The role of high-power converters in efficient and stable smart grids

David Hegarty: Hello, and welcome to Sound On. Power On.

David Hegarty: I’m your host, David Hegarty, and today I am talking with Prof. Marco Liserre, about high power converters and how they can contribute to the optimal and stable operation of electric grids.

David Hegarty: On one hand this is about advances in their efficiency.

David Hegarty: On the other, it is about the information that converters can contribute to the smart grid, and thus help to deal with the increasing numbers of wind and photovoltaic systems at one end, and significant loads, like charging stations, at the other.

David Hegarty: Marco is Professor and Head of the Chair of Power Electronics at Kiel University, in Germany.

David Hegarty: Apart from a book and 7 patents, Marco has published more than 700 technical papers. Perhaps most significantly he has 50,000 citations.

David Hegarty: In 2023, he joined the Fraunhofer ISIT as deputy director, and head of the new "Electronic Energy Systems" group and of the Kiel branch.

David Hegarty: He an active member of several other organisations, including being Co-Editor of the IEEE Open Access Journal in Power Electronics, and Technical Committee Chairman of the Committee on Electronic Power Grid Systems.

David Hegarty: Over the years, he has been the recipient of 16 awards, including the prestigious 2018 IEEE-IES Miitelmann Achievement Award, and in 2023, the IEEE-PELS R. David Middlebrook Achievement Award. Also in 2023, he was awarded the title of "Ufficiale" by the President of the Italian Republic.

David Hegarty: He has chaired or co-chaired several IEEE conferences, over the years, and is looking forward to 2025, when he will be Chairman of Powertech, which is taking place in Kiel that year.

David Hegarty: Welcome, Marco.

Marco Liserre: Thank you. Thank you, David.

David: Maybe we could start with the efficiency of power converters, as this is a topic that will affect not just energy losses, but also thermal aspects, associated costs, and reliability across all the different applications.

Marco: So, let's start maybe with wide band gap devices. For sure the silicon carbide devices, they can play in the, higher voltage category – so let's say ten kilovolt – also a role in smart grid technologies like solid state transformers.

Marco: Also, the other components around, they play a role when it's coming to the aspects related to the efficiency – like passive components. Of course, we try with the modular technologies, like in the case of the higher voltage, to the modular multilevel converter. And in the case of higher current, puting in parallel power converters to reduce, basically, these passive components, and to, lead to a reduction also of cost and losses. And this is very interesting, because it's not only to do with the use of these devices, wide band gap devices, and to the topology like modular multilevel converter, but also sometimes with aspects related to modulation and control, like having an optimal modulation for reducing the losses.

David: And what advances are we seeing here?

Marco: Well, the advance is that when it's coming to the management of these modular topologies, in the past, there was actually seeing that one by one, when it was coming to the parallel power converters, it's just puting this parallel converter in parallel and synchronizing them – what is actually still a lot used in the industry. In the last 10, 15 years, we have learned how if one can reduce significantly the inductances, which are used to connect this power converter to the grid, by using so called interleaved modulation. And also, we have seen how we can effectively use neutral point clamped converters, which are multilevel converters, and they could also push to higher efficiency and further reduction of passive components. And now we can optimally choose the modulation to reduce for example, circulating current, which can arise by this particular operation.

Marco: And when it's coming to modular multilevel converters, we have seen a number of innovations. One can also operate them differently to take into account for example the lifetime of the component. Or one can use different kind of modulation, like a discontinuous modulation to optimize their use.

So also, in power electronic, we are more and more using new technology, like for example, digital twin or power hardware in the loop. So, these tools allow us to design synergically hardware and software together. And it's not needed to design the full system. We can design a part of it and then test in a virtual environment, with this approach of power hardware in the loop, in the sense that the rest of the system is emulated. We can, out of a part of this system, like one cell, create a digital twin, which allows us to make much more test offline, before testing something in the real laboratory. And then using artificial intelligence, we can come to a higher level of optimization of these two levels: this hardware and software level, together.

David: You have you been looking at these solutions in the lab. What have you been able to achieve?

Marco: What we have been building different kind of prototypes and like using parallel neutral point clamped converters. For example, in this case, we were achieving the use of interleaving inductance, so small, like 0.01 per unit – so 1% – which is an amazing small size.

Marco: And also, we have been focusing on using of modular multilevel converters. In those modular multilevel converters, we have also special features like the possibility to measure the junction temperature of the device, and then give to them the capability to operate with higher value of currents during the grid faults. These are some of the examples.

Marco: When it's coming to efficiency values, in that case, we were more in the DC-DC converter, demonstrating that also multiport technology topologies – they typically in the past were acknowledged to have a lower efficiency because of reactive power circulation – they can come to efficiency in the order 98% and higher. And this is thanks to silicon carbide and special design.

David: If we look at how these advances can be applied, for example in the case of medium voltage DC converters for connecting remote photovoltaic or wind parks to a distribution system. One of the stumbling blocks here has been about protecting against DC faults...

Marco: We have been studying intensively the use of modular multi-level converters to manage these DC faults and to limit DC fault current. And this of course can be done with an MMC, all with full-bridge cell. But this, of course, is a significantly higher cost, with respect to the use of a half-bridge cell.

Marco: And then we have been studying the use of hybrid topologies with partly full-and half-bridge cells, which can be also effective. And the combination of this property of the MMC to break the DC fault current with the use of switches, and say, protection switches – so, a DC breaker – which are of lower size.

Marco: So, these are all solutions, which they are not only focusing on the power converter itself, but they are showing what the power converter can do for the system level. And how the power converter can also reduce the need for other components and create an advantage at system level.

David: At the other end of the grid, a lot is changing with the loads too, isn’t it? And they are bringing their own challenges.

Marco: Charging stations is very strong emerging topic. Because we have to consider that we have to build a large infrastructure. And this infrastructure is not only for autos, but also for trucks. There is, I think also the use of batteries in trucks seems to be very interesting alternative to use of, not only fossil fuel, but also to hydrogen.

So, you can imagine that you will have all these different autos and trucks which are running in our streets, and they needed to be charged with different kind of needs: because those, they need the fast charging; those, they can accept also to have a slow charging. And so, we’ll have in the future charging stations, which will be multifunctional, because they have to answer to all these needs. And quite a complex distribution network with the different kinds of charging points, which are also far from each other. So, it's like a small grid.

So, you can imagine that you will have all these different autos and trucks which are running in our streets, and they needed to be charged with different kind of needs: The charging station then can be built with very different technology, with centralized or decentralized one. And again, there is the challenge to reduce the component size and increase the efficiency. At the same time to have a power converter, which they can also interact with the electric grid in terms of reactive power when this is needed.

David: And how is this being handled today, and what are the advances that you're seeing in your research?

Marco: Actually, today we are using still a very classical approach, with a 50 Hz transformer rectifier. And then some DC-DC converter in the charging station. So, already there has been a proposed solution with the solid-state transformer, where this 50 Hz transformer became a high-frequency transformer, leading also to a reduction of use of copper and iron and also a reduction in size.

Marco: And then we can try to reduce the number of isolation stages, so to increase the overall efficiency from the point which we take the power to the point where we use the power. But I think that we will go also in the direction to create in this future charging parks, different kinds of supply. To have a medium voltage DC when it's coming to bring in power a little bit far away, maybe where you have to connect trucks, and using low voltage DC or using still AC, in the case of slow charging. And in this respect, we have the feeling that multiple active bridge technology, which is based on the use of multi-winding transformers, could be a key actor there.

Marco: I think that we are never to forget, when we are talking also about charging stations, that despite these charging stations, they are in principle absorbing power. But also in the case of these large loads, the requirements in terms of grid connection, they will increase. Because this kind of system will come very large loads. And in the future electric grid – all based on power electronics – then the difference between generation and load will became smaller and smaller. We have been doing some study about the capability of the charging station to get coordinated among them and to inject reactive power, to have an influence on the voltage profile into the grid. Because of course, when you absorb active power, you can have some voltage drops and then you have to also care about the voltage profile into the grid.

David: Which sounds to me like we are talking about smart grid technologies...

Marco: I would say that in this way we enter into the topic of smart grid, where there is this concept of the future electric grid, 100%, or like 80%, based on power converters. And in this new electric grid, which became an electronic grid, then I would say that the control of power converters became a dominant topic, because it's important how the power converter, they interact with each other to try to maintain the voltage stable. So, the power converter, they are not anymore operated in so called grid-following, but now all the power converters, they have to take responsibility in forming this grid voltage. And then it's emerging the topic of grid-forming converters, where these power converters, especially if the power of these power converters is very important in the power balance of the grid, then they have to contribute to establish the voltage and guarantee the stable operation of the grid.

Marco: And then we have to ask ourselves if in the future grid power converters, they are only a side actor or really becoming a central actors of this electric grid. And I think yes, they are becoming a central actor, and they should take management role into the electric grid. Which actually means they also have to contribute, in case of failures, to guarantee that the grid is recovering out of this failure. So, this will be a big topic. Because up to now we have been able to have a system based on classical generators, and we are really managing the electric grid. So, I think we want to stress the concept of power electronics becoming a management node for the electric grid.

Marco: And so, this of course then comes always back to the current which power semiconductor device they can provide. But then as powertrain engineers, we know that to the junction temperature, which is really the boƩleneck, and then to the thermal management of semiconductor.

Marco: So we did start from efficiency and from devices, we did come to a system level and looking at the challenge into the electric grid, we come back to the devices as a bottleneck, to actually provide this service, but not only devices as devices itself.

Marco: And if I can say, we come back to the concept of digital twin, because when we are able to make a digital twin of the device, and we are able out of this digital twin to get all the characteristic and condition of the device, we can use at the best, this device in real operation and minimize the requirement of this device and the cost of this device.

David: The paradigm today is that the supplier and the distribution system operator are typically separate organizations. So, are the issues here technical, or are they also issues of companies working together in this respect?

Marco: So, the transmission operator, they see quite clearly the advantage of requesting power electronics to have this grid-forming feature, and then directly contributing to the stability of the grid. And then the distribution operator, they see more the challenge of having power converters, that they are now voltage sources and, they are not anymore in this following mode, which of course, distributed system operator is easier to handle. But that's only a matter of updating the knowledge, because the solution they are there also, if in the future we'll have 100% of grid-forming converted also in the distribution.

Marco: And we have not to forget that power electronics converter, they are also an incredible source of data for system level optimization, because we have a lot of sensors. And on the other hand, at system level, we have a lot of data which are maybe a different kind of granularity, but still, they are very useful for us. So, we have introduced in power electronics the use of data fusion approaches, which try to combine these different sources of data to make the best for the power converter. And this of course we did in cooperation with informatics experts, to really try to get the best out of power electronics.

Marco: And in this attempt to also look to the power electronic in the wider way, in a little bit broader sense, also, in the case power electronic is interacting with such component like battery, or in the case of like electrolyzer and so on, to look also in the chemistry and see exactly how power electronics can do the best for this component, as well as we did when power electronics had to interact with electrical machines.

Marco: So, I would say my last comment is to the also younger engineer to look to the new tools, like we are saying, like the digital tool and so on, as also to transdisciplinary topics like chemical storage. And don't be scared to take these challenges.

David: And with that look to the future, I think we are at good point to end. Thank you very much, Marco.

Marco: Thank you very much, David.

David: If you are interested in any of the topics that we have touched on in this episode, please check the links in the description, where you can get much more information.

David: For Sound On. Power On., and the PCIM Europe, I am David Hegarty, and I look forward to our next episode.