With its digital platform Interconnector, EnBW operates its own virtual power plant. We talk to Simon Schweda from EnBW about who can participate, what this means for the individual and the energy transition, and why small-scale systems will play a big role in the future.
Virtual power plants connect smaller, decentralized power generators. What requirements does the system operator need to meet in order to become part of a virtual power plant? Is it also possible even with really small plants?
Yes, it’s possible! The smallest PV installation that is connected to our virtual power plant has an output of 5 kW, basically your classic single-family home. That is quite a rarity though. In Germany, there is a provision that requires systems with an output of 100 kW or more to take part in so-called direct marketing. This means that anyone who operates smaller PV systems that are subsidized by the Renewable Energy Sources Act (EEG) can sell their power directly to grid operators. Once the output exceeds 100 kW, the plant operators themselves are responsible for selling the power. That’s when people start looking for a virtual power plant, for example, that will take care of sales. But it’s becoming more and more attractive for smaller systems to voluntarily connect to a virtual power plant and take advantage of direct marketing.
Are there also technical requirements?
Yes, in order to be able to curtail the supply in certain situations, the EEG unfortunately requires that virtual power plants are able to control each system remotely. They also need to be able to transmit real time data, so that we always have accurate information regarding how much power each system is currently feeding into the grid.
What advantages are there for system operators to participate in direct marketing?
That depends on the situation. Anyone who is currently installing a new, small-scale PV system will typically receive EEG subsidies for 20 years along with a set feed-in tariff per kilowatt hour. But if you opt for direct marketing instead, you will get a bit more. In the past, this was known as a management premium, which was set at €0.4 cents/kWh for PV installations. On the other hand, we charge a small service fee that allows us to manage the system, but at the same time, we offer the opportunity to profit from high market prices through a market price tariff. Then the operator receives much more than the current feed-in tariff. This is particularly relevant for larger systems at the moment. Financially speaking, direct marketing can definitely be worth it.
Conversely, when the subsidy for a system runs out, so to say a post-EEG system, for example, the power has to be sold elsewhere, since grid operators usually will no longer accept it. But there are a few special provisions that will keep things running until 2027.
So, is it true that direct marketing is mainly used for larger systems with an output of more than 100 kW?
Exactly, at the moment that’s still the case. But, we are already preparing for there to be more small-scale systems in the future. When their subsidies run out, they will need another solution. In Germany, there are currently around 100,000 systems with an output of more than 100 kW, but it’s estimated that there are roughly two million smaller systems that have been installed. That means that there’s going to be a lot happening with small-scale systems in the future. If it’s possible to efficiently and automatically market utility-scale systems, then there’s really nothing standing in the way of doing the same for huge numbers of small-scale systems, which we really need in the future for the energy transition.
Are virtual power plants able to connect an unlimited number of system operators, or is there an upper limit?
Technically speaking, there’s really no limit. It’s simply a question of how the IT system has been designed. But in principle, the IT system can be scaled up or down accordingly. That means, whether we have five or 5,000 systems now or maybe five million in the future, it’s all just a matter of the IT infrastructure. In fact, there are actually quite a few positive effects. As more and more systems are connected, efficiency increases and costs are reduced. Particularly for issues such as forecasting processes, it is advantageous when more systems are connected. This allows us to create better forecasts and maintain a better grid stability.
Virtual power plants are controlled using software. Does artificial intelligence play a role?
Yes, artificial intelligence plays a role, particularly intelligent, self-learning algorithms. Above all, these are useful for creating forecasts. As a virtual power plant and direct marketer, but also as an energy provider, we have to accurately forecast how much power our customers are feeding in or consuming down to the quarter-hour. Since renewable energy is volatile, this certainly presents us with a challenge.
On the one hand, we have to be able to take into account and forecast the personal consumption of prosumers. Artificial intelligence helps to keep the learning phase brief, for example, when we connect a new system, especially since we often don’t have all the information about the system and the corresponding customer. If we already knew everything, like how large the system is, how it is set up, which technology it uses, its maximum efficiency level, and so on, then we would be able to make excellent forecasts. But the problem is that we want to make it as easy as possible for the customer to connect to our virtual power plant. That means that we have reduced the information required from the customer to an absolute minimum, which has the disadvantage of not knowing everything from the start. Good self-learning algorithms can understand a system and its behavior in a very short time.
What are the main roadblocks to the expansion of virtual power plants today?
On the one hand, there are a few regulatory issues that we are, of course, not so happy about – for example, the 100 kW limit for direct marketing. Personally, I’m not a fan of putting an arbitrary limit in place that excludes many systems from the market, and also prevents innovative solutions. I would hope that they do away with this sooner or later.
Another thing is that not all regulations for utility-scale systems should be transferred to small-scale systems, for example remote controlling. At the moment, the law stipulates that every system that is directly marketed must be continuously adjustable, which makes things more complex on a technical level. I don’t think it’s really necessary to be able to continuously adjust a 5 kW system. It would be completely sufficient to simply be able to turn the system on and off. If I am able to turn two million systems on and off, then I’m already able to make more than enough adjustments.
To sum things up: What do you consider the two main advantages that a virtual power plant offers for the energy transition?
The first advantage is that we gain a lot of flexibility from having so many systems connected, for example, from energy storage systems that allow us to shift the peak of solar power toward the evening. The more we expand renewable energies, the more we will face the challenge of having way too much solar power during the middle of the day. But, if we have some energy storage systems where we can feed in the power and use it later, we can alleviate this dilemma of the energy transition a bit. A virtual power plant helps to distribute the power among everyone, so that people don’t have to deal with this on their own.
In my opinion, the second main advantage for the energy transition is the simplicity, making things less complicated. If we want to install large amounts of renewable energies and connect them to the grid, but system operators are left with the entire responsibility of implementing and managing everything, then it’s just no fun for anyone. With a virtual power plant, everything is simply plug and play for the customers. That really bolsters the energy transition.
Could it also be helpful to integrate electric cars as storage systems into the virtual power plant at some point?
Absolutely! That’s a big topic and would offer a lot of flexibility. But to do that, we first have to figure out a few issues, for example, bidirectional charging. One thing that we’ve already started to do, however, is to provide more flexible charging options for electric vehicles. When I park my electric car at work, for instance, it’s just sitting there for eight hours, and when the vehicle is able to be charged quickly, then it’s already fully charged after two or three hours. So, in this case, what we’re optimizing right now is to enable charging that contributes to grid stability in order to reduce energy costs. Rather than charging the vehicles as quickly as possible and at the same time, we can achieve this by slowing down the charging process, maybe for 15 minutes at a time, shifting it to the next quarter-hour. By doing this, we can also offer cheaper tariffs for the drivers of these cars.
This interview is an excerpt from an episode of The smarter E podcast. You can find the German language interview here.
Do you want to receive monthly updates on recently published interviews, best practices and industry news? Then sign up for the EM-Power Newsletter!