The future of the power grid: Re-engineering for renewables

Posted: 4 April 2023


Decarbonizing the power grid is going to take more than just replacing fossil fuels with renewable energy sources to maintain our current power generation capacity. We have to increase capacity as the population grows.

The first episode of our three-part series, “The Future of the Power Grid.”




Electrical energy is the most valuable real-time commodity in the world. It’s the commodity that differentiates us from anarchy. I actually stood inside a large equivalent of a Home Depot during a blackout 20 years ago, at night, and I saw looting break out in 10 seconds—soon as the lights went out. And that’s how quick you go from organized civil society to anarchy when you lose this one commodity.

We can’t afford to get it wrong for 10 milliseconds, because if the system is out of balance for 10 milliseconds it can collapse. It's frightening, but as soon as that commodity is lost in a blackout, lives are lost. So, these are all the nightmare scenarios that we worry about, and we are going to transition into uncharted territories. Data will be a defender and a way of managing that transition like never before. 


From AVEVA studios this is Our Industrial Life, the podcast that brings you stories from the essential industries and investigates how data and technology are shaping the future of the connected industrial economy.

I’m your host Rebecca Ahrens, and today we’re going to be kicking off a three-part series that we’re calling, “The Future of the Power Grid.”

Over the course of three episodes, we’re going to hear from three different industry experts about a few of the most pressing energy questions of our age: What will it take to decarbonize the energy grid? What challenges stand in the way? And what role can new industrial technologies play in bringing us a carbon-free, energy-secure future?

First, let’s start with a quick lay of the energy landscape.

Here in the U.S., the Biden administration has set a national goal to transition the grid to 100% carbon-free electricity by 2035. It’s a tall order, but the U.S. isn’t the only country striving to fulfill it. The rest of the G7 have also pledged to achieve “predominantly decarbonized electricity by 2035.”

It’s not technologically impossible, but it’s definitely not going to be easy, either.

Two of the biggest challenges we face are:

  1. Replacing our fossil fuel generation capacity with renewables isn’t as straightforward as you might think. And…
  2. Just replacing what we have isn’t going to be enough. We’re also going to have to increase our generation capacity dramatically. In other words, we’re going to have to have more electricity available on the grid—and fast.

According to the UN, the global population is expected to grow by 2 billion in the next thirty years. As the population rises, so does the need for electricity—not to mention the demands that will be put on the grid from other kinds of societal changes.


I would have people, first of all, just think about what the electric vehicle is going to do to the power. If you took a big urban area—like the LA basin—if you had a million EVs being charged, you're using more power than the CAISO has ever delivered.


This is Pat Kennedy, the creator of the PI System, software that’s widely used throughout the energy sector. Pat’s been studying the power grid and making software to help power producers and grid operators all over the world run their operations for decades.

Thanks for being here, Pat.


Glad to be here with you today.


And a quick aside for our listeners, the CAL ISO, or CAISO as it’s often called, is California’s independent system operator. CAISO oversees the operation of California's power grid and it’s one of the largest ISOs in the world.

Ok. Back to EVs: that’s a lot of power, isn’t it? Do you know how much that adds up to?


Cal ISO I think was a little over 50 gigawatts during a huge heat spell back in maybe 2006. And you add the million cars together and you get about 65 gigawatts.1


Wow. OK. Let’s put that number into perspective. Your typical lightbulb is between 60 and 100 watts. A gigawatt is one billion watts. So 65 gigawatts …that’s a LOT of lightbulbs.

This increase in energy demand is something we’re going to have to figure out soon. Here in California, the Advanced Clean Cars rule mandates that by 2035, all new vehicles sold in the state must be EVs. New York plans to follow suit.

But this shift to electrified transportation isn’t just happening in California or New York. It’s happening all around the world. The EV revolution is already well underway.


The electric cars just changed the game, totally. Because (A), they're so big from a power perspective. And (B), they got their own storage. And (C), they move, which in the power grid business—this is a horrendous issue because they move from the city to the homes in the evening. So, they tend to synchronize themselves around start-of-workday.

Then, what do they do? They unplug, they drive home, they plug it back in. And they turn on their appliances, they turn on the TV, turn on the lights, and they start to enjoy their evening. And so that results in a double peak. And the second peak: all the sun has set and the wind’s calmed down.


Meaning, those standard renewable energy sources aren’t producing energy for the grid anymore.

This might be a good time to introduce an important issue that gives people who care about the energy grid (which, by the way, should be everyone) a lot of grief. This issue is known as intermittency. For the most part, renewable sources of energy are intermittent, meaning that source is not always available: the sun isn’t always shining, the wind isn’t always blowing. We can dispatch power from, say, traditional gas-powered power plants practically whenever we want. Renewable power sources, on the other hand, aren’t so straightforward.

The topic of intermittency and the problems it causes the grid can get a little complicated, so if you’d like to understand more about it, you can check out the third episode of our “Future of the Power Grid” series, where we try to untangle the intricate relationship between DERs, intermittency and grid stability.

So—renewable energy isn’t always available when it’s needed. Another problem is that it’s also not always available where it’s needed. If a whole city relies on renewables to feed their power grid, what happens on cloudy days, or when there isn’t any wind? What about parts of the country that stay cloudy for months at a time?

One way grid operators deal with the problem of intermittency in the short-term is to use what are called, “peaker plants.” Power producers use peaker plants in conjunction with baseload power plants to support the grid at times of high energy demand, or peak demand—hence the name, “peaker.” The time of peak demand depends on the location. In hot areas, it’ll often be in the late afternoon when temperatures are high and residents are running their AC units. In colder climates, it’s often in the morning when households are starting up their heaters.

The problem when it comes to peaker plants is that they’re significantly less efficient than the combined-cycle plants that provide the bulk of our power today. In fact, peaker plants, on average, require about 50% more gas to operate.2 Right now, we use peaker plants to fill in the gaps when renewable sources of energy aren’t available. But they aren’t a long-term solution when we’re trying to completely decarbonize the grid.


Now—if you were a developer—let me give you two options, and you tell me which you would pick. One of your options is that you have to site and entitle a power plant in downtown San Francisco.

[RA laughing]

And actually, people do chuckle at that. If you've ever tried to entitle something in a major city, then it's, it's like 20 years. It took 18 years to build—and several blackouts—to build a transmission line across upstate New York. And there you were disturbing the cows.

The other option, though, is that we could take software, and better manage the resources we have. Because right now, the grid treats the lights in the parking lot as an equivalent use of power to running heart/lung machines. And those have different needs.


I don’t think anyone would argue with you there.


And so by using the intelligence of the grid to better allocate those resources, then perhaps you don't need that power plant. It's really relatively straightforward to use controls instead of building peaker power plants.


Well, then put me down for option #2: smarter resource management. But I wonder if you can help us understand how far can smarter management really stretches our power resources. Are we talking about a huge difference, or maybe like a medium-size difference?


When I was growing up, which was a long time ago, the local power company gave my father eight dollars a month to put an interposing relay in his air conditioner. And he did that. He thought it was a great savings. And what they did is that, at any one time, they could take out his compressor for 15 minutes. The blower ran, the thermostat ran: people in the house didn't even know the air conditioner wasn't running. And then they would sweep that through the territory. And unbeknownst to anybody, they were managing the peak loads to clip the peak loads off.

And one of the big power companies once gave me a statistic from their research. And that was if they could cut 100 hours off the peak loads of the grid, they could save construction of a power plant. So these are very, very high-benefit projects.


Wow. OK, so these little savings here and there really do add up.

But what does it really look like to use advanced, industrial software to create smarter grid management? To tackle that question, let’s introduce another guest. Joshua, would you introduce yourself?


So my name is Joshua Rhodes. I'm a research associate at the Webber Energy Group at the University of Texas at Austin, where we study macro energy systems. So how, you know, the electricity sector, natural gas sector, water sector, food sector, agriculture, everything, transportation, how it all interacts.


Thanks, Joshua. Thank you for being here. I want to ask you about some of the new technologies emerging today that can help us better manage our electricity resources as we start moving to a decarbonized power grid. One of the things we hear a lot about these days—and I know this is a bit of a buzz word—is “smart grid.” Can you just tell us what that is?


So I really see the smart grid, generically, as basically a bi-directional information overlay to our resource network, or resource systems. So, our electricity system, our natural gas system, our water systems.

So instead of sending a commodity or, you know, an energy source or water from just from point A to point B, you know, we send that along with information about that. And then that information can go two ways. So it can be actionable. And so the grid can maybe not only send you electricity, but it could, if you allow it, make smart decisions on how you consume that electricity.

Like, if you’re not home and you sign up for a program that allows your thermostat to be set up a few degrees, or down a few degrees, depending on what season it is, allowing us to more smartly use our energy, use our water, to tell us when there are leaks, or, you know, your air conditioner is not performing as well this year as it was three or four years ago. Maybe you need to have service or something, you know, like that on the system. So information that can, flow both ways, be actionable and actually increase efficiencies, or decrease costs. I mean, I really think that's the crux of what the smart grid is aspiring to do.


So, I mean, I know some of these kinds of technologies—like specifically when you're talking about, you know, maintenance, just to take an example, say—like, industrial companies have sensors in place, and they have devices for gathering data about their various machines, and then analyzing it. And then they can get these automatic notifications when something seems to be going awry and it needs maintenance.

So is the idea, you know, equip homes that are using these devices that draw tremendous amount of energy with these same things? And so if it's operating less efficiently than it should be, instead of the user just kind of going on their merry way, and never knowing that whatever it is—their refrigerator—could be running more efficiently than it is, we have that data, we have that insight. And whoever it is, whether it be the homeowner that's collecting that data and looking at it, and you know, making these choices, or if it's the—I don't know if it would be the power company that's collecting data from these smart devices—but somebody can do something about it more proactively?


Yeah, and those relationships can be, you know, third parties: companies that analyze data as well. It doesn't have to be the homeowner, who may not know what they're looking at, or the utility. But I think the thing is, yeah, for large, complex systems, you know, oil refineries or manufacturing facilities, I mean, it's someone's job to make sure that they're operating at a certain level of efficiency.

So if we can actually use data to say that, OK, today's a 95 degree day, and your air conditioner is running at a certain efficiency, and four years ago, on a 95 degree day, very similar, it was running at a higher efficiency, that could be an indication that maybe there's a filter that's clogged, or a compressor that’s going out, or something, you know, like that, that would allow us to maybe make smarter decisions. So we could do maintenance instead of having to replace something, or letting it break on 100 degree day, or something like that. You know: preventative maintenance and things like that. If we can use data to help us make better-informed decisions, that's something I think the smart grid can be useful for.


And, of course, if we’re using the energy we already have more efficiently, maybe we don’t have to build and incorporate quite as many new resources to meet the growing demand.

So, a smart grid is a large, utility-level power system. Now, if we shrink our focus way down, that brings us to another technology we’re hearing about more and more these days: the microgrid. I asked Pat if he could explain what that is.


So what a micro grid is, it basically handles the functionality of a grid, but on a much smaller basis. I'll give an example. Consider that you have an office building with solar PV on the top, and you have batteries. And you want to implement, say, a high reliability bus in your building. This is totally on the downside of the meter. It's behind the meter: power companies can't even see you doing this.

So what you do is, you take your batteries, you use your control circuits to store excess power in those batteries. And then in the evening, as your solar tapers off, you start bringing power out of the batteries so that you have a nice, constant presentation to the power company.

You could do the same thing if you wanted to say, “I want to island the building.” What that means is that you say, “I see fluctuation coming in the bus. I don’t want to participate in this. I want to separate from it.” And then you have to consider: how do I re-attach later, when the problem’s gone?

You go through these functions—these are what a microgrid does. It uses the intelligence to actually manage all of these resources that sit behind the meter.


Pat brings up another important technology here. He mentioned rooftop solar panels, and other resources that are “behind the meter.” I want to tell our listeners here that these behind-the-meter resources are sometimes called “DERS,” or “distributed energy resources.” DERs might be small-scale energy resources individually, but together, in their cumulative capacity, they will likely play an important role in achieving the energy transition.

But, they also pose their own challenges to grid stability because of their intermittent nature—and because they are usually behind the meter, as Pat said, which means that the utility can’t see them. You can check out episode three of this series for more on DERs.

OK, Pat. So, we have the microgrid, which is managing energy resources that sit behind the meter at a particular site. And then above the microgrid, we have the power company. What’s next? How do the two work together?


What I see coming is that there are layers of these things. Because if you did, in fact, want to manage a grid using intelligence instead of using power plants, you can send out the signals, but you have to have something ready to accept those signals.




Like, if you go to a large building with 50 or 100 EV chargers and air conditioners and windows that tint and all of this, you can't realistically control all of those from a transmission control center.

But you could send out a signal saying, “we're in trouble: please reduce your load to what you can reduce safely.” And then at the level of action, which would be the building, your micro grid would take that instruction and it would reduce power. Maybe it would turn off the lights in the parking lot. Maybe it would do a…turn up the air conditioner, tint the windows—whatever it has to do, it could do these things locally. So you have this micro grid that sits at the lowest level that basically is a control circuit. And it's accepting these pieces of information and taking action at the load.


And so then is there another layer on top of that?


The next level up you have the substations. And in the Bay Area, I don't know, there's perhaps 15 big substations. These substations now have these feeders that go out to all of these intelligent buildings. And they accept power from the grid to go and implement what they have to do for managing their work.

There's got to be a microgrid around that as well. Because something has to be able to say, “I need to go to solar resources here. I need to go and shift power over to here. I have these different reliabilities.”

Then you have the distribution grid, which has its own control system. Then you have your transmission grid, which has its own system. And so you have these layers of grids.


Microgrids, like in the example of the big commercial building that Pat described, are generally considered to exist within the boundaries of a single site. It’s one site, managing a system of DERs: the solar panels, the EV charging station, and so on.

Now, you can also use advanced industrial software to aggregate the generation and storage capacities of these DERs, forming what’s called a “virtual energy network.” You might call it…


A virtual power plant.


This is David Bartolo.  He’s the head of asset intelligence at AGL, where he leads a small team that concentrates on operational technologies supporting power generation assets—especially distributed and renewable assets, like solar and wind.

Again, we’ll go into virtual power plants in greater detail in the third episode of this series. But, just to get us rolling, David, could you tell us what a virtual power plant is?


So, AGL has entered the area of virtual power plant. We call it a VPP. And our vision of virtual power plants is at the residential level: to have many customers who have solar systems on their roofs, combined with battery storage technology. And to be able to aggregate all that data, understand how much energy we could supply to the network at any time, and also how much energy we could take from the network at any time to charge those batteries. And dispatch them as what we call a virtual power plant—so a gigantic battery—to support the network, either at times when there's not enough energy discharging those batteries, or at times when there's too much energy, especially during solar max, to charge the batteries for times of shortage, especially during peak arrival times when everyone gets home from work.


Do you think that this distributed model will make the system ultimately more resilient? Or is it hard to say?


It's hard to say. I've been brought up with massive machines very, very reliable, centralized on the thickest part—the most, you know, heavy part of the network, that could always hold up and could dynamically change to support the network. This is what the system was engineered around. The system was not designed and engineered around batteries in houses, distributed across the state, holding the network up during times of emergency.




We have to invent that. So the answer to that question is yet to be invented.


It’s kind of hard to wrap your head around how massive the change really is, and how much work it’s going to take to get there.


There’s so many equations and so many dynamic electrical problems—deep, dynamic electrical engineering problems—that have to be solved: voltage control, frequency control, regional congestion. You know, what happens in a street, your street, where your street is producing 20% more than the transformer at the end of your street can push into the network? How's that going to be managed? Right? All of these things—we need to invent those solutions.

It's the biggest change in this industry since the 1890s, when we first electrified. The next ten years will represent the biggest change since then. And if we're not set up correctly, the data problem, which we've explored in detail—I think you're going to find yourself at square one, again and again.


Alright. So, we’ve been through a lot so far on this episode. Let’s take a moment here to quickly run through the challenges and complications we’ve covered so far:

1. We have renewable resources that can only produce power intermittently.

2. We need to get better at moving power around because, for now, renewable energy sources are not equally available everywhere.

3. We have an increasingly unpredictable and shifting energy demand landscape with the rise of things like electric vehicles.

4. There are a growing number of small energy resources that are “behind the meter.” These resources will need to be controlled and communicated with so they can all work together to support the system and keep it from going down.

Like David said at the beginning, when the power goes down, it can be incredibly hard to get it up and running again. And in the meantime, people can die.

Controlling the grid by using technology and data will go a long way toward helping us deal with these issues. But how are we going to coordinate all that data? How is it going to be collected from all these different entities and exchanged between the right groups? How are we going to make sure that just the right amount of power is put onto the grid at the right time in the right place so no one ends up tripping the whole system off line?

There’s one more technology we haven’t discussed yet—one that will likely be a big part of the solution to secure data exchange that will need to take place between all these different entities.

It’s called the industrial cloud. The ability to rapidly access, share and analyze data in a secure way will become even more essential during this energy transition. That’s where the cloud comes in.

We’re going to do a whole other series on the cloud and its role in the future of the industrial economy. But for now, just know that it’s definitely going to play a big part in the future of the energy grid.

Alright, so we have the cloud. We have more intelligent grid-management systems, like smart grids, microgrids. We have DERs and the virtual power plants. Is it enough?

I wanted to ask Joshua what he thinks. Is it a realistic goal—decarbonizing the grid by 2035?


I think we can decarbonize quickly. Texas is already 35% zero-carbon generation. And that's been changing quickly. That’s gone from 5% fifteen years ago to 35%.


Fifteen years—I mean, that doesn’t sound too bad. But what’s it going to take to get to that 100% mark by 2035?


I think, you know, if we had the right incentives in place, I actually think that's possible, too. And one of the reasons is because, like, the average coal plant in the US is 45 years old. And that's about as long as they last. And so that's one of the reasons that coal’s been retiring really quickly is that it's just coming to the end of its life. And then it's just being replaced with things that are cheaper, be they renewables or natural gas.

You know, as things retire—to the extent that we replace them with zero- or lower carbon technologies—I think we can move pretty quickly. Because we have a lot of things to replace coming up.


Yeah. And this is—I'm glad you brought this up—because, you know, there's a lot of concern over this aging infrastructure that you mentioned. I know you've written about this recently.

So, given that we need to make updates one way or the other—I know there's been talk about, for instance, right now, as you mentioned, like—we have the eastern interconnection, we have ERCOT, and then we have the western interconnection. And they don’t really communicate all that much. And people have proposed, you know, well, why don’t we try and make a greater entanglement of these—of these systems? Maybe that would go a ways in providing even more stability, or at least, you know, create more opportunities for power sharing, especially when you have sort of these pockets, right?

Because there are states where solar is abundantly available. And there are places where wind is abundantly available. And then there are places, you know, where there’s a lot more demand or load, but they don’t necessarily have those resources. So part of this is updating the infrastructure in such a way that we can move those variable resources to the places that they’re actually needed.

So, what in your mind—where should the focus be? Given that we have this infrastructure, it needs to be updated—this could be an opportunity to make some needed changes. What do you think are the priorities?


If I were king for the day—Infrastructure King for a day—I would really focus on transmission. Moving electrons around more freely, I think, is going to be the cheapest and best way to move into the future. Because, as you mentioned, there are places where solar is great. There’s places where wind is great. There’s places where solar and wind are great. And there are places where it’s poor. If we can move those more freely, it’s like any market that, you know, you take down barriers to. If you allow the movement of goods and services—in this case, electricity—in a less constrained way by building more transmission, prices will go down. And given current technologies that are most cost effective, we would build a lot more—probably—we’d build a lot more wind and solar. And we would move it around to places that need it.


We could talk about all the myriad challenges that are facing us, forever. But I also want to look past that for a minute. Let’s talk about what we stand to gain. What does the world look like once it’s running on decarbonized energy?

Joshua, let’s say it’s 2100. You’ve been—what did you call it—King of Infrastructure? OK, so you’re a very, very old King of Infrastructure. What’s your dream for the future of energy? What would you like to look back on and say we’ve accomplished?


Wow—going out to 2100. That’s a…[sigh] that gets into the realm of science fiction, I think.

Yeah, well, “Back to the Future” said that we’d have flying cars, and we’re not there yet. So, I mean, I guess we’ll temper expectations a little bit.


This is your dream world, Joshua. You can dream as big as you like.


Umm.. yeah, I mean, I think, you know, at the end of the day, one of the reasons why I care about this—and climate change is going to affect everyone, it’s going to affect those that are poorest first—those of us that are in rich countries will be able to adapt a lot better than those that are poor. And so when I think about the future, I really hope to see not just an electricity system, but an energy system that basically allows someone to have the standard of living that Americans—or Texans for that matter, because we use so much energy—have today without, you know, the consequences that we have today, or the pollution, the climate impacts, you know, that type of thing.


That's such an interesting point to bring up, you know, just about—energy access is not just about kind of the luxuries or conveniences that we think about in the West, right, like air conditioning and things like that. It really has to do with—I mean, it's at the it's the core of everything, right? It’s at the basis of almost—I can't think of one economic sector in which energy is not relevant.


I really think if we can solve the energy problem, we can solve a lot more problems. Because if we can have unlimited amounts of clean and abundant energy, we can solve our water problems. Because we have so much water available on this planet, but a lot of it's locked in the ocean, and it's salty, and we can't do anything with it. But with energy, we can clean it up.

And if we can solve our energy problems and our water problems, we can solve our food problems. Because then we can irrigate the deserts and we can grow more food. If people are not worried about where they're going to get their next meal, they're freer to go to school. And if they’re freer to go to school, they’re less likely to engage in terrorism, or other types of things. And if you free up the ability of people to see the world as bigger, because they're not focusing on an acute problem, be it energy access or water access or food access, I think it gives people a chance to think bigger, and that makes society better.


Well, that's a very beautiful place to end the conversation. I’m glad I asked you that question.

I want to leave us today with Joshua’s King of Infrastructure vision. I know it’s easy to fixate on the sheer difficulties of the work ahead. But it’s useful to remind ourselves: when we talk about the future of the power grid, we’re talking about the future of society. We’re talking about the fate of the planet, and the wellbeing of everyone on it today and in the generations to come.

Ok, that’s our show for today. Remember to check out the next two episodes in our Future of the Power Grid series, one on aging infrastructure and the other on DERs and their impact on grid stability.

Special thanks to our guests, Pat Kennedy, David Bartolo, and Joshua Rhodes. And for the rest of you, we’ll see you next time.

This is Our Industrial Life.




1As of 2017, the U.S. Department of Energy estimates that charging an electric car generally draws about 7,200 watts—or about 7.2 gigawatts for a million cars, or nearly 15 gigawatts for two million.

2EIA data showing that natural gas turbines, often used for peaker plants, generate about 50% more BTUs per KWh than do combined-cycle plants.

Background facts

The White House April 2021 fact sheet on its greenhouse gas reduction targets, including its goal for “a carbon pollution-free power sector by 2035.”

The G7 May 2022 communique committing to “a goal of achieving predominantly decarbonised electricity sectors by 2035.”

The U.N. forecast of population growth of nearly 2 billion people in the next 30 years.

The U.S. Energy Information Administration findings that global electricity consumption has been growing faster than the rate of world population.

Reuters reporting on California power outages, including the July 2006 and September 2017 events that took the grid over 50 GW.

The State of Texas Comptroller’s report that, in 2021, fossil fuels comprised 61% of the state’s energy use, with wind and solar at 28%, and other sources, including nuclear, making up the rest.

More from our guests

Discover AVEVA™ PI System™, the advanced industrial software our guest, Pat Kennedy, created.

Learn more about how similar advanced industrial software is transforming power generation, as well as transmission and renewables.

Read more about how our guest David Bartolo is using advanced industrial software to manage the grid at AGL.

Visit our guest, and King of Infrastructure, Joshua Rhodes, at the Webber Energy Group at UT Austin.



Contact AVEVA
Live Chat
Schedule Demo