Jeff Smith: As we look to these deep decarbonization targets, and we’re going to need to build out a lot of transmission to support that, what happens when we can’t build it out fast enough? That means we’re going to start needing to connect a lot more to the distribution system. That means the distribution system providing local energy and capacity to meet these growing customer needs. And all that needs to be done while maintaining affordability. So, all of these increased expectations mean we need to reevaluate how we plan and design for these coming changes.
Teri Viswanath: That’s Jeff Smith, the director of grid operations and planning at EPRI. And he’s discussing the need to boldly re-envision and plan today, for what our future electric distribution system might look like tomorrow.
Hello, I’m Teri Viswanath, the energy economist at CoBank and your co-host of Power Plays. I’m excited about today’s conversation. The grid has been in a constant state of change for more than a century, with many moments in history feeling much like today.
There are a lot of headlines like Washington Post’s “Amid Explosive Demand, America is Running out of Power.” But consistently, creative adaptation has occurred, and the grid has evolved. The discussion you’re going to hear on this program is about what that evolution might look like. Joining me for this conversation is Tamra Reynolds, a managing director here at CoBank. Hello Tamra.
Tamra Reynolds: Hi Teri. The news these days surrounding our industry is ominous, but as a banker, working with electric cooperatives on their own efforts for delivering reliable and affordable power, I’m encouraged. We wanted to deliver that message and explore that story of “creative adaptation.”
So, we asked EPRI’s Jeff Smith along with Larry Bekkedahl, Portland General Electric’s senior vice president of advanced energy delivery, to explore the specific planning that distribution utilities are undertaking to “future proof” their systems.
First, we begin with our conversation with Jeff Smith, who alongside the U.S. Department of Energy, held a workshop earlier this year, focused on how utilities should approach planning for the future. He explains some of the critical takeaways from that gathering.
Smith: We brought together about 20 or so utilities from around North America to discuss some of these challenges and what we need to start doing to right size the distribution system. So some of the things that we specifically heard: There are hurdles everywhere, right both regulatory, financial, supply chain resources. You also heard that infrastructure costs have skyrocketed. Supply chain delays are out there. Some the utilities are trying to leverage every and any vendor they can to meet the growing needs of their customers.
We also heard we heard that it’s not just about building out new capacity. It’s mostly about replacing legacy assets. There’s a lot of really old legacy equipment out there, cables, poles, substations. Much of this has been in the in the field for more than 60 years and needs to be replaced. It needs to be modernized.
Some utilities had shared that they still have a lot of 4KV on their distribution system. Some really old 4KV. One utility specifically said that 60% of their miles of line are still 4KV. And that is nowhere near the voltage that’s necessary to support decarbonization electrification load.
We also heard utilities that talk specifically about, you know “what are the no regrets investments?” Many utilities are proactively undergrounding their system right now. One in particular said, I’m replacing a lot of single phase with underground and I would hate to have to go back in 10 or 15 years and replace that with three phase.
Viswanath: As a starting point to “no regret decisions” for the investments, it is essential to first figure out just how big those system might need to be down the line…and, this is where it gets tricky.
Tamra, it’s really hard to pin-point the exact moment that our mindset changed about electricity growth. Sometime last year, for sure. By the time NERC released their updated “Long-term Reliability Assessment” in December, I think most of the industry recognized that there was going to be a bigger future ahead. And even though NERC justifiably leans toward a conservative forecast, they had also come around to a growth mindset. In fact, doubling their demand projections.
But, 20-year forecasts are tricky. Here’s Larry Bekkedahl to explain…
Larry Bekkedahl: I’ve been at it for 40 years, it has changed a lot. Even if I go back five years ago and we were all projecting out our clean energy targets, we were 80% by 2030 and 100% by 2040. Now we knew that 2040 was pretty aspirational, but 2030? We could see our way. And that was going really well until a couple of things happened…
There was a CHIPS bill that took place, a bipartisan bill that was in support of chip manufacturing. And if you’re familiar with the Portland area, Intel’s headquarters here, 15% of all the chips manufacturing in the U.S. are here in the Portland area. So, there’s a lot of attraction, $36 billion dollars coming into the area to grow that industry. Well, along with that comes data centers and the thought of how AI is going to take us into the future.
When you consider a data center, it to me looks on off a lot like the aluminum industry of the last century. So, if we go back to the New Deal in the 1930s, putting people to work and they built hydro facilities, the dams here in the Northwest. Well, right alongside of that came World War II and the aluminum industry came. And these facilities are typically a hundred megawatts and they had five different pot lines. Each pot line is about 20 megawatts in an arc furnace. You know, that’s huge for our industry, and it built out all the transmission that we know as it is today. But also with those five pot lines, you could actually turn one down. And when it was needed in the Northwest to reduce energy for a peak day, for instance, or you could turn two of them down. But each building was about 100 megawatts. Well, that’s what a data center looks like today. Coming on our system, 100 megawatts is pretty normal for one of these buildings. It’s a concrete tip-up building with all the accessories associated with it. It takes about 18 months to have it up and fully running.
That’s staggering when you consider what we’ve been doing for the last 40 years in terms of building out buildings. We usually have plenty of time to react and put that together.
Reynolds: Larry makes an important point here. Even if we solve the load challenges at the distribution level, gigawatt data center campuses could still threaten grid reliability further upstream.
In fact, as we move beyond the distribution system to transmission and generation — maybe, where the true bottlenecks lie — it’s not just that cost allocation becomes more complex. It does, as Larry mentions, but also that there is a fundamental mismatch in the build-out time. There’s a pretty big gap between the lead time required to construct a data center versus building new generation or wiring new large-scale transmission lines.
Viswanath: That’s right. Larry talked about an 18-month timeframe to build out a data center. Above ground high voltage transmission has historically taken about a decade to develop. But going back to your reference on who will ultimately pay for the system. Larry shares his insights on this billing challenge.
Bekkedahl: It’s really complex in the sense that those that cause the issue, needing to pay for it. The distribution, it’s pretty clear: that substation, those feeders, go to that customer. Boom, it’s clear. Now I step up to the transmission. Well, if it’s a radial line up to that customer, again, it’s pretty straightforward. But all of a sudden you get into the network and the way that we operate all of us as utilities, the transmission…and it becomes, okay, what is being upgraded? How many parts and pieces? What percentage of that network then are you or causing the improvements on that you would contribute to that? So, it starts with the transmission, but then you can also go out right on up to the generation and say, okay, how much of the generation? And, if you want clean generation, how do I then carve out that for your company as well? And I think that’s where those questions become a little more difficult.
And finally, you don’t just build it for today, you try to build for the future. So that means you put a little extra into it. So, if you could take those projects and parse them out by percentage as to which group is paying for which piece.
Reynolds: Teri, in the conversation you had with Jeff, he emphasizes that there really doesn’t exist a one-size-fits-all view of future growth and there are challenges to the sort of crystal ball gazing required for figuring out the right investment today that will meet future needs ahead.
Smith: Not all jurisdictions are seeing the growth just yet, while others, they can’t keep up with the growth, right? One of the main takeaways that we heard is that we know we need to do something. But, if we upgrade early, it would appear as though we’re “gold-plating” the system. We need to be able to ensure that there is regulatory support to enable these this these changes to occur to reliably and affordably serve today’s customers as well as tomorrow’s customers.
Viswanath: I really like that expression, “a no regret investment.” Being careful that we don’t overbuild because we’ve gone through a similar cycle, the 60s, 70s, and 80s… Because of efficiency gains, because of offshoring, we built systems that we found were expensive. And then all of a sudden, the load just did not materialize.
Smith: Exactly. “Right sizing” is to consider what those future needs are so that you’re not over-investing or putting things where the loads may not materialize. You don’t want an underutilized investment out there. One thing that we heard are the utilities that were specifically seeing growth in their areas is that they’re less concerned about building it and the load not coming.
They’re more concerned about not being able to build it fast enough to keep up with the load, but it varies. Every jurisdiction is a little different. Some areas, they’re just not seeing that growth or it’s coming in specific pockets and they’re working with their customers to try to understand where there that load’s coming, where like maybe fleets are electrifying, that sort of thing, where they do need to provide that that capacity, that new service in an expedited fashion. But those are very specific, those are specific locations. You can’t just build everywhere. You have to build where there where there’s need for the for those services to be provided.
Reynolds: Teri, there has long been a concern in the development community of stranding generation assets, or frankly any utility assets for that matter, but in this case that’s really developing a power plant that either could not connect with downstream consumers because of the transmission gridlock we mentioned or that maybe proved to be sized too large because the growth didn’t materialize the way it was predicted with the long-range forecasts. Larry weighed in here, shedding light on the changing marketplace, where consumers are now having to be more proactive in securing supply. Playing a more active role, if you will, in planning for their future needs.
Bekkedahl: When was the last time you saw a stranded asset in our system? We are using everything and it’s being called upon in such a great demand right now. I would say that data centers are searching for wherever they can fit into the system. And many times it works really well in rural communities. And we’re seeing that, at least here in Oregon, Umatilla comes to mind or Prineville area, you know, where data centers have built.
And it’s been a great benefit to those communities from a tax base, not a lot of jobs. So I guess when it comes to schools, you’re not building new schools. So they end up with revenue that they can build other things, parks, whatever it might be. However, it is then the real challenge. Are you a vertically integrated utility? Are you dependent upon someone else to provide the transmission? Are you dependent upon somebody else to provide the energy? And so when you consider various states, you think about how who’s responsible for what element. And there has to be coordination amongst those parties to make it work.
You know can aren’t going to serve that data center if there’s no generation out there. You’re not going to serve them if there’s no transmission. That cooperation, we find, is becoming really crucial. And what we’re also finding from the data centers is that they are coming and asking, where’s the best place to build? And there are places on the system where that does make sense. And we’d like to focus them. Nothing worse than developer buys the land and then comes and says, here, we’re going to build here. Well, that may be the worst place in the world. It’s going to be the most expensive for them. And it’s not going to you know move the ball very quickly. So we always appreciate it when they come and talk to us and we think about where is the best place to put them.
Reynolds: Both of our guests discussed the dual challenge of meeting load growth with the evolving generation mix.
Viswanath: Specifically, as Larry points out, “We’ve got to build out the new grid. And, oh, by the way, we’re shutting down our carbon-emitting resources. We’re adding renewables, which are flexible….But, I can’t always tell when they’re going to come on or off. So how do I do all that? How do I manage the system with all these moving variables, and how do I do it quickly?”
What’s more, the power flows on the grid are changing, as he illustrates about his own system, but so too is the potential leverage that large consumers now bring to the table. The idea of driving more efficiency and load flexibility. Of data centers that are spread across the nation and maybe are able to process at different times coast-to-coast.
Bekkedahl: We’re kind of two sides. One, obviously the flows, where that energy is coming from. In the West right now, we’re seeing flows from California where there’s a lot of solar, a lot of hydro versus Canada where there wasn’t good snowpack. And so we’re seeing South to North flows where in the past we were always North to South flows. So you got to start thinking about that in your planning models. But I also say with these new larger loads, I like to term it bridging strategies. So what’s their backup generation? And is that something that can be utilized during those peak periods to offset the peak?
And there are lots of opportunities, I think, in that space to help. That enables them to bring their loads on sooner, at the same time managing the better utilization across the whole system, which also then drives price down for everyone. Instead of the spiky-ness of some of the price considerations. So we’re trying to better utilize the entire system. And that flexibility, any time that they can either flex their load, which means maybe I’m going to run the large learning model in Georgia instead of Oregon today, that’s a different mindset we haven’t really thought about in terms of energy usage from a data center. But if we do that, we can spread out and do a better job of utilizing you know the energy that we have across the nation.
Those are those are different thoughts in terms of bridging, using the backup generation, obviously getting as efficient as possible, whether it’s their cooling or their operations. We’re going to see benefits that way. It’s kind of like the old homes. When we first built them, we didn’t really think about energy efficiency. But now when you think of a home, you try to get as much efficiency out of as possible. Try to do the same thing with data centers as well.
Reynolds: Larry stresses the importance of coordination from the consumer all the way upstream to the power plant, but here too there are foundational differences in our planning horizons as Jeff explains.
Smith: So when we think about right sizing, it means being proactive and taking a mid to long-term look to identify the grid service needs, your new designs, your new operating requirements that enable that deep decarbonization of the energy sector in a reliable and cost-effective manner.
This means looking beyond the typical two to five year planning horizon and looking much further out, say 20 to 30 years, and determining what we need of the distribution system in that future scenario, and then backing that back to say, okay, what can we be changing now that helps us to meet that future scenario? So we could continue down our sort of a business as usual approach where we upgrade the system, by looking out at just two to three to four to five years, where there’s near-term need and looking out over those next few years and adjusting to meet those systems’ needs. We could do that, but by doing just that, does that potentially set us down a non-strategic path that requires us to replace infrastructure within the next 10 to 15 years? The last thing we want to do is to continue to build on our system with a short-term look and then 15 years from now regret that investment and having to replace equipment.
When we think about integrated resource plans, you know, you’re typically looking much further out in time, right? You’re looking 20 to 30 years. But with distribution, we’re specifically looking typically in that two to three years out timeframe, perhaps maybe out to five. There are obviously reasons obvious reasons for those differences. It takes more time to build out bulk generation/bulk transmission. Distribution, however, can pivot a lot more quickly.
The further out we go, we know that uncertainty increases as go further out in time. And, when we’re particularly looking at distribution, we know that location is king. So, when you consider that a long-range forecast has a lot of uncertainty as to when and where changes are going to occur on the system. So that’s sort of sort of a broader challenge as it relates to reconciling the planning processes between sort of your traditional IRP type planning to your distribution plans.
But, as it relates to customers and the choices that they have, resources like solar or storage where that is dispatchable for various use cases, EV loads that that may be dispatchable or are controllable, that sort of thing, these are technologies that can consume resources from the system, but they could also contribute resources to the system, right? In some cases, these sort resources may be dispatchable, providing local distribution services, or providing bulk system services, but through the distribution system. And that within itself creates challenges for planners. In essence, they’re trying from a planning perspective, trying to figure out what that what that new resource is going to look from a long-range perspective.
Viswanath: From a resource planning perspective, what are the ways that the industry is adapting and preparing for the growth ahead?
Smith: Many utilities are going through various stages of grid modernization. What we’re seeing across the board is an investment in foundational elements that will enable the distribution system to be to provide increased reliable service to all customers. Deploying more distribution automation, deploying more sensing capabilities to monitor the distribution system more effectively so that we can quite honestly push the system a little harder to get more capacity, delivery capability out of the distribution system.
We’re seeing investment in management systems that will enable the system to be optimized, to be include resources like solar, storage, EV into the mix. Incorporating DER aggregation and along with other resources like microgrids. So there’s a lot of work that is happening that is foundational.
All those things are occurring, but there are certainly a lot of challenges around that. Because as we want to advance and need to advance the system, affordability is a critical aspect of that. We’ve got to make sure things remain affordable. While there’s opportunities and some things are really excited about that’s happening on the distribution system from a modernization perspective, there’s certainly like the challenges around making sure everything remains affordable. There’s certainly movement that’s happening across the industry as it relates to enabling both of those to occur.
Reynolds: The bottom-line question that we posed to Larry is, why is he so optimistic that we as an industry are prepared to meet the challenges of the day?
That is… if we agree that meeting these new large loads is important for achieving the next cycle of economic growth, and that ongoing decarbonization of the grid will continue, how do we break the gridlock that appears to be occurring?
Bekkedahl: Well, I would say, number one, that technology is coming with us and we’re being able to put it to advantage such that when solar is being produced and how it’s stored, when is it used, you know, the old duck curve being limited. But, at the same time, what batteries are doing for us on our system? Large scale, I know we’re putting on close to 500 megawatts of batteries onto our system, just how do we operate how do we think about that differently? Many of us are heading towards a digital twin of our system so you have that visibility.
We’re also moving to grid-edge operations where we don’t bring all the data back to a control center. You leave it out there and you have it operate autonomously in a neighborhood. Those are huge advances that you know we will all be making. And it’s exciting to think you know that technology is now keeping up with the electricity, so to speak. And so our digital and our electrons are starting to look the same.
And I’m very, very excited about, you know, where we’re heading, how customers can participate with us. Now, many of them don’t want to have to think about it and play with their phone and some app. They want it to happen. And so how do we build that in? How do we make that simple and that yet they know they’re doing the best job they can to participate and be a part of a green energy future?
Viswanath: Right now, there is a terrific opportunity for the utility industry to be intentional in how future demand is met, setting an important precedent for handling growth in a way that better supports the grid.
There’s a unique opportunity to invest in the broad array of solutions outlined by our guests — starting with consumer efficiency and demand management and working upstream to provide ample, reliable and secure power in an age of rising electricity demand.
Reynolds: As our electric cooperatives head into their budgeting and planning season, I hope that this conversation has been helpful. Join us next month as we discuss an updated research paper from REMDC with some co-op leaders I had the privilege of working with.
Viswanath: Goodbye for now!
