The Biggest Machine In The World Is Being Rebuilt While It Continues To Run

Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!

---

Last Updated on: 12th March 2025, 02:40 pm

Recently I had the opportunity to sit down with Mark O’Malley, Leverhulme Professor of Power Systems at the Imperial College of London and founder of the Global Power System Transformation organization, which was based on the principle of grids moving toward 100% renewables. The grid is the biggest machine in the world and as we shift from legacy technologies like coal to next generation technologies like wind and solar, we’re rearchitecting it from the ground up. What could go wrong?

What follows is a lightly edited transcript of the first half of our discussion. I promise, the rabbits in the lead graphic will be explained.

Michael Barnard [MB]: Hi, welcome back to Redefining Energy – Tech. I’m your host, Michael Barnard. My guest today is Mark O’Malley, Leverhulme professor of Power Systems at the Imperial College of London, among other hats. Welcome, Mark. 

Mark O’Malley [MO]: How you doing, Michael? Are you well? 

MB: I am very well, thank you and looking forward to this conversation. I’m going to learn a lot, which is my entire purpose for doing these podcasts, having really interesting conversations with really bright people who know stuff. So let’s start with you. I always like my guests to share how they got to where they are because they’re usually in really interesting places and the journey is fascinating. So how did Mark O’Malley end up at Imperial College? 

MO: What I’ll do is I’ll just go back about five years. I’ll tell you where I was before the five years, I suppose as well. So about five years ago I did a stint at McGill University in your country in Canada. And then I went to the National Renewable Energy Lab In Colorado, the U.S. Department, where I was chief scientist from 2018 to 2020 and 2020 with COVID Etc. I’m Irish herself. My wife decided to move back to Ireland at the time for personal reasons about COVID And I then became the chief scientist of the Energy System Integration Group, which is one of those other hats I wear, as you said, which is an organization that’s run by a great friend of mine, Charlie Smith, and I became the chief scientist of that for a couple of years. 

And then Imperial College London made me an offer I couldn’t turn down. So I went to Imperial in 2022, so September 1st. So I went Imperial about two and a half years. That’s how I got to Imperial as leaving professor of Power Systems. It’s funded by the Lever who Trust, who I much must acknowledge because they pay my salary, so to speak, and they’ve funded a big research group around me. Prior to that, I spent a long time in Ireland in University College Dublin as a professor of electrical engineering. I built a very large research group called there called the Electricity Research center that was very well known across the world. And I suppose it set me up for some of the international stuff I did since then. So that’s my story. If you want some more, I can give it to you. 

MB: That’s probably enough for now. So this journey started because Laurent Segalen introduced us because we’re in somewhat overlapping spheres in the sense that I’m at least dabbling in this space. Laurent is attempting to build an HVDC transmission cable across the Atlantic, which still occasionally triggers my holy crap mode. But part of large grid balancing, and that’s something that professionally you’ve been focused on for 14 or 15 years, is what do we do in 100% renewables based grid? And so I’d like to cast your mind back to something you told me about, which was the 2011 toward 100% workshop. Why don’t you start by just saying what was that, what was the intent of it and what were the primary outcomes of it? 

MO: So if I can go back a little bit further than that, because maybe in my introduction I should have also said that my area of expertise has been on wind power, integrating wind on this onto power systems. But then as solar became economically viable or close to it, then it was wind and solar. So that’s my background and I’ve been doing it for more than 15 years. It’s. Yeah, it’s hard to say, let’s say 30 years just for want of a good. 

MB: You did tell me that, you did tell me that you knew Eddie O’Connor. 

MO: That’s right, yeah, I do. Well, I did. Unfortunately, Eddie, poor old Eddie departed. I knew Eddie O’Connor in Ireland when he was in Ireland setting up air. So Eddie was actually Eddie’s company Airtricity at the time, or can’t remember what they call themselves. But yeah, they used to fund my research group along with several other companies. So I knew Eddie well. A good man, a big loss, I’d say. 

MB: His legacy lives on in SuperGrid Super Solutions. 

MO: So I suppose to go to your 2011 and 2011 is the wrong date. I saw that in the email this morning. I should have alerted you to that. So when I went to the US as chief scientist at NREL, Martin Keller, the head of NREL, gave me this job. He had a fair play to me. He facilitated me in a huge way and gave me a lot of latitude as what I did. So I teamed up with Charlie Smith, who I mentioned before, who I have known for years as well. Charlie was the founder of UWIG, the Utility Wind Integration Group. It then became UVIG and then it became ESIG. So he’s the real founder of ESIG. We teamed up together in 20. I think it was 2018 when I just arrived at NREL. 

We teamed up because we were noticing that in the media mainly and actually interested off in the academic literature, which is another bugbear I have. I thought the academic literature was supposed to be a bit more objective, but unfortunately I could speak at length on that, but I won’t. We were finding there was a lot of material coming out that said that getting to 100% renewables was impossible. And a lot of the literature was also saying that it was dead simple. And we believe that it’s actually somewhere in between. And as a group of people, we’ve never advocated for 100% and we’re still not advocating for 100%. We’re not advocates for anything. We simply are people who do integration. If you want to put lots of technology on. We look at the problems and issues that may cause. 

And wind and solar are the big growing areas at the moment. And that’s been the big focus at the moment. And it is a pretty dramatic change in the power system, how it operates as a research area. I’ve made my career in it. So we as a group of people around the world, people in ESIG, people at the National Renewable Energy Lab and all over the world, my friends and colleagues around the world were all of the view that while it wasn’t impossible to get to 100%, it was difficult and it was somewhere in between. So we decided to run this workshop called towards 100% in 2018, 2019, I think it was in May 29th in Denver. Invite only, but it was. 

I mean, we attempted to make it as international as we could, but of course being in Denver in the middle of the United States is pretty much dominated by North America. But we did have, I can’t remember the numbers, but 10 or 12 people from Europe, a couple of people from Asia and then mainly North America. And out of that came a report that’s published on E6 website called Towards 100%. That was the genesis then of a lot that came after that, which was eventually the founding of the global power system transformation, which is. It became obvious after that workshop that if that, you know, for policy reasons, for all policy reasons and economic reasons, there’s a lot more renewables going up and it’s accelerating and no amount of it’s unstoppable. 

From what I can see, the economics are such that, you know, economics are such now that they’re favorable. And then if there’s a policy aspect to it, then it just drives it up. So renewables are increasing dramatically around the world, it’s going very fast. And let’s say that the workshop sort of highlighted the fact that while we got to where we are reasonably easily, it wasn’t difficult to get to where we are. There was some research to be done, I did some myself. But what was about to come was a little bit more substantial, a bit more dramatic. And some of the system operators around the world that we know were feeling the heat, so to speak. So between those system operators and the researchers, we came together to form Global Power System Transformation, which is essentially an organization. 

It’s not a legal entity. It’s not a legal entity yet. I think that’ll actually change fairly shortly. But the idea is that system operators, these are the people who operate the power system, the big transmission system. I’m going to leave out the distribution system in this discussion. It is important. I’m not downing it, but we’re mainly interested in the big truck and of course the little distribution systems hang off it as well. So they certainly are part of the story that they were facing very difficult technical challenges, particularly a few of them who were at very high penetrations, and that these problems needed to be solved quickly and in fact so quick that in fact they didn’t have time to solve them. If, I mean, most of these changes that happen in the power system have happened reasonably slowly over decades. 

The rate at which wind and solar in some parts of the world are coming on is so dramatic as to be. The pace of change is so dramatic that they just simply keep up. That’s not a criticism, it’s just a reality. They’re small organizations, they’re not large, you know, they’re not large, you know, mega companies with huge balance sheets. They’re small organizations generally owned by the government and they’re essentially civil service type organizations. They serve the people, so to speak. There’s also no profit and you know, it’s not a commercial entity. But they’ve been asked to do something very quickly that’s very difficult and they’ve been asked to do it by the government, etc. And they’re small. So the idea was to bring them all together, share their pain, so to speak. They had a lot in common with each other. 

And so the GPST, I suppose it’s a concept that says in order to get there quickly and to do it well, we all have to work together globally. And that’s the power system operators and everyone else has to collaborate. So that’s the GPST. It sort of came out of the closet, so to speak, in 2020 or 2021 I think. Yeah, four years ago in fact. Four years ago, I think around this time it was officially launched at the Biden Energy Summit. That was held mainly online, but there was something in person. So it sort of got officially launched about four years ago, but it existed for at least five, maybe six years. So that’s the GPST.

MB: I am curious because I compare and contrast to another organization which is overlapping with a different global intent, the Global Energy Interconnection and Development Cooperative Organization. And it strikes me that because they’re focused on energy interconnection between countries, They’ve got about 145 countries overlaps strongly with Belt and Road initiative countries. It’s Chinese led, founded in 2016. I’m curious about whether you’ve had interactions with GEIDCO representatives and whether they’re part of the consortium, you know, and your perspective on the differences between those organizations. 

MO: No, no, we’re well aware of them and they’re well aware of us because they did approach us for sure. We have, we have talked to them, but we don’t work directly. Most of these system operators, you know, are sitting there right now having to deal with this. You know, interconnection will definitely work for sure and it’ll help. But their focus is really on interconnection across the globe. Our focus is on power systems and getting high penetration and renewables. They are compatible with each other. And also I think they’re, let’s say their business model, so to speak about the Belt and Road etc,. is not compatible, just to say that diplomatically. 

But we are aware of them and we are, we do work with China as well. We’re starting to work a little. China is a very important player in this. 

MB: Something I live by, you’ve probably seen it on my emails if you bother to scroll down to the SIG line, it’s a William Gibson quote, you know, Canadian author, the future’s already here is just unevenly distributed. And as I and I’m sure that you, from your perspective can point to the most advanced geographies for this portion of the power. Power engineering and then for this, you’re just seeing the solutions emerging through the global experiments that are different in different places. 

MO: The GPST recognized that because of the system operators who we’ve dealt with initially, we deal with far more system operators now than we used to. But the system operators we dealt with initially were those ones who are right at the top of the spear, so to speak. They were the ones with very high penetrations. They were the ones who weren’t talking about it, they were seeing it. It was real, it was there right in front of them. 

MB: What are the top three or five that you know, that were seeing that were, you know, pockets of the future? 

MO: Yeah, well, I mean I’m not here to sort of rank people and all the rest of it. I’m an Irishman, so I’ll start with Ireland. Why not? I, you know, it’s objectively true. Ireland is one of the Parsons. It’s an island. It’s not connected, it’s connected to the, to GB by HVDC and I think we’re about to, we’re in the process of building connected to France as well at the moment. But anyway, as a synchronous power system, Ireland would arguably have the highest penetration of variables in the world. So it probably would be the number one in terms of statistics then. 

Obviously Denmark is a very important player in this huge amount of wind, but it’s actually geographically it’s one country or politically it’s one country, but it’s actually in two power systems and those two power systems are pretty large with respect to the amount of renewables they have, so they’re an important one in terms of the metrics. I’m in the UK now, so I suppose GB is definitely one of them as well. Great Britain. The other ones of note that we dealt with at the start were California, Texas and Australia. There’s other ones as well. We could have. Who would be in that list as well. I think I should mention places like Spain, Portugal are definitely up there as well. 

And it wasn’t that we ignored them, it’s just that we went with those six at the start but now, like I said, we are interacting with at least 20, 25 system operators. Because your point is quite right. The future is here now just hasn’t appeared in your district. And that’s starting to happen in the last four or five years. A lot of the issues that were identified by those system operators are now in all system operators. It’s spreading fast. 

MB: So there’s one thing to tease apart here because I’m in a very high renewables penetration geography. I’m in British Columbia in Canada and you know, it’s 96% renewables. But there’s a key difference between that and wind and solar that I think it’s worth you drawing out. 

MO: No, no, absolutely. And in fact, you know, like I said, I started this most recent sort of evolution of my journey, so to speak, in Quebec, which is even higher. I think Quebec would claim to be the highest. I think Quebec is 99%, as is Norway, for example. You know, the difference is very important. It. There’s actually the differences that fall into two categories and they’re the two. The two major challenges, if you just put them in two big buckets, one of them is the fact that the old story, when the wind don’t blow and the sun don’t shine, what do you do? Some people use the word. The intermittency issue. 

I hate the word because intermittent indicates on, off. I sometimes call it variable and somewhat predictable.But it’s a very awkward word. We need the Germans to invent a famous German word that explains that. But anyway, that’s one of the issues. And the other issue is the fact that the. I’m sorry. And that issue doesn’t really impact places like Quebec, B.C. And Norway because they’ve got dams, so they effectively dam the water. So effectively they don’t have this issue, they don’t have this balancing issue, etc. Because they dam the rivers or whatever, the lakes, etc. And they actually hold all the water behind the dams. I think Quebec is interesting. I have some statistics in Quebec that could be wrong. I think Quebec has 187 terawatt hours behind — and that is an enormous amount — behind their dams. So that’s one difference. 

Whereas wind and solar, as we all know, you know, well, solar in particular, you know, it’s at night, there’s no solar. It’s simple as that. So they definitely have that balancing challenge. The other challenge is the. Is the way that they’re connected to the power system. So if you look at hydro, it’s connected to the power system with, you know, turbines and then it drives synchronous generators connected to the power system. And that’s why they’re called synchronous power systems in wind and solar though, for all sorts of reasons to do with cost, you know, basically cost and performance. Basically the better way to connect them to the power system is via power. So there’s been a term coin now that’s quite popular. 

I think the North American Reliability Council may have an informal definition, but I don’t know who coined the term first but they’re now known as inverter based resources. So I think the definition that NERC has, if I can remember, it’s any resource that’s connected to the bulk power system via power electronics and to give examples as solar, wind storage. And there’s other ones as well. So they’re called inverter based resources. Whereas your classic hydro plant is connected with a synchronous generator, as is your nuclear, coal, oil, gas units. So synchronous generators have been replaced by, as we put more wind and solar on, they’re being replaced by inverter based resources. That is from an engineering point of view, the most dramatic change in power systems ever. 

The other problem is still there and I just want to mention it before we proceed with this. The problem of balancing, you know, wind and solar, you know, over large scale, wind and solar over season. The daily storage problem is not. The daily balancing issue is not a major problem. That problem is definitely still there. But that problem won’t bite us for a while yet if you don’t. It’s coming, it’s starting to impact some people, but it is out there, if you know what I mean. We have some time, but the inverter problem, we don’t have time because what’s happening now is that the inverters now have so many of them, they’re starting to cause real problems. Power systems or let’s say real challenges. Let’s be positive. 

MB: Well, let’s start with what is it that synchronous generation units are providing to power system stability that inverter based resources don’t automatically provide? 

MO: So just to be clear on this, a little sort of, you know, a legal statement here, I’m a power systems person now, mind you. I work with, like the famous story says, I know some good power electronics, but I work very closely with my power electronics colleagues. That’s one of the reasons I went to Imperial, because they’ve created power electronics. So I’ll give you my take on it. A power electronics person might get a slightly different particular. But basically a synchronous generator is a large piece of metal. It’s two magnets basically. Well, one magnet that rotates and induces currents and another one to produce another magnet. But basically it’s a classic electrical machine. But its behavior is largely based on physics and the metal. So big lumps of metal and the physics, so it’s, its entire characteristics are based on physics. 

There is, you know, some of its characteristics can or do have some control algorithms, etc. Whereas if you go to the inverter based resource, which essentially is power electronics, their behavior is based mainly on software, the sort of software devices. Now they’re hardware as well. But the differences are. Then, for example, let’s talk about the physics in a big large lump of metal. If you want to stick, you know, it’s rated for 100amps, you decide to stick 600amps through it’ll be fine for a while. If you have a piece of silicon that’s rated for 10amps and put 20amps through it won’t last long. So that’s one of the issues. Synchronous machines can produce a lot of. They can take a lot of excess current, which is very useful of various characters, they can reduce fall current, etc. That’s one very simple difference. 

Another very important difference is to do with the fact that the degrees of freedom that there are in the inverter is much higher. And the reason is that the degrees of freedom in a synchronous machine are essentially physical characteristics. You know, weight, size, you know, put simply, the physics that, the literally the physical makeup of how big it is, how, how much metal is in it and the dimensions, and they’re all pretty fixed, you know, you can’t vary them. Whereas the equivalent to some extent in the inverters is actually again the control algorithms and they’ve much more flexibility to change things. So the dimensionality of the individual inverter with regard to what it can do is much higher than the simpler machine. 

The problem there then is that each of the manufacturers, the OEMs, I won’t mention any by name, but the large branded companies who make these devices, they all have intellectual property. It’s in the software essentially and they don’t like to share it because it’s, you know, it’s secret, et cetera. But the problem is they all have, they all do it differently. Whereas most synchronous machines are the same, but in this case they’re largely different because they all have different tricks at a trade and the dimension is much higher and there’s far more of them. So I think I’ll go back over that again, you know, and in fact, I’ll go the opposite direction. There’s far more of them because, you know, a wind turbine might be. Yeah, we’re getting 10 megawatt wind turbines, aren’t we? Eight, nine. You know, we’re getting pretty but… 

MB: Offshore 21. 

MO: … but a large nuclear power plant is a gigawatt. Yeah. So no matter what way you cut it, if there’s a replacement going on, the simple fact of the matter is there’s far more of them. And particularly if you go down to household solar, for example, which I know some distributions of them, you know, every house has it, so you’re into thousands, so there’s far more of them. So from a dimension point of view, way more. Number two, then, in terms of their actual diversity, in terms of the different manufacturers, because the diversity can come about because of software as opposed to the physical makeup, there’s far more dimensions in there. And then if you actually come down to just the sort of, the basics, even if you had one manufacturer, you’d still have way more the software. 

There’s way more control parameters. And the problem with that is there’s way more ways. And that’s the problem. And in fact, to try and understand how they all work together is very difficult. In fact, we had a phone call with. I won’t mention any names of any companies for this. I don’t think it’s appropriate. But, you know, I was on a phone call recently with one of the system operators and this problem of this dimensionality problem is just. They just don’t know what to do with it. Because before this they would have, yes, roughly speaking, say, let’s say two dozen or let’s say 50 synchronous generators, for example. Yeah. And largely speaking, it’s an approximation, you know, once. And they’re all roughly the same. 

You know, they might be different ratings, but as a piece of metal they would behave roughly the same. So it was homogeneous to some extent and they’ve been messing around with it for the last 50, 60, 70 years so they got to know. So they knew it well. It was largely speaking homogeneous and its dimension wasn’t too large. The next thing they’re having to deal with way more of them. Way more, way more hedges, substantially very little experience with them and they’re growing dramatically. And you can see now why the problems are so dramatic. And I go further. It’s not even that the tools, the techniques like this is not a case of taking the same tools and techniques and just scaling them up. That doesn’t because the tools and techniques or happens as well because the very nature of the individual devices. 

I suppose one of the characteristics that I was getting to when I was talking earlier, that we’re controlling now is that it’s very well known in the system. Everybody talks about the so called inertia question and that’s a big difference between them as well. So synchronous generator essentially if you think about it very simply, a synchronous power system is effectively all the rotating masses of all the synchronous generators together rotating in synchronous synchronism just like 10 people on a. I give a lecture on this two days ago, I think in Imperial. I show a picture of a 10 person tandem bike. That’s essentially what a synchronous generation. All those 10 people are working together to drive the bike. Anyone decides to go at a different speed, they can’t because you know, the crowd rules, so to speak. 

That’s been replaced with something that’s not synchronous. So there isn’t 10 people on a bike that are locked together. They have much more independent. And the inertia issue is the fact that combined you can think about synchronous power system, all the rotating masses on all the generators essentially being in one L and that. I know if you do your physics, sure, you’ve done your physics. Half I omega squared [half the moment of inertia times the square of its angular velocity] is the amount of kinetic energy stored in a rotating mass. Where I is the inertia. That inertia, thus energy store, although small, is actually fundamental to how we operate power systems because it’s a buffer, it’s an inbuilt. You get it for free. It’s there no matter what you do. It doesn’t need any control. It’s a pure natural sort of buffer. It’s, it’s a, it’s a beautiful thing as a certain person might say. 

But that’s now essentially disappearing. It’s not gone, but it’s getting light. That’s the inertia question. That’s one of the big issues that has issues not just for the speed of the system but for all sorts of. That make it different. 

MB: Historically we had multi ton chunks of spinning metal that provided a lot of stuff for free to keep power that comes out of our wall sockets stable, reliable, within the right voltage range, within the right frequency range. And now we’ve got a bunch of rabbits multiplying madly and hopping off in all directions. And they’re all different breeds of rabbits but from different organizations who decided it would be really fun if our rabbits had really long powerful legs. And other people said no, we need short legged rabbits to crawl through holes, you know, jackrabbits versus other rabbits. And so yeah, and there were no standards about how they should do it. So they all tried, but it’s all custom. And this leads to I think the consortium. 

So why don’t you tell us about your role with the consortium and then you know what that leads into in terms of the big chunks of stuff the consortium is focused on. 

MO: Yeah. So I’m glad this is not an academic exercise when you’re talking about rabbits. But it’s a good analogy. It’s a good analogy, there’s no doubt about it. It’s definitely a good analogy. So sorry, remind me what you said. 

MB: You got stuck on the rabbits? So the consortium and the research agenda. 

MO: So I suppose to be clear as well, like I’ve just mentioned the sort of the major, you know, but in that bigger chunk of the sort of the balancing and the inverters, there’s all sorts of side issues that all come into play. So it’s not just about inverters. So the consortium came together and identified research areas or research questions. They taught collectively that if these research questions were answered then you could get to very high penetration. So note the word we use towards 100%, we’re not advocating, but you could get to very high penetration. And in fact one of the targets, one of the sort of the short term targets was could you operate a power system with just because, you know, 100% people might think it’s 100% energy. That means operating at maybe 100% of the time, only with it. 

Now I think at this stage we need to be recognized as, you know, that may only apply to very few systems because if you have nuclear and it’s low carbon, you know, so there’s all sorts of combinations of permutations here. But one of the sort of short term targets that you could say is could you actually operate it even for an hour, that was a challenge and it’s still a challenge that still hasn’t been met from what I know. I know that there are microgrids around there that can do it for sure. They’ve always been there. Large scale power system somewhere in the world operating 100% off inverters for an hour. I don’t think this. There’s some people getting very close to it for sure. There’s definitely some people are getting close to 100% carbon free. 

I think in fact one or two have probably done it. They would typically have nuclear as well. So they would have nuclear, wind and solar and they would then be for a few hours maybe carbon free etc. 

MB: To be clear with Iceland geothermal. 

MO: Yeah, yeah, that’s. Sorry. Actually there is lots of people but I suppose what we’re focused on most of the world doesn’t have geothermal. Most of the world doesn’t have hydro resources so I know they’re interesting cases but you know, Norway, Quebec, bc, New Zealand, a few places like that certainly have huge water resources but most of the rest of the world doesn’t have the same scale. So most of the rest of. Most of the rest of the world doesn’t have nuclear. So if you think about this globally most of the rest of the world tried to decarbonize electricity are probably going to have lots of solar so we are concentrating mainly on inverter based use. 

So we identified lots of research topics that if we could solve them we believe that you could operate a power system at 100 and could probably do it for a large period. Again we’re not advocating going to 100 etc the storage. So first of all on the storage issue, on the sort of, you know, the balancing issue that wasn’t taken as a near term problem. It’s certainly a long term problem. There’s no doubt about it. I would underestimate it at all. But it wasn’t taken as a long term. It’s a long term. We definitely need, you know, some form of. Well, I’ve got other ideas in that but I won’t go into them. But you’re going to need some. So in the short term it’s the inverters and those research questions largely speak broke into six areas now. 

In fact we’re actually updating the research agenda right now as we speak. So there’ll be a new version of it available in about two months. So I’m actually in the middle of it. Tonight I’ll be working on that over the weekend. So the research buckets that we have now, they’re roughly the same as the ones we had before are inverters themselves which is. It is number one, I’d say it’s the most important way the most short term number two would be, we’ll call it planning things like adequacy, things like, you know, storage, etc. Do we have enough generation? So that’s the second one. The third one would be distributed energy resources. 

Again we’re not dealing with DSOs, but we, if somebody has 14 gigawatts of solar on their roofs, whatever then the power system operator for the bull power system has to know about, you know, so the d are distributed energy resources very important. 3 or 4 is control room of the future. So if you think about it, and again I don’t want to simplify things but you talked about the rabbits. Can you imagine trying to orchestrate the rabbits in a control room? So that’s, you know, it’s a very simplistic way of putting it but it’ll do. So the control room has to deal with far more moving parts, far more, many more things that are changing dramatically over short periods of time. So the control room that they control, this has to be, has to change. 

The fifth one then is on, you know, I think I mentioned it earlier. Not only are, you know, not only is the dimensionality changing. But the tools and techniques that we need to analyze these things have to change. The tools and techniques we have at the moment may not be, are almost certainly not fit for purpose for lots of issues that are happening. So we have to develop new tools and techniques and while we can develop, while we do actually have some of these tools, they’re far too slow. So we have to develop new tools, techniques. So that’s another. So we’ll call it tools and methods. 

But mainly I suppose the real caveat in that is mainly to detect instability on the system which can occur because of all these things. Let’s say that there’s so many of them, they’re all acting a certain way, they can cause instability. But you can’t be just the dimensionality so large. There’s as well the dimensionality is so large, there’s so many, the space is so large, there’s so many different places. It can go wrong, essentially to keep it simple. So those tools and techniques are mainly focused on detecting and stopping systems from unstable. And the last one. 

MB: Yeah, I just, I was just thinking about feedback loops there. You know, I, I’m imagining this is my hypothesis, not being an inverter, power engineer, but, you know, they would be detecting the frequency from the grid and feeding that frequency back to the grid is my assumption. 

MO: Now we’re now into the. I’ll tell you what, dude, let me finish off on the sixth one and then I’ll get into that. Yeah, so that’s fine. No, no, I think you brought up a very interesting point, which is where I could do it. One of my colleagues, like Baris Kasikci from Imperial or Tim Green, they could be doing sit beside me because I could get it wrong because that’s where they, that’s what their expertise is. But the sixth one is services, which is the one that I particularly focus on myself. My Leverhulme professorship in Imperial is based on system services. Services are those. Energy is the main service, if you want. That’s what, that’s what we want. But in order to get you the energy delivered to your home and factory, there’s other things that are required and they’re called services. 

So they’re very technical. And again, back to our synchronous machine that we talked about before. The synchronous machine, effectively, to a first approximation, it could do nearly all the services. Right, so once you had a synchronous machine, you know, not entirely, but, you know, largely speaking, once you had synchronous machines, you didn’t have to worry too much about it. They could do it if you now with the inverters, they can’t do the same things at all. And there’s much more of them, etc. So the whole problem of supplying services is become fractured atomized, I think is the word, you know, and not only that, but we probably need new services as well because of characteristics. So that’s services. So there the six. And we’re currently reviewing it. 

But back to your comment. Inside the power electronics world, this is a bit like, oh, I’ll get shot for this, but this is a bit like beta Max versus VHS, let’s say. Remember that two stand two different standards, whatever. Now it’s slightly different. So there’s two main technologies, so to speak. Grid following and grid forming, they call them now, to be absolutely clear about it. There’s not a very distinct difference. There’s sort of a continuum in between. So it’s not a case that this is grid forming, that’s grid following and there’s a very strict definition, but grid following, grid forming. And to your point about the frequency in a grid following inverter, which is mainly the type of inverters that exist on power system today, effectively what they do is they follow the grid. That’s what they do. 

So they’re like you said, they look at the grid and they see the frequency and they see the voltage. And what they do is I’ll do as they do. So they follow. That’s all they do. Now that’s fine until they get to the position whereby there’s no one to follow. If you know before the sinks of machines disappear, all of a sudden someone has to form the grid. So you have grid forming it. They can actually form the voltage and you know, they maintain different. And they to some extent they replace the sort of fundamental thing about the synchronous machine. They actually form in a synchronous machine, they all work together at 50Hz or 60Hz. Where you are in a grid forming situation, they will all work together to form the grid at 50 or 60. So there’s two different technologies. 

Your feedback loops you talked about, it’s not there. The feedback loops are there though. The feedback loops are right inside the system and they’re things we don’t like. In other words, if one inverter does one thing and another inverter reacts to it in some way, it might feed something back. That’s where these interactions between these different devices causes some problems in the system. So there are two big buckets, grid following and grid forming. And there’s a really fundamental question. I wasn’t really aware of it until recently because it wasn’t my area, but my power electronics friends are very much into it as, you know, it’s not just simply a case of as someone says, well, why don’t we make them all grid forming? 

As you know, it’s a reasonable first order approximation because essentially it replaces all the grid forming. But there’s several problems with it. So number one, in order to have a grid forming type of operation, you have to have an energy resource behind it. So you probably have to back off your wind or solar or put storage into it. So it makes it more expensive to start with. And second of all, it turns out if they’re all grid forming is a very Bad in many, it’s even worse than not having. So it’s not a case that you can just say replace them all. So there’s all sorts of cases of, you know, you’ve grid following, grid forming. Grid following does well in certain circumstances, grid forming does well in certain other circumstances. You do need a balance between them. 

You can’t have all one or all the other. There’s a sweet spot in the middle. That sweet spot depends on the circumstances. It depends on the system, depends on the location. So it’s a very complicated sort of sweet spot to be worked out. In all power systems. Systems. That question is one of the big questions. We do not know how to calculate that suite. We don’t know what the right balance should be. And at the moment the technology is evolving because like I said, there’s different flavors of these inverters between grid following and grid forming. And the manufacturers are changing, you know, changing their knobs and bells and whistles all the time. So it’s a very difficult question as to what the combination should be. And it’s a question that people want to know. 

Most power systems around the world do not have too many grid forming inverters. They mainly have grid falling because they didn’t need to be grid for, but they’re now going to be needed. Some of them are going, some of them are going to be grid forming. Just to be clear, grid forming is not a new thing. It’s a well known thing, if you know what I mean. Microgrids are grid forming. You know, they have, you know, micro grids have to be grid forming because they have, you know, if you have a micro grid with just wind or solar and a battery, that’s it. They have. But they’re microgrids. Microgrids are not power systems. They’re, they’re micro. I won’t go into that either. So that’s basically where we are. 

Then there’s a whole bunch of sort of very, very detailed research questions within those sort of buckets. If you, I’m right now actually trying to review it and renew it, so to speak. So I’m in conversations with lots of people. So this is probably a very appropriate time to have this. The sausage is being made at the moment, so it’s a good time to talk about it. Might actually help me. That’s what I have to write about it later on. 

MB: Well, I hope I don’t introduce too much of a flavor. 

MO: No, no, I was going to say, I was just seeing the new report coming out. We have a picture of a rabbit or no hundred rabbits running around the field. 

MB: One of the questions that occurs to me, so obviously inverters because as you say it’s so much software based and I come from a big software background. I had like a couple or three decades in large systems globally. The variability of grid forming or grid following can be a software configuration in an inverter. So you can put a whole bunch of ones, the capability to do grid forming and then just presume that one of the things you’re working on is what is the control system that configures all these things to be well behaved grid forming, grid following inverters in the complex system per geography. How do you even figure that out? How do you send the signals? What are the interfaces, those types of things? 

MO: I mean again, my power electronics colleagues should be here beside me to keep me on the straight and narrow. But my understanding is that you could actually make an inverter that is software configurable to be grid formulated. You know, does not now apparently it’s not that straightforward, but it’s not, I’ve been told that it’s not that difficult. You could actually have a, it’s an inverter, you can make it, you can decide yourself how you want to operate. Which sort of illustrates the dimensionality of this problem. So you know, if you have a situation where that’s possible. If they all woke up some morning, your rabbits woke up and they all decided to be grid following, they wouldn’t be good. Or if they all woke up inside of the grid for me. So that actually, that’s a good, it’s a good thing. 

They can all do whatever they want every day. We can’t have that. Actually we do need them to be in some bounds and they’re the bounds and we just don’t know. We simply do not know. And it’s not a case that the bounds for system X and the bounds for system Y will be the same. Now they share a lot in common but you know, just because someone knows the rules of the road for one system doesn’t know, doesn’t mean it’ll apply to the other system because they’d be subtly different. 

MB: Yeah, well let’s just take a compare and contrast as an extreme example, let’s take the south eastern provinces of China versus the northwestern provinces of China. So my last conversation like this was with David Fishman, Shanghai based resource with the Lantau Group and he spends all of his time in power economics in Asia and mostly in China. And so we’re talking about the reality that in China, 50% of the solar market is rooftop. And in southeast China, which goes from densely populated to more densely populated to extremely densely populated, there’s solar on everything because that was a policy thing. So they have millions of tiny inverters and then they also have some coal plants locally which are diminishing in number. 

But then they’re getting this massive amount of energy from the northwest and northeast from mega bases which are grid forming arguably because they have massive amounts of solar, massive amounts of wind storage and coal plants and they’re putting in HVDC for 2000-3000km to feed these power hungry southeastern provinces. And so you’ve kind of got, in the northwest you’ve got these, the grid is being formed by the coal plants that are part of the mega bases, but it’s also a very low demand area. Then they’ve got this high demand area with all these microinverters all over the place. And so it’s a very different balancing act in those two places. And how do you even figure out what’s appropriate? It’s a challenging problem. 

MO: So you’ve raised several things here that are worth commenting on. So number one, HVDC, let’s talk about that because that’s an important part of this. That’s just a bunch of power electronics as well. So that’s the other thing, you know, so I didn’t go, you mean. And they’re not inverter based resources like a solar or wind or by, they’re different, but they are essentially inverters. They are. So that’s the first thing. They are also inverted to a large extent. So they are part of the issue. The second thing though is once you connect something by HVDC effectively what’s on the other end is your, you’re, what’s the word? You’re blind to it. You don’t, it doesn’t make any difference. All that power electronics will determine what you see, so to speak. 

But you know, and again, without naming names, I was on a call, another call, what about a month ago, with some system operators and all this issue of mixing DC and AC for example. And I think China is a great example. China is, I’ve spent a lot of time in China, my friends. I did some teamwork with the Energy foundation there and were trying to help the Chinese. I don’t know if were helping them. We were, let’s say, advising them. And China has a really fascinating situation where it has essentially a HVDC overlay, essentially the major amounts, but major amounts of energy been shipped around the country by HVDC and then under it, so to speak, is the AC system. 

So they actually, to some extent, and I’ve been in the control room in Beijing and anyone who’s been in it’s pretty interesting place to see. And I think the thing to say is they sort of manage the HVDC bit and then it drops down to different AC systems. Whereas if you go to European and North American places, there’s a little DC. And what it is AC mainly is the thing to control and then the DC is on the side, but that’s going to change dramatically, particularly if you look at a lot offshore wind, for example, in the North Sea, etc., DC is going to be used to connect them together. So there’s going to be an awful lot of HVDC. So that’s another. It’s in the research agenda there. 

But I think it’s going to be an even bigger question going forward. How do you make these AC and DC systems work together?