The antiquated one-way system of poles and wires that comprise the U.S. electric grid will one day be almost unrecognizable. It has to be, otherwise it means we failed to meet the increasing demand heralded by experts across the energy industry.
Adnan Chaudhry, the senior vice president of digital grid at Siemens Energy, is dedicated to making power systems smarter, more efficient, and more reliable. I caught up with him at CERAWeek by S&P Global to learn how generation, transmission, and distribution are changing in the age of digitalization. The following is a rough transcription of our discussion, with minor edits for clarity and conciseness.
Paul: In order to navigate this unique time of load growth, we’ve got to think about the grid a little differently. How does Siemens Energy see T&D in the future?
Adnan: “We are very much focused on digitalization. We believe that especially now, in the time of data and AI, there is a huge possibility to bring the grid to the next level. In the past, we were used to doing grid planning and grid operation in a static way. We didn’t exactly have information on what is happening in the outside world or the exact condition of the grid and its surrounding environment. It would take a lot of assumptions in the past. Now with the new technology we have, you can put sensors in a transformer. You can put sensors on the switch gear. You can put sensors on a line. You get a lot of data from different analytics, which was not possible in the past.
If you get the sensors, if you have the data, you actually can use this information by combining it with learning models and AI. You can use data from the past to learn on AI models how to think about how to increase the efficiency of the system, because in the past you were always using some static limits and some assumptions. Now you have real-time data that can be used in a real-time operation. Analyze it, and we can use the grid more in a more efficient way; I can drive more electrons through the lines, more megawatts to my substation. That’s a basic advantage we have with digitalization and AI.”
Paul: We talk a lot these days about the ways we can maximize the grid we have. From your perspective, how much benefit can we expect from grid-enhancing technologies (GETs) and other smart grid stuff, and how much is actually being done right now?
Adnan: “There’s a great example that applies to what we just talked about, dynamic line ratings, or DLR. If we just put a sensor on a line, for example, on a 400 kV line, what we see with current use cases and customers we have, is that you can average putting 15% more on that line. Of course, there are cases where you can’t do more. There are, of course, cases where you cannot do more, but on average you can see 15% more with a simple sensor, without big investments, without any heavy equipment. It’s simple to install with no disruption to operation, you can just go and do it.
Is it used that much? Well, I think it’s a starting phase. In the past, we were used to having static limits. We never drove the grid to its limits at the edge like we’re doing today. The accuracy (of GETs) in the past was not that reliable, but combining the data with AI makes the analytics possible.
The need is there. Electricification is rising. Data centers are coming. The speed of the energy demand is increasing, but we can’t always get equipment (fast enough). So now there’s a different need, to digitalize the grid and think about grid-enhancing technologies. Change is coming as the utilities and customers think about GETs.”
Paul: You have a bit of a unique perspective, having worked at a utility in Europe (for RWE). Do you think there are any energy transition lessons Europe could lend the United States as it encounters a lot of the same roadblocks?
Adnan: “A lot of European countries started early with renewable integration and discontinued operation of fossil fuel plants. You could actually see how the grid was driven at the edge. At the limits, we had an issue where we couldn’t transfer power from northern to southern Germany, which is why there has been an increase in technologies like HVDC.
On the digitalization side, Europe was quite slow, but it’s picking up now. This lesson (for the U.S.) is how to actually increase the capacity with new technologies like HVDC, which is not used in North America like it is in Europe. Europe started 10 years ago, and we were talking about offshore connections. We were talking about large corridor projects to transfer large amounts of power. That’s something starting now in North America.
On the other side, North America has some different advantages. If you look at California, the renewable integration started at the distribution level. They started a different way, and this is something which is now coming to Europe, thinking about rooftop solar and distributed generation. Both continents are going more or less hand in hand on digitalization, although Europe started a bit earlier with the automation. There are things both continents can take from each other.”
Paul: Is there an emerging technology that stands out most to you? Something that really makes you say, hey, we can do something with this?
Adnan: “Everybody’s talking about AI. AI in grid operation provides a huge advantage, and it’s going to be everywhere. Now we can say, okay, there is a match tomorrow and then we can plan on what and where the load is going to be. In the past, we were making assumptions, but with the data and AI and the combination of different models, we can actually improve the efficiency of the grid and availability of the grid in a completely different way.
You can use AI to predict something. Just look at the case of California, for example, with the wildfires. You can’t avoid wildfires. You can’t avoid extreme weather conditions, but what you can do is put sensors on the line, digitize the grid with a digital twin, and then combine that data with some weather forecasting data and some historical data, and you can actually do a lot of things.
What is the impact of a wildfire’s distribution on the grid? How can I predict the impact on the grid? Then the next step is prevention. How can I prevent bigger impacts on the grid and on energy availability in order to take the right actions? Nowadays, you can also create a digital twin of the line and you can then put the surrounding vegetation into that and combine it, which was not possible in the past.”
Paul: What are the hold ups to some of these technologies? Is it just utilities being slow to adapt? Is it a technological divide sort of thing? Integration challenges?
Adnan: “We’ve been talking about smart grids for 25 years. We had a nice idea and a nice vision, but the computing and data processing technology we have today was not possible then. Now you have the technical solutions to put in a massive amount of data, that you can collect easily, and process and analyze it in quasi-real-time. These were the enablers that were not there in the past, but I would not see them as roadblocks because yes, the utility industry and power industry are quite conservative in that area. For those operating the grid over the last 50-60 years, the equipment was reliable. If you just put a transformer somewhere, you can rely on it running for 30-40 years. It doesn’t need that much maintenance like some other equipment. Switch gear is the same.
But the demand increase, like we have today, is actually the game changer which comes with the enablers of the technology, and that’s why I believe as we go into digitalization, the roadblocks of the past are not the ones for today and the future.”
Paul: Have you learned any particularly interesting insights here at CERAWeek? Any takeaways that have changed the way you think about your job?
Adnan: “A few years ago, CERAWeek was maybe about talking about generation and addressing the energy demand, but now almost every session is about how to actually tackle challenges with digitalization. Everybody’s talking about data centers. Everybody’s talking about AI. This is not only a huge chance, but it’s also a huge challenge. Everybody has realized that these are big topics and now they’re something you see in almost all sessions at CERAWeek.”
Paul: Is there anything else you’d like to add?
Adnan: “You mentioned the roadblocks and the bottlenecks. We believe the utilities and operators in the future will be talking much more about digitalization than they are today. We also believe AI is something that is going to be crucial to power system operations in the future, which is why we are building an AI lab. As a company, we think AI is not just something to use in applications, but that in the grid of the future, the utilities are also data centers that make it safe to operate. With the speed at which grid operation is happening, we are not able to manually operate a grid from a control center, every minute dispatching and re-dispatching to keep the congestion low. That’s why I see a transition coming with utilities, small data centers themselves in the control room of the future.
We had a case in Italy where we put a vibration sensor on a line when a customer was surprised at frequent outages of insulators, connectors, and other equipment. We figured out with the sensor, and nobody knew before that, that there was a high vibration on the line causing a lot of stress on the equipment from the movement. And then you wonder why you have several outages and blackouts in certain areas? Data acquisition can open new horizons. You get a lot of data and analytics you weren’t expecting or couldn’t assume, and you get a lot of transparency, also.”
