The energy transition cannot be achieved without intelligent solutions for the transmission and distribution networks. What is the current technical status quo of smart grids? What role does the grid edge play? And why do engineers need to talk more to the people? Dr. Michael Weinhold talks about this in our interview. He is Head of Technology & Innovation Smart Infrastructure at Siemens
Michael, to cover the basis first, what's your definition of smart grids beyond what we usually use as the buzzword?
In an overarching mission perspective smart grids are electricity networks that allow the intelligent integration of all the stakeholders and users connected to it, like generators, consumers and prosumers to deliver a sustainable, economic and secure electricity supply system. In a technology perspective it is a electricity supply network that uses digital communication technologies to detect, to react, to forecast to changes of energy supply or usage.
If I compare smart grids with the conventional power grids that are already around, what are the differences?
In the 80s when I studied, we had a basically conventional power generation. Renewables were around hydro. Solar energy was more or less for satellite or very exotic applications. The grids were built as needed to supply the consumers and the loads, the private consumer was more or less passive all over the world. This has changed quite dramatically. Now we're integrating a huge amount of renewables on all voltage levels from a very small size, like rooftop PV, to very large size, like offshore wind parks. That has to be integrated and transmitted and brought to the usage of energy. At the receiving end we have more and more loads connecting to the electricity grid, for example for e-mobility.
Now, let's look at the stakeholders out there in the field. A lot of them are just basic consumers, such as me, maybe having a solar panel on the roof or an electric car. So I might be producing my own electricity and using, storing or feeding it back into the grid. I think my use case would be falling into the so-called grid edge which is another buzzword. Could you please explain that to me?
Well, the grid edge is actually the interface between the electricity grid and the usage. The usage could be, as you describe, the feed-in of photovoltaic power from your rooftop PV plant, the charging of your e-car, the operation of your heat pump or the HVAC system of a complete building. This grid edge actually gets more and more interactive. One word we should mention at this point is flexibility, because that is one of the key elements of the future system. Flexibility to take in the variable in-feed of renewables and store it, for example, in thermal storage. Or a building has typically quite decent thermal time constants so preheating – precooling would become a key in the future.
At the German national grid, even just the basic conventional grid at the moment is a huge problem. We're not even being able to transfer all the power created in the windy north to the industrial south. So if you look at Europe as a such, then I think the next level would be a smart grid. How close are we to it and what are the biggest hurdles in getting there?
Maybe to start, you mentioned the, let's say, grid bottlenecks that we have in Germany and this has been analyzed also taking into account future supply and demand characteristics. And we are looking towards for HVDC, high voltage direct transmission lines or cable systems running north south to bring down the vast renewable energies from the north to the load centers in the southern part of Germany. One quite exciting project is the so-called Ultranet Project in the southwestern part where existing transmission lines are converted into a HVDC line. So this is a power electronic application where energy is taken at one point, in this case the northern part of Germany, and is transmitted to the receiving end, to the southern point, where it is injected again into the AC system. And this shows one beauty of HVDC systems. You can very precisely take and inject energy at another point of the system.
In my eyes, probably that's from an engineering perspective one the more basic things to implement compared to like the whole scope of a smart grid. Yet we still have huge challenges in getting it implemented, really from a regulation point of view from citizens not wanting to have this huge power lines on their field.
Two aspects, here. The first one is that HVDC systems and how the terminals of the HVDC systems work, they are part of a smart grid. And you can see it in the 10-year development plan or in the papers around what should happen in Europe. You can see there are several already existing HVDC lines and others being under construction or considered. So high voltage direct current transmission is in Europe and in other parts of the world, already an established technology. And I would include it into, let's say, the box of smart grid technologies that we have at hand.
The other aspect that you were mentioning and which I think we should touch on that one, is us as people with our wishes, what we want and what we do not want. And so this is really important because one of the key elements of our future energy infrastructure is that we as people play an important role. We decide what will be implemented and what not. And therefore it is important to reach out to the people and tell them, OK, if we move towards more wind, more solar, these are the options that we have in order to realize a sustainable infrastructure.
There's a big effort being made to establish a Europe wide charging infrastructure. How closely is that link to the effort to create also a smart grid across Europe? Are they happening parallel to each other?
That happens parallel and needs coordination, of course, because if you think about it, where are the big challenges in our grids? The closer you get to the low voltage level, the bigger typically nowadays the challenges become, because of the low voltage grid was a very passive part of the electricity grid. In former times, you could very easily forecast the usage pattern and build the grids just as they were needed with respect to capacity, for example. But as we discussed, nowadays especially the distribution grid and the low voltage grid have become quite active. We talked about charging infrastructure, about heating, cooling of equipment, connecting PV plants on the rooftop. A majority of I think 90 percent of the renewables are connected to the low voltage grid in Europe. So there's a big challenge and it has to go hand in hand.
When we talk about smart grid, this includes the integration of charging infrastructure. Especially fast charging is a quite decent powerflow. And fast chargers are very necessary in order to promote the application and the acceptance of mobility. And, just think about it, the average household in Germany probably needs, on the average, one kilowatt of power consumption. But if you connect maybe an 11 kilowatt charger and several neighbors do so, you see there is a challenge for the distribution grid. That's what a lot of projects are nowadays about, and also technology developments, to control the downstream low voltage charging infrastructure, for example, or heat pumps very efficiently. And I could talk about a project where we are applying artificial intelligence, an intelligence substation in order to manage it.
Yes, please tell about that.
So what is it about? You have a, let's say dynamic state estimator which gets trained in the cloud and the parameters get downloaded to the neural network in the low voltage substation. The neural network manages the charging infrastructure downstream such that there is no exceeding of voltage or current levels. In case the actual measurement values differ too much from the values of this state estimator, the neural network gets trained again. This could be a very low cost but very effective solution to low voltage smart grids. That's why we are prototyping it with customers like Stromnetze Hamburg.
So the power consumption would go up significantly due to electric car charging. And if all of us do that, we just need an intelligence to make sure we're not all pushing the button at the same time?
Yeah, that's maybe a little bit simplified because I was talking about average values. And of course, you don't run your eleven kilowatt charger all the time. But in essence, you're right, since we're moving charging infrastructure into our cities, into the parking garages of apartment houses and we will have charging along the streets in our cities. We will have to see how we very effectively can accommodate the power flows. And part of it is, of course, having smart charging. So an interaction of the charger with the smart grid where it is connected to another one, could be a strengthening of the grid. This is a system planning task.
This interview is an excerpt from an episode of The smarter E podcast. You can listen to the full interview here.