Dominic Kohler with Siemens

SUMMARY KEYWORDS

simulation, dominic, model, siemens, grid, technology, customers, built, utility, renewables, data, user group meeting, network, talk, scada system, industrial, simulate, management, manager, modeling

00:04

Welcome to the Industrial Talk Podcast with Scott MacKenzie. Scott is a passionate industry professional dedicated to transferring cutting edge industry focused innovations and trends while highlighting the men and women who keep the world moving. So put on your hard hat, grab your work boots, and let's go.

00:21

Alright, once again, thank you very much for joining Industrial Talk. Thank you for your continued support of a platform that celebrates industry professionals all around the world. You are bold, brave, you dare greatly you innovate, you collaborate, you solve problems. That's why we celebrate you, you're making the world a better place. That's why we we just love you on Industrial Talk. We are broadcasting from DISTRIBUTECH here in Orlando, Florida. And it's day three, wrapping up, people are still milling about the place, it's still got a lot of energy of a bunch of problem solvers here in the world of energy and utilities. That's what it's all about here. Big time. In the hot seat. We have the doctor, Dr. Dominic Koehler, he's in the hot seat, we're gonna be talking about a new product offering coming from Siemens, and he's gonna go into details. I don't have to do that. So that's that's his responsibility, not mine.

01:16

I'm fine. It's great to be here. Thank you. Great to have

01:19

you. You were all over in the other. How far away is

01:24

that? Half an hour? Drive by car? We're there with our user group meeting for all portfolios with all the all the utility planners. Planners? Yeah,

01:35

yeah, we were. You had a wonderful event. I was surprised at how many last night right. Yeah, it was great. We went to Epcot.

01:41

Yeah. We also had the meter data management colleagues and protection. user group meeting. Yeah, it was just,

01:48

it was off the chart. All right. Before we get into the topic, Dominic. Was that user community? Has user

01:56

group meeting we call it was it Yeah, it was a productive. I actually loved it a lot. You know what, we have this network model manager that we are talking about today. But we also released like one and a half years ago, our first cloud offering. And it was just so incredible to see how the industry is picking up now we had like one and a half hour as a slot as presenting a bid to our customers, like first customers now using it. And we had so many questions like the 90 minutes were not enough, we had to cut it short in the end. And we had more questions in the break. So cool. It's cool to see how the industry picks up on the cloud and the new technology. Right? It's, it's

02:31

it is awesome. All right. For the listeners out there. Let's, let's cut to the chase. Dominic, give us a little background on who you are. And then we're gonna go into this network modeling manager and what that's all about?

02:42

Yes, my current role is I'm the product manager, the Head of Product Management for the power system simulation suite at Siemens. We are part of the business unit grid software within Siemens, and we help our customers plan the power grids, we do basically modeling and simulation, so that our customers can study how the power grids behave. Yeah, go ahead. No, sure. And you asked me who I am, right. So that's

03:09

where I was going. I'm going, Oh, great. You

03:12

just, you just talked about the doctor coming, right? The US guys usually make fun of the Germans always write this on the business. So I'm a doctor in mathematics. Actually, I used to do a lot of simulation technology back then in my PhD times. Back then we simulated water grids, you know, and now we're now we're looking at the electricity side. In the meantime, then I discovered my passion for for the business side and was a consultant. And now product management is for me bringing those two worlds together the technology and the business side and talking to customers and really identifying how this technology can be used.

03:50

Here's, here's the challenge that I always have. And this was the segue into your your solution just because it is a dynamic world out there in the world of utilities. It's a dynamic, it's distributed. It's I as a system operator, I wasn't one but I know what the system operators. What a challenge. What an absolute, you have all of these renewables. And if we're trying to be net zero in a certain period of time. How do you ensure quality of power, achieving those objectives, distributed energy, all of that stuff? Take us through why from from a network modeling manager just start taking us through

04:37

that what you're asking is the holy grail, right? Yeah. How do we do all of this and you know, we have one contribution there. One contribution, but one of the things you want to have as a utility to model this, to start looking into how renewables help to the grid, and how the challenges can be overcome is actually to first of all model how the grid looks like. If you want to install the new Renewable plants a wind farm, so you better model it and simulate what's going to happen. And whenever you want to really do analysis and simulation, all these kinds of things, the starting point is always just knowing what is where what those parameters are. And that's what actually Model Management is.

05:17

I like the fact let's see if I can, can your solution take existing asset base and an existing asset base, Florida got the solar's I got, I've got all of these renewables that exist, and they're on my, my grid, can we can we perform the simulations to optimize the way those interact with what it's

05:39

about? That's what it's about. So the starting point is the as is that as built modeling, right? Yep. And then you have as a grip planner, for all the multitudes of different scenarios, you have 1000s of scenarios, there is more load there, there is more renewables there, and so on for all of those scenarios, you have then ideas on how you can improve the grid to actually then include those new loads or generation. And now for this multitude of all those different changes to the grid, you somehow have to track them and manage them. This is what Model Management and is all those changes all those incremental changes as to how they compare to the base case, managing this over time. That's the Model Management basically,

06:21

see, this is so fascinating, because a number of the conversations I've had here at DISTRIBUTECH, have been wrapped around the speed at which the market is changing, right? Like my conversations associated with what was taking place within the utility energy space, two years ago, is completely different than it is today, two years, that's what we see as well. And when we start talking about Model Management and simulations, that's an ongoing thing. It

06:53

is an ongoing thing. And that's actually a good point, this is not new. So we at Siemens have been offering a Model Management solution for 1015 20 years. But the thing is, now we are now releasing the next generation of network model management. And the change basically, is we make use of all the new technology that's out there. So we have a next generation UI, we have it cloud ready, we do have interoperability. So in the past, this was all manual stuff that you had to do now it's automated with API's, really using the new technologies. That's the that's the change, that our customers also rightfully are asking from us. You know, what I just this morning, I'm coming out of a session back in our user group meeting. And one of the one of the one of the talks was about how they want to use the new model, network model manager. And then this guy told me Look, Dominic, the model is always wrong. But sometimes it's helpful, help us get more helpful. And I just love this, right. So there is there has always been a model. And now we want to make it better higher quality so that you can really run the simulations that are needed. I just love this guy is one of our customers there. Yeah, the

08:06

the one you say we took at the look at the as is or as built? And you have to run those models, I would imagine, how do you begin to feed the model with specific data, there's there's the as Bill but then there's, there's the economic component of saying, hey, this area is building out, Hey, we see a projection that there's going to be XYZ over here, you have to take that all into consideration, right? That's

08:34

actually true. So you have to look at the transmission and distribution side a bit differently. So the transmission side is traditionally you you take network models from asset management tools from SCADA systems, and then you feed this into a network model manager, and then you make sure with this, what people then like to call the single source of truth or a digital twin, they then take the data and then they feed it into all the systems into planning into operations into protection systems, and so on. So that's the transmission side, on the distribution sides, the change is actually much bigger. And there is so much more happening that has not been there in the past and also for the Model Management that's different. So on the distribution side, there is no sometimes no existing grid that you have in in a SCADA system or so. So what you have to do there is start from a GIS system or so take the geographical data about what is where and then start to develop a network model that has an electrical parameters that you can then start also using for simulation.

09:34

Do you have to upload the assets into that is like there's modeling, but that's a trance. That's a physical transformer. This is a reclosure this is you know, a real, all of the assets. You have to pull all that stuff. That's what

09:53

you're modeling. So look, this is the interesting thing about this. This is not an a model of an I said with 1000s of parameters for one acid, this is rather like 10s of parameters per acid. But then for 1000s, hundreds of 1000s of acid in the power grid, it's a huge, huge system. People actually say this is one of the biggest machines, or if not the biggest machines that humankind has ever built such a power grid. Right? So modeling, this is on the right level of of abstraction that you can still do the simulations there. I actually find it so fascinating, you know, with my background in simulation those years huge systems. Yeah, that's, that's cool.

10:31

The huge systems that have been around for years and years and years and years and years, and that have certain ratings and degradation and all of the stuff that's associated with it. I don't know how you how, how does the utility let's just use them as a, as a user, it's

10:48

our primary customers, right? So

10:49

you take I'm a utility? And how difficult is it to say, hey, Seamus, come on over, we want to begin doing this because we're having whatever the whatever the business reason is, take us through that sort of process of working with a utility,

11:06

right? So that is actually for a utility, a big, big question, how do you because this can be a major effort, right? So how do you get value out of a pilot project or so immediately. So for example, one of our big customers is fingered in Finland, but also AEP in the US. And they are all starting small with one concrete value add. So for example, and they start connecting the SCADA system, the EMS to the network model manager and start filling the network model manager and then pass it on, say to planning. So then you have now the same model being used in operations and in planning. So in planning, you know, what is really the starting point for all your incremental changes to the grid or in the planning side, you can then commit to that's the change that we're going to implement, pass it on to the operation side, and then those guys can can use it. So higher efficiency, higher data quality. As I said before, this is also a lot about automation. In the past, those guys have been sending emails back and forth. And well with this model manager, REST API is you can get access to it right away, pass it on notifications, all this good stuff. You know what, actually, when we introduced this, two days ago, or so we had our first big lounge about this. And one of the things we said is, you know, I said it before it was in simulation, right mathematics and so on. You might have heard about Alfred North Whitehead. It's a 20th century mathematician. And he actually said, the one KPI or one indicator of how a society evolves is how many processes can be automated. And if you think about this, this way, this model manager, automating how data flows in the in the different systems, it's quite remarkable what the whole industry is doing here. Like the level we are, we're getting to, it's amazing.

12:50

But it has to happen, we have to be more efficient with the with the network, our grid, if we're to achieve, because there's so many, there's so many points of pressure impacting the grid itself, utilities, there's just constantly changing. You have to drive out in efficiencies as much as you possibly can get to a point where you know that the network is operating at its optimal. I very much agree. And then and then and then go from there, but you have to drive it out.

13:25

I very much agree. So it's about efficiency. It's about data quality. Yes, it's finding the right spot for where to include all those renewables, having the right simulations all accessible then at the fingertip. Basically, what do you want to do?

13:41

How do you deal in the world, though, this, this draws upon my past experience with utilities. What is in house, the diagrams, the the maps, all of the stuff that's in the house, can very much differ than what's out there in the field. That's a great point, there's a disconnect. It's true. And these don't get updated. But anyway, how do you deal with that,

14:10

that's the s built versus the S model status, right. And this is actually exactly one of those points. So the value that our customers get out of this. And that's always the center point for what we're doing. Right? The value is that you get an aligned knowledge across the company of what actually is built out there. And then you start from this making all the changes. So the planner knows that's really out there in the field and starts making the changes there. The operations guys is starting from what is really built, they're starting to evaluate the switching states or the evaluate switching strategies and so on, all based on what is actually built out there. So this is this is why this is actually is so so important for us. You

14:50

can't you can't upload or whatever connect to bad data. You can't say you want to make sure you want to validate that that that switch was out there at that particular location. And that has that type of rating, right or whatever it is, or that transformer that pole top, whatever it is. And validation

15:08

can have many dimensions. So validation can mean you kind of first of all, you check that the data fields for the parameters are not empty, then you want to run a power flow and want to see if this is really a good configuration of the parameters that you have. So, this is kind of the things we are then putting on top right. As you rightfully say, the parameters need to be correct, the data quality needs to be there otherwise bad data in that data out right, then this simulation is not really helpful.

15:40

It's Go ahead.

15:41

No, I was I was just thinking, the other thing that we are now doing is I said it before, right? We have we have been offering this network model manager, and this is now where my engineering heart starts starts beating. We have been offering this for 1015 20 years, right? So there is has been almost a monolithic software. And we don't rock solid and proven in implementation by customers. They're using it and very happy with it. But now the challenge with this new technology for us really was and how can we now take this monolithic software, make it modular, slice it apart, right and then introduce our new technology around it kind of wrapped in certain pieces, put this together, put this on a new cloud stack, have API's accessible there. So this is also what then our whole r&d teams together with our CO innovation customers have been discussing, in the end design software from engineers for engineers, right? This is, yeah, this is also what the discussions are now about

16:41

me, I'm super geeky, when you can simulate. So am I I can I'll just eat that all day long. I don't care what it is. Can I do some? What if I did this? What if it's this of what? And I placed this out there? What is it? And just does the model sort of help with that optimization? Like instead of me just playing around with the data? It says, so let me make a recommendation for you. And do this,

17:12

you have all the right questions, right. So I mean, right now, and that's what we've been doing in the industry for quite some time is the engineers use the simulation technology, and then use their engineering brains kind of in the magic to find out what the changes might be, and then start evaluating what they exactly need to be in detail and so on. And now if we now kind of look into the future, like where this can take us, this is then all those new technologies around AI and so on, right? Do we have the possibility to create an recommendation engine for the engineer, so say, as an engineer, you know, there will be a new power plant at this part of the grid. And then the system in the future might tell you, Okay, that's the place where it is, and you need to, you know, have three options, you might want to have a substation in addition there, or you want to have you know, those and those changes. And then as an engineer, you're just kind of check what the what the machine is actually proposing to you. And this is, you know, this is probably a bit further out. This is Gen Gen AI, but on the other hand, then Gen AI in the sense of generative AI, these are things we're also looking into, like how can the simulation system kind of help you with documentation accessible with natural language almost ready to type in I want to increase the samsam load there? Yeah. Where do I have to model this? Like, which, which which buttons to press in the software, and then you get some some kind of responses there. But that's kind of next things to come right with? We're, of course, piloting here and there and trying out things. But that's not in the market yet.

18:42

Yeah. But it has to happen, I think.

18:45

And it will, it will be our customers rightfully ask for it, right? I mean, that's the technology. This is also why we're now launching this next generation model manager. The software has been there for a couple of years, as I said before, but now with cloud and all those things. That's the first step we're introducing, but now it's a future proof technology platform is tech stack, that we can then build other things on top.

19:08

Yeah, I don't see if if we're, if if we're tasked to achieve a certain level of whatever the net zero and, and increase the capability of three acts, whatever the all of those, all real, all real, that that is an efficiency. Those

19:29

changes that I mean, everybody talks about the energy transition. And to be quite honest, this is why at least I myself, and many of my colleagues get up in the morning and do our work and love working with the software that we that we put out there for the energy transition, but everybody's talking about it. In the end those it's those those concrete instances of kind of mind blowing changes out there. So for example, I was just talking to a partner from Australia one day ago yesterday, right? We had a chat and so on, and then he told me they are I'm having so much new generation renewable generation in Australia, that they have to kind of start looking into building an own transmission grid just for this. This is just so many changes or another one, another one. Now we're in the US. You're right there is Hiko, Hawaiian Electric. Yeah, they they now have plans to go fully renewable, 100% renewable. And they are kind of now looking into how, how does the system actually have to behave? And so they're using some of the Siemens technology to assess what the grid is then to optimize with, you know, if you have renewables, those IVRS, you have 10 changes, and you have fluctuations from the AC DC coupling, and so on. All those dynamic phenomena you have to control you have to optimize. There is like these are the concrete instances that make me really think there is a huge, huge thing happening there on the transmission side. And even more so than on the on the distribution side.

20:53

Yeah. And the reality is, is that it almost gets down to something simple. I'm a, I'm a consumer of power meter sky, I go to my house, I flip on the light, everything's good. I don't want to come to my house. I don't want to go to my house and flip on the light and it flickers as does something that I don't like, and and something's wrong. And then if it happens with more people, then then whatever effort is being taken, you have to do this right. One of

21:23

our absolute one of our one of our users told me yesterday, we are here in Orlando, right? Yeah. So there is all this disney world around. Oh, yeah. And he told me when I was young, I always wanted to be one of those. I think you call it Imagineers. Right? Yes, yes. They build all those things. And then he said, yeah, it hasn't really worked. I'm not an Imagineer. But at least I make sure that Disneyland is powered and that it is energized. How cool is that? Right?

21:47

Absolutely.

21:48

Keep the lights on. Good. Yeah.

21:51

You're just absolutely. You're the bomb doctor. Oh, Doctor delivered, you are different. Measure. Hey, how can people get a hold of you?

22:01

I think the best way is just via LinkedIn. If you are, oh, yeah, for sure. You can perhaps put it on the on the notes of this interview. Yeah. Thank you so much. Shoot, Dominic,

22:14

you fantastic. All right. Listen to him. We're gonna offer all the contact information for Dominic out there on Industrial Talk. So if you're not reach out to him, because it's all about education, collaboration, and innovation. He brings all three of those components to you. So reach out document, and then do some innovation. All right. We are broadcasting from DISTRIBUTECH. It is in Orlando. And we're gonna wrap this conversation up on the other side. So stay tuned, we will be right back. You're

22:43

listening to the Industrial Talk Podcast Network.

22:52

Dominic, Dr. Dominic, make a note of that. Was he great? There's a lot of great stuff happened. I mean, when you listen to what's happening at Siemens, about the ability to be able to simulate to access that data, that information, and I'm telling you right now, in the world of grid management and, and what's happening in the utility space, this is vital. This is vitally important, right there, dummy contact information all out on Industrial Talk, you know that I'm not gonna hit you just need to connect. It's all about the connection game, right? Finding those incredible Sherpas to help you along on the journey. Dominic fits that. Siemens great company. All right, we're building a platform. Industrial Talk is that place you need to reach out to me go out to IndustrialTalk.com Say Scott, I want to talk amplify my voice. People. Be brave, dare greatly hang out with Dominic, change the world. We're going to have another great conversation shortly some statement