Listen: Next in Tech | Episode 41: IoT's Role in Energy and Utilities

View Full Transcript

Eric Hanselman

Welcome to Next in Tech, an S&P Global Market Intelligence podcast, where the world of emerging tech lives. I'm your host, Eric Hanselman, principal research analyst for the 451 Research arm of S&P Global Market Intelligence. And today, we'll be discussing energy, utilities and their interaction with the Internet of Things with Johan Vermij, a senior research analyst with our 451 team. Johan, welcome to the podcast.

Johan Vermij

Good to be here.

Eric Hanselman

Well, it's great to have you on. We've been talking a lot about various aspects of the Internet of Things or IoT in a number of different areas, all tech-focused in various forms. But when we get to energy, it really takes on new roles. What do you see as the most important factors?

Johan Vermij

I think IoT in energy is a game changer. It's the IoT sensors that provided telemetry and give you insight in what happens. And if you can count it, you can control it. IoT is the foundation of the smart grid. And if we add artificial intelligence and machine learning to the mix, we get the sentient grid, the grid that becomes aware of what's happening in the world.

It takes the data and learns from it to predict and fine tune. And that gives us the tools to take on a number of challenges in the industry. You may be aware that this week, the COP26 conference of party started in Glasgow. And basically, that's the driver that we're looking at right now for the next decade. IoT is the enabler for the digital transformation of energy.

Eric Hanselman

There's a lot going on at COP 26, really helping to sort out what we can do from a governmental and regulatory perspective. But IoT plays a big role in simply understanding where we are.

Johan Vermij

Yes, absolutely. It's across the world, countries and companies have committed to decarbonize, and many of them have made net 0 emission pledges. And in the oil and gas industry, many companies are investing in new carbon-capture technology, but IoT is the technology that's here already and can really help us drive down emissions empower consumption.

Eric Hanselman

Well, if we're well instrumented, we ought to be able to understand both where we are in terms of output, but also hope to optimize and build efficiencies so that, that energy that we are using is used more wisely.

Johan Vermij

Absolutely. We're all aware that oil and gas production and consumption comes with a lot of CO2 and methane production. And -- but it's an illusion to say that we can stop using dino juice in a few years. I mean, the transition needs time, lots of cars, ships, factories will need fossil fuels for years to come. So it's an allusion to shut down the oil majors. So what we have to do is reduce the emissions on current production.

Eric Hanselman

So you talked about the sentient grid. Are we heading to an IoT Skynet controlling all of our -- both consumption and production?

Johan Vermij

Yes. Well, in some ways, it's Skynet. It has to have the ability to predict. I mean, currently, we're in the transition to renewable energies. And wind and solar produce power at all the wrong times, it's one big balancing act to fine-tune demand and supply.

And the sensor data can give accurate insight to power consumption of machines. And with AI, ML, we get the ability to forecast, for instance, by combining that consumption data with real-time weather reports as temperature has a big impact on how much energy we use. So if we know it's going to be hotter, we know people will turn on the air conditioning, and we know we have to generate more energy to keep the balance.

Eric Hanselman

So the sentient really is that balance point. And it's interesting. Both Dan Thompson and Perkins Liu who are on were talking about the challenges of creating a fully green grid. And I guess that's one of those things that I've always found is interesting about sort of the grid in general, which is it's inelasticity.

The thought, I think, for many, is that we've got a list capacity. We can create energy when we need it. We've got the ability to really manage the distribution because that's always been a simple and invisible process. But yet, matching production and consumption is a really complicated thing for the grid because it is pretty inelastic.

And I guess what you're identifying is that the sentient grid part is helping to forecast and understand where we have to be able to manage that production, anticipating consumption and bringing new suppliers online so that we don't take the grid down.

Johan Vermij

Well, with the renewable energy is being brought into the grid, you get power generation everywhere, whether it's in small micro grids, energy islands or state-operated wind farms or solar farms. And it really changes the way that the grid works because it's been designed from the start to be one directional.

It's -- you've got this huge power plant that's been ferried up with coal and produces a pretty constant load of energy power that is sent from the power station to the end user, one way. And now you get the generation at the plant of the end user who's using it himself or supplies it back to the utility. And that's a 2 way.

Eric Hanselman

And that creates that congestion that you're identifying, which is the issue of here, we've got this great, big, huge grid that was set up to have a small number of producers and lots of consumers. And suddenly, we've got all these producers.

Johan Vermij

Absolutely. And most of these smaller producers are maybe prosumers producing and consuming it at the same time. They're using and producing renewable energy like wind, which is great when you got a steady breeze, but you have to shut it down when a hurricane is coming or when there's no wind. And the same goes for solar.

It's in the day, in the summer, you have lots of power generation, but on a rainy winter day, you don't have that much. So it's really not predictable. Unless you take in that data, and you really know the angle that the solar panel is oriented and you can take in the weather account, then you can start to forecast.

Okay, this is how much this panel will generate during the next 24 hours or maybe even further on in your 14-day forecast. And that's what the utilities need to work with in saying, "Okay, if this is what we expect, then we have to have this much energy reserve."

Eric Hanselman

And so it really winds up being a matter of getting to a more predictive capability. Because as you're saying, you ought to know that at this particular time of year, with this particular set of solar producers, you can know how much solar energy is going to be produced. And it then account for it the way the grid operates.

Johan Vermij

Yes, indeed.

Eric Hanselman

And it sounds like a lot of IoT's value is in creating that operational telemetry. But all of this data and data in great volumes that we talk about, being able to model the performance of individual solar arrays on individual houses, I mean that presents some challenges. That's a huge amount of data that's being generated. How should organizations manage that shift?

Johan Vermij

Well, in the energy sector and that's powering utilities as well as oil and gas, we're talking about massive loads of data. In many cases, the data is already there. It's just not information. Power utilities, oil and gas companies have been gathering terabytes for ages.

For instance, all the time series data from the Scada systems and other machine telemetry. I even heard some estimates that oil companies only leverage a mere 1% of all the data they have. And yes, IoT is adding to that, like a modern wind turbine produces approximately 1 petabyte of data per year. So if the number of sensors keeps growing at this rate, and we need to transfer all this data to the cloud, we would get to the point that there simply isn't enough bandwidth available to transfer the data.

Eric Hanselman

Yes. You simply can't move all that data around.

Johan Vermij

Yes. At a certain point in time, all the radio towers are at full throughput capacity. And how do you move the data? So the question is, do we need to backhaul all that data to the cloud? Or can we process it at the edge?

Sticking to the wind turbine example. For instance, there is a company doing acoustic analysis and by listening to the sound that the blades make, you can hear if it's covered in ice or if it's hit by lightning or if it's suffering from corrosion. And if you download all those sound according to the cloud to analyze, you would be transferring gigabytes per day.

But if you analyze it at the edge, the turbine itself, you only need to backhaul the results, which can be as small as a few bits.

Eric Hanselman

So we're getting back to those shifts to being able to just push metadata of the analytical results that we're accruing at the edge. And as our listeners know, we spend a lot of time talking about transformation at the edge, but this is one more example where you've got to have analytics just simply to deal with the data volumes.

And that can really be the thing that helps organizations to figure out how they deal with this substantially greater amount of data. Because you're saying a wind turbine is generating a petabyte worth of data a year, nobody wants to accumulate that kind of information. But if they can accumulate what are the important parts of it and pull that data volume down to a more manageable level.

And now they can do the performance analysis for how well is it running, get the maintenance information, which is one of those just core aspects of IoT, do that predictive maintenance, but be able to manage that, but that depends on having enough capacity close enough to those sources.

Moving on from edge, you were talking early about COP26. How do you think that's influencing some of the thinking around IoT applications? And where do you think that's headed?

Johan Vermij

I think when you started the IoT channel some 5 years ago at 451 Research. And at that time, it was emerging and people were tinkering about and finding their way, what could we do with it? And it's really maturing at this time. In the last years, we've seen an evolution with all sorts of new use cases coming.

Prior to COVID-19, it was about cost savings, predictive maintenance, obviously, during the pandemic, it was about enabling remote operations. And starting last year and this year, it's really, really taking on. The narrative is changing into energy efficiency and carbon emissions.

So with the targets set by COP21, known as the Paris Agreement a couple of years ago, IoT is our best shot to act on climate change and wider sustainability challenges, not just in creating energy efficiency. If you know how -- what the optimum conditions are for machines to operate, then you know how to adjust that based on, okay, it's getting warmer so maybe it needs less energy. And therefore, you cut emissions.

But also in preventing and early detection of wildfires and biodiversity monitoring are the use cases are really expanding into those areas.

Eric Hanselman

So it's really covering a pretty broad spectrum of capabilities. And clearly, coming out of COP26, there are a lot of concerns about managing and monitoring emissions, which, of course, we need IoT out there to be able to manage it. I guess when we start thinking about where some of those shifts start to go, a lot of this is being driven by now greater visibility for ESG concerns.

And the first of those environmental is, of course, smack dab in the middle of energy and utilities use and utilization and that overall sort of carbon emissions aspect of where this fits. But there's that next stage that we need to be able to get to where organizations are actually able to integrate all the data that gets produced and start to do that reporting piece.

Because clearly, one of the imperatives that's driving a lot of this and a lot of discussions at COP26 are that aspect of really understanding in more detail, what is the emission profile look like? And what are those realistic goals? At the outset, you were saying, we're not going to be able to get ourselves out of fossil fuels by just simply flipping a switch and next week, heading off.

It's going to be a gradual process, and it's one in which we're going to need to know a lot more about our environment and how this really is, how it exists today and how we can make it better.

Johan Vermij

Absolutely. And IoT will play a role in that. I think President, Biden said something yesterday at the COP26 about helping farmers to get the new technology to capture methane from the ranches and stuff like that. You can use methane from oil production or from farming or from mining and use that gas to convert to hydrogen, for instance.

And if you monitor the whole supply chain, you can say, okay, this is dirty, but maybe a little less dirty than that type of fuel source. So you really get into the supply chain and see how much CO2 emissions have been there. So it's about making that supply chain visible and auditable.

Eric Hanselman

Well, and ensuring -- for that supply chain piece, ensuring that you've got the ability to know -- if you're capturing methane, to know where those methane leaks are. And that's, guess what, IoT sensors will help you figure that stuff out.

Johan Vermij

Absolutely. And -- but beyond the governance, there's also the social aspect in ESG. And that will be a challenge as well because for a large part, our residential homes are responsible for the CO2 emissions through our heating. And we can stimulate homeowners to buy solar panels and batteries for energy storage.

But we have to be sure that it's inclusive. Is it only the well to do that can afford to buy these and reap the benefits? Or can we make the energy transition available to everyone?

Eric Hanselman

And that's that larger and more important question of how do we extend this broadly across society? I guess that's something that we start thinking about the future. As COP26 will wrap up this week. Everybody is going to head off from Glasgow after you they've knocked back an iron brew or 2.

Where do you see this going next? You're talking about expanding that -- those capabilities and bringing those greater benefits to a broader swath of society. How do you see IoT's role changing in the future? And what should organizations be considering?

Johan Vermij

Well, what we just discussed, it's focusing more and more on the ESG component, the goals. One of the use cases that's pretty new is we talked about predictive maintenance. Predicting when your machine equipment will fail, so you can shut it down in time and do planned maintenance instead of waking up and finding all your production is gone.

But with these evolving use cases, we can start thinking about emission prediction. Like, okay, what are the effects of the weather on your emissions? And what can you do to prevent that and optimize it? Whether it's in manufacturing, you need optimum cooling for -- or certain temperatures in your factory to process food and beverage.

If you start thinking about measuring those emissions and predicting the emissions for entire cities, you would probably know where you will run into, bad air quality or have the entire city covered in smog. So I think those are the use cases that, especially cities, need to focus on because cities, they play an important role in the energy consumption globally.

It's not just the energy use in manufacturing. And sure, manufacturers, they're pretty eager to adopt the technology because if it saves like 10% in energy by optimization, it helps them to meet their ESG targets, but it also lowers their energy bill. But really is the trick to extend that Internet of Things into the cities and into the residential homes.

We have the smart meters that tell the utilities how much I use without the need of them sending an engineer to take the readings. But it needs to get smarter. It needs to be able to predict what's going on behind the meter and what the demand behind that meter will be. So if we're talking smart home, usually, we're talking about gadgets for entertainment that make our life easier, but it really should be about saving energy.

Eric Hanselman

Well and about greater benefit for society as a whole. It sounds like what you're identifying is moving from what has been that reactive response to, wow, we've got a smog alert, we've got bad air quality days, to getting to the point where we can anticipate that, that's happening and really starting to roll back emissions. And really have that benefit extend to society at large because as that's something that the population benefits from when we've got the ability to manage it.

Wow, well, I appreciate all of these perspectives, Johan. It is so far-reaching and especially now when you've got the climate summit going on. A lot of these aspects where a lot of the pieces of technology, often we talk about, are barely esoteric. Here, we're in one that has some very direct impacts on society and the planet. So thank you very much.

Johan Vermij

You're welcome.

Eric Hanselman

And that is it for this episode of Next in Tech. I want to thank our audience for staying there with us. I hope that you will join us for our next episode, where will dig into more fascinating aspect of technology because there is always something Next in Tech.