Jan 15, 2020
| 2 min read

Podcast #78: Digital Industry: A Utility Perspective - A Conversation with Thierry Godart

With over 25 years of experience, Thierry Godart is a veteran of the digital utility space. Thierry has developed an extensive knowledge on several emerging technologies including IoT, AI, edge, and IT/OT. He has worked with a number of prominent organizations including (but not limited to) Intel, Schneider Electric, Siemens, ABB, and GE.

In our conversation, Thierry and I discussed digital technology’s impact on the utilities industry with Thierry elaborating on how the sector has changed over his career. We delved into keys areas including the evolution of power consumption patterns and grids. Thierry provided insight into what he believes constitutes a strong utility organization digital transformation strategy and offered his thoughts on what the future holds for the space.  

Recommendations:

The Grid: Fraying Wires between Americans and our Energy Future, by Gretchen Bakke

Low Carbon Energy Transition, by Kathleen Araújo

 

orange-line.png

 We'll notify you weekly about new podcast episodes, upcoming guests, and news. You can subscribe to the podcast and if you'd like to be considered to appear on the podcast contact us.

 

View Transcript

My name is Leif Eriksen, Insights partner at Momenta, and our guest today is Thierry Godart, a technology executive with a long history of driving growth and innovation in the utility industry. Today we’re going to explore how digital technology is transforming the utility industry, and what we can expect in the coming years. First off, welcome Thierry.

Thank you.

Can you provide some professional background; how did you get involved in the electric utility industry, and particularly use of digital technology by the industry?

Yes, and thank you again for having me. Since college really I’ve always been involved in the application of technology to the power industry. Most recently I spent four years at Intel promoting the computerization of energy and industrial automation system. Prior to Intel I worked at a large equipment manufacturer, like Siemens, Schneider, ABB and General Electric, in various roles but always growing the IT and OT portfolios for utilities and industrial users. So, I’ve worked with utilities as my clients for over 25 years, and also across the world. I was very fortunate to be directly involved in some of the major shifts of the industry which required innovation with digital technology.

Yes, there have been a lot of shifts haven’t there over the years, it’s been an interesting and exciting time to be involved with both the utility industry and digital technology. That experience you’ve had, how does that inform your views on digital technology where we are today, and where we’re going?

As a vendor of these complex systems to utilities, I share your skepticism on how a digital technology initially designed for the IT world can actually work in harsh, dangerous, and mission-critical applications. So, my background of understanding this legacy world, and the fundamentals of power system operations, allows me to put a critical eye on digital technology that are not mature enough. Then I become the trusted adviser that recommends proper solutions to utilities, who want to try digital technology to solve their problems effectively. To get there I do spend considerable amounts of time in learning the new technologies and analyzing the true cost benefit ratio. This analysis often results in significant modification and adaptation of these technologies, for them to work in the power industry.

Interesting. You mentioned something about the skepticism of IT, and it brought to mind another term that we have used in industry called OT, or Operational Technology. Just quickly, what’s your view on those different worlds, and how they converge or how they relate.

Mostly when we talk about digital technology, I’ve seen their origins in the IT world, so today of course we’re talking about using the internet, and the internet of things, and more and more using AI because people like Google, Facebook, have perfected technologies like deep learning, machine learning, and of course have perfected the cloud computing approach. Now, how these nice IT applications can truly work in operation environments where of course data is not clean, failure of systems can lead to catastrophe, that convergence often requires a bridge and a lot more maturity, because the integration of these IT techniques is not that straightforward. But absolutely there is a convergence going on, there is an obvious benefit to be ripped from IT into day-to-day operations.

Yes, it’s interesting because I can remember a time many years ago when a lot of the technology innovation was happening in the business world, the industrial world, and as you pointed out, a lot of it today is happening in the consumer-facing world, and we’re benefitting from that in the industrial world. But that leads to my next question which is, what are the biggest changes you’ve seen in the utility industry over the course of your career, what are the most fundamental changes?

One thing that doesn’t change is the fundamental role of the utility, which is to operate and maintain a very reliable infrastructure to deliver electricity to all at the lowest cost possible. So, utilities are not going to jump on the latest and greatest trend, just because it’s kind of cool. So, the shifts I’ve seen are really non-digital shifts, I’ve seen three major shifts since 1990. The first one was power electronics, the second one was gas and wind turbines, the third one is photovoltaics and battery. Those are non-digital innovation that have driven the need for digital transformation.

So, let me explain a bit. In the nineties we could not store electricity at scale, but we started controlling its flow by using power electronics, these are called fact device, and this essentially allowed to modify the physics of the circuit by using power electronics. Then within the emergence of gas turbines, very efficient gas turbines, and now of course wind turbines, and that combined with the previous innovation, modern transmission systems, really allowed for deregulated markets. So, that’s when you saw a big regulation event which is in the US which allowed the power production from the power transmission.

Then in recent years of course now we see a major shift with battery storage that will finally make solar power scalable and reliable, as a major source of electricity. So now we see another big regulatory initiative which is to allow people, even you and I, but most importantly industrial customers to generate their own power, and maybe to contribute to the overall society needs of clean power.

So, I am really fascinated with the fact that for more than a century, electricity was the only product that got consumed as soon as it’s produced, and that now a major shift is happening with affordable storage, and that will really fundamentally change the way we provide electricity to all.

That’s huge isn’t it. Of course, there’s still a lot of challenges there, but we seem to be getting closer which is as you say very exciting. A lot of what you talked about just there are the things that affect the utility industry itself and the production of power, how that’s changed, and production storage I guess as well. But how do you see particularly as we look forward, the changing consumption patterns, we’ve seen a lot of investment and effort go into improving efficiency and energy consumption, reducing electric consumption over time, but we also see a parallel particularly recently new sources of consumptions such as the growth of the electric vehicle. What are your thoughts on that, and how that ultimately impacts the industry?

Yes, you’re right. The energy efficiency initiative which started a long time ago, during the oil crisis of the OPEC in the seventies went on and on, and it got really better with LED lights, HVAC appliance and so forth, and this was a major disruption because from a utility point of view this was a pure reduction of kWh consumption, resulting in a direct impact on utility revenue, but really without mitigating the cost of operating the grid. So, essentially the utilities had to maintain the same infrastructure with a lot less revenue, just because for a few hours of the year the grid would see a peak demand, requiring its full capacity.

So, the next wave of energy efficiency was demand response, which is about giving incentives to people to not contribute to a higher peak, and eventually re-defer investment in great infrastructure. So, that has a very nice benefit for utility, because it does reduce some of the revenue, but more importantly reduce their capital expenditure. And now with electric vehicles, solar, and storage, we could finally achieve a much more efficient energy system where peaks are shaved continuously, because you could have a concept, we call a virtual partner. This is essentially batteries and solar, including charging stations that are now connected and dispatched at the right time.

If you don’t do that, one thing that could backfire is a very plausible scenario of people starting to charge their car, and creating a new peak at the wrong time, leading to for example starting a peak generator, a gas turbine, just because people are charging their car in the middle of the day. So, you have the need for digital information for data to be able to really monitor, predict, and dispatch these resources. So, the new consumption pattern will definitely benefit by shaving the peak, and making the grid a lot more stable, but it requires of course an intelligent system, more than automation, but an intelligent system that can predict.

That’s a good segway to the other question I had, which is, as you point out there’s an opportunity of a much more efficient and maybe even balanced grid, and hopefully I’m not putting words in your mouth there, but that’s the way I interpret it. It also creates a level of complexity that we don’t have today, although we’re more complexed today I imagine than we were not that long ago. But you’ve got all these changing point consumption, plus a distributed generation network, it creates a lot of complexity. How do you manage that, and how critical is digital technology to managing that?

Exactly. Like I was saying, these distributed resources are often not owned by the utility, and they’re not centrally controlled and dispatched. So, they could run too costly and inefficiently, and even risk of outage and black out. Then when you look at digital technology today, there are two aspects where they can be useful, the first one is to operate the grid itself, today most of the grid is self-protected with devices that are pre-programmed to shut off the power if a problem occurs. So, if a tree falls on a line, or whatever happens, various scenarios, we know how to do it very well, especially for standard power flowing from central powerplant to consumer houses. But that’s no longer sufficient, and these flows now are more complex and often unpredictable, so you need more sophisticated analytics, and new controllable devices such as inverters in batteries, and solar, can now be used to absorb these excessive flows.

So, not only do you need to predict the unpredictable, but the way you control and correct is with new types of devices, not the traditional, ‘I’m just going to shut down the power and then wait until the problem is fixed’. In short, it’s no longer just energy automation, but it’s really energy intelligence that is required to operate this more complex energy system.

On the other side of the meter, we also have digital technology which can help consumers better use electricity. We’ve seen at the residential level the advance of smart homes and so forth, but I’m also talking about industrial customers, people who run factories, data centers etc., they can really optimize for the lowest cost and the highest value by balancing their self-generated power with their agreed connected supply. We call them prosumers, so these prosumers require more granular data to better predict their energy needs, and then also to anticipate potential events; if they know there’s a storm coming they can reset their system and make sure they can survive much longer with their self-supplied energy storage, and energy production.

So, digital technology will enable intelligent devices on both side of the meters, the grid side and behind the meter side.

Very interesting, yes it’s truly going to be a brave new world. I guess that begs another question, and that is, if you think about the consumer side of the meter, how much of an issue is it that the utility themselves don’t control that, how do you see that playing out? And I guess the second part of my question in general is, do we have the necessary technology today, and if so what will hold us back? Will it simply be the cost, the rate, structures, what’s stopping us from getting there?

It is an issue, but if you look at what happened at the power generation side, and transmission side, during the event I mentioned earlier when regulators, around the world by the way, unbundled the business of power generation from the business of power transmission. You had transmission utilities facing the same exact problem, suddenly you have a large powerplant that’s no longer owned and dispatched, we created market systems and independent system operators to handle the transaction between power generation owner, and users.

So, I see a very straightforward parallel here now applied to the distribution system, so the lower voltage system, the last mile of the power distribution that comes to your home, to your buildings, and to your factories. I see where a similar concept can be used to create a distribution system operator, so this has been discussed, this has not been implemented yet, there’s some attempts in New York, Texas, and in Europe. But as you’ve said it’s much more a political and regulatory concern, than a technical concern, it’s very-very possible with today’s technology to do it. Now, after saying that, of course the big difference between transmission and distribution is the amount of data, you’re talking about a lot more data, and also the lack of predictability. So, you’re talking about systems that have to react fast in the face of unpredictable events, because now if you will the system is a lot more sensitive to failure, or to customers changing their behaviour suddenly etc.

Yes, plenty of challenges ahead. One of the things I wanted to talk to you about briefly is, there are significant differences between how utilities are operated around the world, how the utility industry is structured around the world. As some of these changes that we’ve talked about come into play, do you see these differences persisting, or will there be some convergence in how these power grids are operated in different parts of the world?

Yes, you’re correct, it always amazes me working with utilities across the world for 20 years, it’s like producing and delivering at first should be a very standard process, and not so different, but it is. There’s been engineers over generations designing unique systems and creating a legacy of complicated electrical systems. Actually, there’s a movie that’s just come out about the war between Edison and Westinghouse on DC versus AC systems, it’s a fantastic movie by the way for those interested in this world. It continued on, you have major differences, physical differences between how the grid is architected in Europe versus how it is done in the US. In the sphere of influence, you have this major difference, which of course prevents a complete convergence. But as a matter of how you design markets and how you for example compensate owners of distributed energy resource today, we should really push for a single technical and commercial guideline.

In the transmission world again, we kind of achieve a lot of that, it was painful, but we ended up having a standard data protocol, so a single way to describe in the digital world what a transmission system is, and that’s fairly worldwide today. So, now it has tremendously reduced the risk because you have less customization, you have less complexity, and you can use standard off the shelf technologies to run the transmission grid. So, again when I look at distribution levels, or even microgrid levels, today it’s still very much early and you see a lot of complex systems that are very proprietary. So, I would say to move the needle we’ve got to push this convergence and standardization, starting with the commercial guideline, the regulatory guideline, to be fairly uniform, and that often leads to a more standardized way of implementing an intelligent energy system. Eventually that leads to a much lower cost.

That to me begs another question, you sort of hinted at this earlier which is, what stands in the way of utility companies greater use of digital technology to move their business and the industry forward?

To me there’s two, on the technology side there’s still a lack of open interfaces between equipment and automation systems, and on the utility side a need for better data-driven operating procedures, so let me elaborate. For digital technology to work in the power system it has to be a well-integrated combination of safe and reliable hardware, with useable and secure software. All this at a cost that is justifiable for regulated capital budget. So, traditionally you’ve seen the equipment vendors like Siemens, ABB, and so forth providing the automation systems to utilities, this on the one hand is great because those systems work very well with the equipment and device of the same brand. But on the other hand, of course they tend to be closed systems with proprietary software, which is costly to upgrade and hard to change.

So you have a dilemma here, because your digital transformation usually implies that you had a data strategy that is transversal to your operation, a diagnostic to your equipment so that you can construct what we call digital twins, you can construct digital twins of your end to end process, independently of the equipment and the device you use. This layer of data acquisition system, sensors, gateways, data, IoT, would then use a standard model to build this virtual operation control center if you will.

So, again I think in the transmission world we’ve achieved a lot of that. If you walk into a utility control center to look at the transmission grid, you see that separation between the way the grid is represented, and the equipment that they use. But the distribution grid and industrial microgrid is still very closely related to the underlying equipment, so I think there’s a lot of effort for utilities to embrace digital technologies such as IoT and AI. They’re going to need to push the vendors, they need to specify equipment and devices that are open that have open interface and can easily connect to an enterprise-wide IoT system, and an enterprise-wide data lake, like a big data system, to construct the digital twins at the level of distribution grid, and event for microgrid.

On the other side, on the utilities side, using more and more intelligent systems requires more and more attention to data-driven process. So, at the early stage when we were starting to deploy energy management systems for running transmission grids, we had a lot of issues with the lack of data representation. So, the equipment was not representing in database, and it was very painful to create those new databases, the more intelligent the system the more it relies on accurate data, and therefore there’s a need to really deploy a process to clean and to maintain accurate data records. That starts the whole debate between IT and OT organizations within the utility, who’s now going to be responsible for what is essentially an IT process, but used in the operation world? Is the CIO taking over operations, or do we have an IT department dedicated to operations, but not under the CIO authority? So that’s over speaking right now.

As you said earlier, it’s an organizational politics issue as much as anything else, which then leads me to ask, you’ve had the privilege of seeing a lot of utility operations in your various different roles, what are the characteristics of a utility organization with a good digital transformation strategy? What makes one better than the other at doing it the right way, and embracing it fully?

I call them smart utilities, they usually have a strong vision, not over 20 years but over the next five years, and they have an incremental roadmap; utilities work a lot on an annual budget, they have to ask for money, sometimes rate case on an annual basis. So, a smart utility would have a five-year roadmap with annual incrementation and ranking the used case; where is the best bang for the bucks and let’s start with those kinds of incrementations. Of course, this requires a supportive regulatory environment, and a somewhat stable financial situation.

But from an organizational point of view I really see it working when there’s a cross-functional committee that’s got its own approved budget, and plan, and it’s all sponsored by the CEO. The cross-functional committee is usually presided by a Chief Digital Officer, or Chief Technical Officer, who knows how to listen to both operators, engineers, customers, let’s not forget the customers, and then regulators, and he knows how to create bridges across the silos of the organization.

So, with this empowered committee you have out of the box ideas that comes from the bottom – up. These ideas are based on sound data and facts and have instantaneous buy-in because they come from the people who are generating those ideas. This typically fundamentally transforms the way utilities operate, usually the committee with organized training programs. I was involved as a vendor being asked to come in and do demos without any kind of procurement process, it was smart for them to do a watch process, and say, ‘Hey, let’s look at what is possible, and learn from vendors for free, before we initiate any kind of procurement process, so we can really keep abreast of innovation out there’.

Then from an execution point of view what I’ve seen is, fast, speed; so fast project execution with very well-defined success criteria, so you define your matrices and you say, ‘Well, if we achieve that matric, we can rapidly scale and deploy to other areas of the company. So usually they would take for example a substation, and an area of the grid, try something, and if it works, they already know how to mobilize and deploy to all substations.

Recently I was involved in creating focus groups, this was more common back in the nineties where utilities would get together and address a problem together. EPRI had that role, it was often for a piece of equipment, for example when SF6 was banned, this gas when they had to come up with a new way of insulating breakers and so forth. But I think that EPRI could do a little bit better getting into the digital technology focus group, that would be really useful because now you have an industry-wide specification that could come to vendors, and really correspond to utilities, instead of utilities being asked to adopt things that were not really designed for them.

The last thing I’ve seen done well is, a proactive transfer of knowledge, so asking the vendor experts to come and train the trainer, make sure that they are internal experts that can take over and make their own decision on how to deploy that technology. So, having to go back to the vendors every time is not a good idea, so you have to build your own expertise and that’s really important.

Let’s jump to the future if we could and get your perspective on where you think we’ll be five years from now, ten years from now, and again the role of digital technology. Of course, it’s like every industry, we know it’s going to be big; it’s going to be important as you pointed out, data and data-driven decisions will be increasingly important for the industry. But maybe if you could paint a picture from your perspective of what the utility industry of the future will look like. You’ve touched on a number of these things, but maybe pull them all together and tell us what you think it will look like.

I think Bill Gates observed something like innovation always takes longer to get started, but often surprises in the speed they get skill into the mainstream, and I see that. We saw that in Germany when the German government pushed for solar, and now suddenly you have areas of Germany that has 120 percent solar capacity compared to their need, and that creates a new set of problems. So, I do believe in general in that aspect where for example, for batteries it takes a while to get utility to adopt batteries, and rightfully because as you said earlier there’s still a lot of technology innovation to be worked on. But then once it matured it will scale very fast, and that’s the characteristic of the utilities, that once they like a particular innovation and its proven, they move fast, and they can really have a very large deployment.

So, I think in the future climate change will impact utilities more and more, with both the frequency and intensity of climate events, like we’ve seen with the fires in California, the storms, the hurricanes, but also with the fact that consumers and politician are going to demand a complete decarbonization of the power system. So, utilities are being hit on both sides, and I do believe that carbon taxes in whatever shape or form will enable utilities to invest in energy storage, and solar powerplants, and further digitize the distribution grid. It could be that your solar panel on the roof is useful, but I see a lot more use of land by utilities to build utility-scale solar plant and battery farms.

I think in the future the grid asset would be instrumented using 5G networks, I’m hoping I’d have several examples of utilities really getting proactive in using networks, either by buying their own network infrastructure, or by partnering with a telecom company. I definitely see that with 5G, so 5G would be finally along with utilities to build an accurate digital twin in the cloud, and with that you would be able to predict failures, consumer behaviors, and really correlate all this in a way that as I said earlier, the power system runs smoothly even though you have all these uncertainties connected to the grid.

You talked about the concept of microgrids that are being run maybe independently of utilities, or do utilities run them for their customers, large industrial enterprise or whatever? How does that all play out, because again you’ve got this distributed energy infrastructure, it’s not clear to me who’s going to own that, and who’s going to manage it.

I do believe that we’re going to see a utility evolution where you have a consolidated regional operator concept for distribution, and a utility just like we saw in the transmission and generation world, you have a DSO that owns the digital operation if you will, so the monetary and the controlling of all aspects of generation and distribution, one owning almost zero of the generating assets. So, it is very possible to have independent owner and also aggregators, because when you mention microgrids it’s not a single individual microgrid that’s going to make a difference, but then you have this entity called aggregators that put them together, it now becomes a fairly stable entity that can market and trade on the distribution network.

So, the utilities role would be a regional grid platform on which markets can operate, so here again I’m talking retail markets where you have people providing resiliency service backup power, you have transactive energy amongst customers, so you can sell your excess power to your neighbors. All this will be done through a cloud-based platform, and the utilities will basically manage the green infrastructure, and maybe the IT infrastructure to do that and get paid on a fee per usage, or per transaction.

So again, in the short I see in my view a very similar evolution that we saw in the transmission world, that created regional transmission operators, similarly in the distribution world where a distribution system operator will come in and become the main function of today’s utility. We’ll see.

It’s a very interesting future, and it seems very logical and clear as you explain it. In fact, this has been a very enlightening conversation for me Thierry. Just to wrap up, if you have any closing thoughts, and we also like to ask our podcast guests if they have any recommendations for reading, you mentioned a movie earlier which sounds interesting. It doesn’t have to be around the subject but just what are you reading, what do you think is interesting these days, and we’ll wrap it up with that.

One final thought is, over the last years we’ve seen a drop in interest from college grads, and really attracting the workforce, especially the IT and the technology-type workforce to utilities. This is a very different world today, you see a lot of utilities with green modernization initiatives, recruiting technology people, data scientists, so to me it’s starting to become once again a very interesting opportunity for engineers and scientists to join the utility industry.

In terms of reading material, I recently read three books related to all this, one that was published in 2016, the paperback came out in ’17, it’s called ‘The Grid: Fraying Wires between Americans and our Energy Future’, it’s written by Gretchen Bakke. It’s very interesting because she’s an anthropologist, she goes back through the history and explains how the culture of America is linked to the grid, and how this evolution impacts all aspects of our lives. It’s a fascinating book.

The other to look at a more worldwide view of this transition is called, ‘Low Carbon Energy Transition’, by Kathleen Araújo. She’s going through examples, and the one that interested me of course was a chapter on France, since I am from France, France has a very unique power eco-system because its heavily reliant on nuclear.

The last book I want to acknowledge here, because it’s written by a good friend of mine Subramanian Vadari, ‘Smart Grid Redefined: The Transformation of the Electric Utility’, this was published in 2018. This book is a bit more technical, but very well written by an expert. I recommend those.

Thanks for that, they sound like great books. I think if someone were to read them all they’d get a pretty good picture of where we’re at today!

Thank you very much, this has been a terrific conversation, and I certainly appreciate your taking the time to speak with us.

Thank you very much Leif, take care.

 

[End]