, we had so many questions. I think somebody said, let's just do a webinar weekly so we can show what's going on, what we're prioritizing, what we're experiencing, and explain what's going on. And then it just--

it stuck, and it's got to be, what, two years now? Every Wednesday.

QUENTIN DRAPER-SCRIMSHIRE: Yeah.

It's a commitment.

Sessions are recorded. So again, we done some--

So we can--

so an omnibus. We can watch the last two years.

NEIL MORGANS: You can do, yeah.

--party.

FAYE RELTON: Well, there were some recordings we didn't--

I think we lost a couple of recordings. So you won't have--

there's--

QUENTIN DRAPER-SCRIMSHIRE: They'll be in the archives somewhere.

Yeah, that's because we were trying to work out how everything worked at that point, I think. But there's loads of stuff on there, and we've got--

there's a Q&A log. And I think if any questions aren't answered, they answer it the following week. So yeah, like I say, it's a big commitment, and those skip rates were discussed on there a lot

QUENTIN DRAPER-SCRIMSHIRE: Yeah.

They're definitely more informative than what we're going to be saying [INAUDIBLE]..

Dispatch transparency.

Apologies.

QUENTIN DRAPER-SCRIMSHIRE: Dispatch transparency. All right, cool, we've run out of time. I just want to say a couple of things.

Firstly, thanks for coming on. You're taking time out of your day to come to our head office and chat with us.

It's been an awesome conversation. We could have gone down some more rabbit holes, no doubt.

Yeah.

And maybe we'll do that in a version two. I'm sure there'll be loads of comments and feedback on this, and we'll pass them on. And yeah, until next time, thank you so much.

Thanks for having us.

Thank you.

Cheers.