Making Batteries Pay: BESS Revenue and Ramp-Rate Limits | Ben Wrigley and Suzanne De Naghel, Montel

Excellent, right, welcome everybody, the energy levels seem quite good, I imagine it's the coffee after the break we just had, yeah my name is Benjamin Wigley, I'm a energy analyst at Montel, and it's my pleasure today to talk to you about batteries and ramp rates along with my co-presenter Susanna. So I'm going to be starting off by talking about why do ramp rates matter, why are they becoming an increasingly talked about topic across not only the Benelux energy markets but also across Europe, but also seeing similar discussions happening in Germany for example. Then we're going to introduce you to our tool Flex Insight, which is really designed to quantify the impact of revenues of not only ramp rate restrictions but also other kinds of restrictions on a battery. For example, what if you have a variable grid connection that is sometimes curtailed, what if your battery is co-located, and so on and so forth. And then to finish up, we're going to present you with some concrete numbers from a case study we did with a Dutch battery to kind of show how the impacts of ramp rate restrictions vary by battery size and also by which markets you're optimizing your battery on. So let's start. Why do ramp rates matter? So what's the unique selling point of a battery? Well, one of them is speed, of course. Unlike other storage assets, a battery can charge or discharge at its maximum power within only a few seconds. Now this has great benefits for stabilizing and managing a system, especially with an increasing amount of renewables in it. But the speed of this regulation can also create huge issues for both TSOs and DSOs and others who have an interest in keeping the system stable. As well as this, we are also seeing increasing ramp rate restrictions in grid connection agreements. And we are also seeing more and more co-location occurring, especially as we've already seen with our previous speaker, talk of the grid fees in the Netherlands being very, very high. We've already heard the number 190,000. Very, very high. Very dangerous for a business case. But one of the reasons why ramp rates are becoming such a topic, especially in the Netherlands, is the battery build-out boom. The rate is very, very high. What you can see on the screen is from our own central price projection forecast from Montel. This is the number of large-scale batteries that we expect to see in the Netherlands until 2050. As you can see, at the moment we have about somewhere in the range of 1 to 3 gigawatts installed, not too many. But we expect to see about 10 gigawatts by about 2038 and close to 20 by 2050. And this is only our central scenario. The high decarbonization scenario has an even faster rate of battery build-out. So if you imagine all of these batteries charging and discharging and changing their position within seconds, you can see how it would become very difficult to keep the system stable over time. And an example of this is the increasing incidence of regulation state 2 that we are seeing in the system. So what is regulation state 2? Well, this occurs when Tenet is forced to activate both upward and downward energy, balancing energy, within the same 15-minute period. Why does this happen? Well, batteries want to earn a lot of money through the passive balancing system. The spreads are very, very attractive. So they see the direction that the system is heading, and they put themselves into a deliberate imbalance to try to help the system, and in doing so, earn some money. If you start to see, this works with a few batteries, but if you start to see all the batteries responding in the same way, you end up with a kind of tidal wave effect, where it overcorrects, and it can basically turn a long system into a short one. As an example here, I've got a, yeah, imagine a 300-megawatt battery, if you will, that's currently charging at full power. That's minus 300 megawatts, and then it suddenly flips to discharging at 300 megawatts. That's a swing of 600 megawatts in only a few seconds. Very difficult for the system to manage. And we're seeing this over time reflected in the number of regulation state 2 periods in the Dutch system. So in 2020, that's 2022 and 2023, we see the numbers are somewhat stable here, these purple lines. But then we get to 2024, 2025, and 2026. It's almost doubling each year, exponential growth. This is probably not sustainable for the system. And then we have the regulation side as well from Tenet. So the understanding that was previously widespread before this year was that your asset needed to be capable of being ramp rate restricted about 20% per minute for battery sizes over 50 megawatts. Now this was understood to just be something that you were capable of doing rather than something that would be implemented the whole time. But what we are seeing after a certain clarification from Tenet this year is that assets are expected to hold to these ramp rate restrictions all the time. And now the question that we're here to answer today is what impact will this have on your battery revenue case? And for those of you with smaller batteries than 50 megawatts, you're not safe. Because earlier in the year, there was a proposal to also regulate smaller batteries in a similar manner. It was ultimately withdrawn due to concerns about transparency and how to implement it. But there is the intention to kind of bring it back and to reform it in some way. So now I'm going to introduce you to our tool, Flex Insight. So a little bit of background on this. Montel has been doing battery consultancy revenue analysis for nearly five years now. In 2022, we saw a huge boom in batteries being built. It was relatively easy to get a grid connection. The banks weren't asking too many questions. They were confident in the business case. There weren't as many difficulties as we're seeing here today. Now we're seeing increasing instances of customers asking us, how can I save on my grid fees? Will my battery still earn enough money if I only have a 75% grid connection? Will my battery still make money if I have a TDTR connection? Or if I make use of time of use? And this resulted in us doing more and more sensitivity scenarios. And then it became a longer and longer process. So what we wanted to do is we wanted to put the power back in the hands of the customers and enable you to do your own runs. And only when you had the runs that you were happy with, and that worked for your business case, would you need to come back to us and we could validate it and help you through the due diligence process. So I'm now going to show you a video of Flex Insight, which I hope will work. Let's see. There it is. So this is an overview of all your different projects. And we have these various different tabs here where you can input different things. So the first thing, it's quite obvious. You have your capacity on the power. So your charging power and your discharging power. Then you have the energy capacity. So the storage duration of your battery effectively, quite simple. We also have the lifetime performance. So we have our own state of health degradation curves that you can use, or you can keep it standard, or you can input your own based on guaranteed state of health from the manufacturer as well. Next up, we have co-located setup as well, if you want your battery to be co-located for Solar Farm. But perhaps the most important for our customers are the grid connection things. So here you can see we have different kinds of grid connections, firm, non-firm, and also other things. But the most important thing that we want to talk about today are our new ramp rate limits. So as you can see here, we've now added to the tool the ability to put a ramp rate restriction on your battery. You can do it as a percentage per minute or per second. And by default, we start at the 20% here. So how have we modeled these ramp rate restrictions via Flex Insight? Well, what we've done is we've assumed a kind of a linear ramp rate restriction here. So on the top left, you can see an example of how a battery will start to ramp up during a quarter hour, assuming a 10% per minute rate of increase. So you start out at 10% and then each minute you go up by 10%. The issue with this is, of course, you only end up with an average power for that quarter hour of 70%. It means that your ability to respond to high or low prices in the market is slowed. Up here, you can see an example of a day, a real day, the 21st of May, 2026, that is. And here you can see the day ahead price across that day. You can also see a lot of sawtooth phenomena as a kind of leftover from the shift from hourly to quarter hourly trading. And here on the bottom right, you can see the differences in how a battery would optimize with and without ramp rate restrictions. So on the day ahead, it's not too bad, actually. So you are forced to kind of charge and discharge over a longer period of time. So you don't get quite to capture the amounts here. You can kind of see the differences in how you need to charge and discharge for a shorter amount of time without ramp rate restrictions, whereas with ramp rate restrictions, a little bit less. And the overall impact on revenues is pretty small. It's only 1.7%. Don't be too relieved. This is only day ahead. We're going to get onto the other markets a little bit later. So yeah, one more thing for Flex Insight. The idea behind it is you create a configuration for your ideal battery, and then you want to see all the different kinds of sensitivities. Or rather, the bank wants to see all the different kinds of sensitivities, right? They want to see the downsides of all these things that could affect your revenues in the future. Well, with Flex Insight, what you can do is you can rerun the same battery multiple times, much like a science experiment. And you can basically compare all the impacts on different revenues like this with a nice overview here. And then you can also download these revenues, put them into your financial models. And with that, I'll hand you over to my co-presenter, Suzanne. Yes. So now let's take a look at those results. Let's start with a simple standalone battery, a 10-megawatt, 20-megawatt-hour battery in the Netherlands, optimized on day-head, intraday, and imbalance. And we used the historical prices of the past year. We ran this with Flex Insight twice, once without any ramp rate limits and once with a ramp rate restriction of 10% per minute. The result, this battery would have lost 4.4% of its revenue the past year because of that 10% per minute ramp rate restriction. But yeah, that number, 4.4%, it's around €5,100 per megawatt per year. But that is just one number over the whole year across three markets. So I would say, let's take a deep dive in what we can find in that one number. So to investigate more where this loss of having a ramp rate restriction comes from, we did some runs with only one market. So only day-head, only intraday and only imbalance. And then we see that in our model, the day-head and intraday only had, the ramp rate restriction only had a moderate impact. But imbalance is really a different story. There, the loss in revenue caused by ramp rate restriction was almost 30%. Where is this big difference coming from? Yeah, imbalance revenue is coming from short price spikes. And having a battery that takes 10 minutes to reach full power, just lose a lot of that opportunity to move on these short price spikes that only last for a quarter or maybe two quarter hours. And they had this the opposite. It's about shifting energy from moments during the solar generation to the evening peak. So on that time scale, your ramping of a couple of minutes doesn't have that much impact. So another deep dive we could do is take a look at these two consecutive days that I found, the 21st and the 22nd of May this year. They look very similar on the first glance. So same sunny profile, same day-head shape, low prices at noon, high prices in the evening. And if you would only optimize your battery on day-head, they would probably, they probably, I calculated, they earn very similar revenues. And if you then would put on there the 10% ramp rate restriction, 10% per minute, on one day you lose 1.7% and on the other day 2%. So pretty small and very similar. But if you then put the intraday and imbalance prices into the picture, those days look very different. So on the 22nd of May, intraday and imbalance prices go all over the place with price spikes, with one price spike even going to 400 euros per megawatt hour. And yeah, next day, what started very similar ended up being a very calm, boring day. So then the difference is pretty obvious. So your ramp rate restriction on the first day, on that volatile day, costs you almost 7%, while on that calm, boring day, it only costs you 2.5%. And this is a good example of what I said before. Your ramp rate limits you not in shifting energy from cheap to more expensive hours. It limits you in reacting to those price spikes in those more shorter term volatile markets. And yeah, also good to keep in mind that how those price swings will evolve in the future also has an impact on how much a ramp rate limit will cost you in the future. So, so far we've seen one, only one example. So a 10 megawatt two hour battery with a 10% per minute ramp rate restriction, full grid access, but every project is different. Your battery might have a different ramp rate limitation, might have different size, another duration, even maybe restricted grid connections. So let's go through some of these what ifs. Let's start with the limit itself. So here we compare no limit, 20%, 10% and 5% per minute restriction. So the 20% per minute restriction is only 2.1%. And then we have again our base case with the 4.4% loss. And then if we make, if we would make ramp rate restriction even more stricter to 5% per minute limitation, that would mean a loss of 10.5% in this case. So what's also interesting is to see is that these losses are less than what you would have expect from our single market results. And that is because the modeled optimizer in FlexSenseight adapts. When imbalanced revenues are harder to capture, it will shift the battery and the strategy more to intraday in this case, because it was less affected by your ramp rate restriction. And you can even see the yellow part slightly growing if the ramp rate restriction is stricter. So yeah, that means that a multi-market optimization partially cushions the blow of your ramp rate restriction. Okay, so what if my battery has a different duration? So we take a one hour, two hour and a four hour, again with 10 megawatts, 10% per minute ramp rate restriction. So the one hour battery loses relatively more than our two hour base case, 6.4%. And the four hour battery loses less. That makes sense if you look at revenue mix, because a shorter duration battery, its revenue depends more on imbalance relatively, because it doesn't have that much energy to shift than if you would have a longer battery. So yeah, that's why the ramp rate restriction hits harder on those shorter duration batteries. Okay, so what if my battery has a different size? In Flex Insight, size does really matter. A larger battery is optimized different because we take into account market depth. A 100 megawatt battery simply can't capture the same share in the revenue in the imbalance market as a one megawatt battery without having an impact on the prices. So that's why in Flex Insight, a larger battery is optimized more conservatively. So the same thing happens as we've seen before. Yeah, so you can see that the smaller battery has a bigger purple part, so a bigger share of imbalance revenue. And what you've seen before, the configuration that depends most on the imbalance revenue, loses most with a ramp rate restriction. So finally, what if my battery has a curtailed grid connection? So we took 25% curtailment on both export and import limit, and then we saw that in the ramp rate restriction, the relative loss was the same for the unrestricted or the more restricted battery. But the battery with the grid restriction in absolute terms loses less because simply it had less to lose with to begin with. So it's always important to combine multiple restrictions your project has instead of assessing the impacts individually. Okay, so we've covered a lot of cases today, but the chance that your specific project was between them is pretty small. So maybe your project has a co-location solar, or maybe even different ramp rate restrictions, or a much more complicated grid connection. That's exactly what Flex Insight is for. So you can build your own battery configuration, including your co-location, grid constraints, and of course your ramp rate limitation assumptions. Then you can run your simulations, either against historical prices like we did today, or you can run them against Montel's long-term curve scenarios. And in the end, you can compare all your different runs side by side to really investigate what parameters your business case is sensitive to. Yeah, and that's how we actually built this whole presentation. So you can do it yourself, replicate it yourself with your own setup. So what have we seen today? A ramp rate limitation doesn't just take a fixed slice of a battery's revenue. It impacts depends on how your battery is making its money. Shifting energy from cheap to expensive hours is not impacted much. What the ramp rate really takes away is the ability to react quickly and to catch those price spikes. So a battery that relies more on these kind of revenues is impacted the most. That's what we've seen in all the cases, the configuration, whether we change the size or the duration, all the configurations with a biggest share of revenue from imbalance were hit the hardest. It also shows that a simple day-hat calculation will underestimate the effect of ramp rates limits. So it's important to have a multi-market optimization model that also takes into account other restrictions your battery is facing. And then we also saw that because the cost of having a ramp rate restriction depends so hard on these price swings, it isn't something fixed over time. More renewables, more batteries are entering the market, but also the balancing markets are changing and how that will impact those swings will also impact the effect of your ramp rate limitation. And then, yeah, so... Sorry. Yeah, so don't... It's not like... Don't treat ramp rate... So the key message... Sorry. The key message is don't treat ramp rate limits as like a small technical limitation. It really has an impact on how your battery can enter the market and what it can do. So, yeah, please take it into account in your business case. Yeah, thank you very much for your attention. Fantastic. Excellent, Suzanne and Ben. Thank you very much indeed. That was a great run-through and I loved the case studies and also how you can modify it for your own specifications. Excellent stuff. Any questions from the floor? Yes. Oh, any questions here? I'll take one here. Mende van Lusgaaf, independent advisor. I was wondering... This is new in the Netherlands, but do you have evidence of other European countries with fixed or limited ramp rates for batteries? The answer is yes. So in Belgium, we're seeing a less restrictive regime, but ramp rate restrictions are something being discussed by Ilya. And also our consultancy team in Germany is dealing a lot with ramp rates in that country as well. And the same kind of issues are kind of universal. It's not specific to the Netherlands. It's the worst in the Netherlands at the minute because of it being so advanced. It's a victim of its own success, like JP was saying before with renewables. The same goes for batteries. So it's not just a Dutch phenomenon now. Question here at the back. Femke Schuurman from Van der Bron. I have a question as well. So for the ramp rates, you're mostly tackling it from a financial perspective. Does Flex Insight also give an insight in, for example, number of cycles or kind of the depths you charge or discharge? Because I can imagine it in that sense, it has a kind of a positive effect on the battery lifetime. So maybe if you look at the whole business case, maybe... Yes, exactly. ...more positive? Yes, exactly. So the output of Flex Insight, we've showed some revenue curves, but you can really download the quarter hourly results of the modeled optimization we did. So you can investigate how much cycles exactly happened. You can even see when those cycles were reduced or not. So, yeah. Excellent. We have a question here. Yeah. Hi. Vincent there with Vadoval. The question, does Monta also have an outlook, forward curve for the rampage? Because the tenant is now proposing something very transparent or illogical and they are working on a system code change next year. What kind of assumptions should we take in our business case? Is 20% the best guess or 10 or 40 or something else? What's your... Do you have a view on that? Well, based on the existing legislation, we would say 20%, but it's kind of like the whole grid fee question. At the end of the day, there's not a lot of transparency on this and we're kind of looking to tenant and the DSOs to kind of tell us what's going to happen in the market. So, all we can really do is give you the ability to model different outcomes so that you're ready for whatever happens down the line. Hi. Ilke Rontuijs, Powerfield, happy user of Flex Insight. Thank you. If we take a look at these ramp rates for 20% of 20%, it's nice on... It goes, say, with our four inbounds, but especially for AFR energy, you can go up to 200% in your energy bidding in the Netherlands. Does Flex Insight... I haven't checked. Does that take... Can I put that in as well? Yes. So, it also takes that into account by the way it is modeled in Flex Insight. For AFR bids, it already took into account some kind of activation rate because you're... In passive imbalance, you can go all the way... Yeah, it's your own decision, but for AFR activations, we model it differently because it's very unlikely that you... Yeah, maybe not unlikely, but then at least it is controlled. So, it's like a different issue in AFR, I would say, than in balance. I think there are several people here, if there are any more questions on Flex Insight, so you can answer questions, I think, you know, after... When we serve the refreshments. But once again, please give a warm round of applause for Ben. Thank you. Thank you. Thank you.