Duck Curve Now Canyon Cliffs

More than a decade ago, the infamous duck curve became the talk of the global energy community. The duck curve — first introduced by the California ISO and named after its resemblance to the aquatic bird — displayed electricity demand net of variable renewables, such as solar, throughout the day.

The duck curve emphasized the challenges associated with a growing solar resource that resulted in net demand declining during the day (the duck's belly) when solar resources were generating and sharply increasing at sunset (the duck's neck), with both the belly getting deeper and the neck getting steeper and longer every year as more solar resources were deployed.

Since that time, the energy industry has been working to improve how the grid operates with the duck curve, since the belly can cause over-generation issues, renewable curtailment, and heavy reliance on inverter-based assets, and the extending neck can make it more difficult for the resource fleet to be flexible enough to ramp up. Increased system flexibility is paramount to address these challenges.

Public Utilities Fortnightly's Steve Mitnick talked with EPRI CEO Arshad Mansoor about what the implications are of all this for the energy and utilities industry. Also, the sense of urgency surrounding the issues.
 

PUF's Steve Mitnick: What's the headline?

Figure 1 - Today’s Duck Now Looks More Like a Canyon

Arshad Mansoor: The way we have operated the power system reliably for a century will need to change quickly.

We always had dispatchable generation to meet forecasted load. We forecast load and we dispatch generation to meet it. That's worked for the last one hundred years.

Now we forecast wind and solar. And we dispatch generation to meet load. And increasingly, we are dispatching load.

Wind and solar are not dispatchable. They have to be forecasted.

DOE Secretary Jennifer Granholm with EPRI CEO Arshad Mansoor during her visit to the EPRI Lab in Charlotte.

As we bring on more and more solar, the other generation that is needed during peak solar is minimal. Sometimes with generation exceeding demand, and we call this net demand. Which means I need to export excess generation to other regions or curtail generation.

The problem is the opposite when the sun starts setting. You can see a significant need for generation that has to be ramped up.

Texas is a solar rich area. But even there, because of electrification, it means that five to seven p.m. during winter, people come home and turn on lights, turn up the thermostat, and — with electrification — charge their electric vehicles. That's when the sun sets.

Worst case. To "climb up the evening canyon wall," Texas (ERCOT) set a record this year, needing to ramp up dispatchable generation and storage by over twelve gigawatts per hour.

DOE Secretary Granholm on a visit to EPRI’s Charlotte Lab, from left, Senior Project Manager Transmission & Substations Rachel Moore; Rep. Alma Adams, DOE Sec. Jennifer Granholm, Vice President Transmission and Distribution Andrew Phillips, EPRI CEO Arshad Mansoor, Program Manager Overhead Transmission Gary Sibilant, and Vice President of Corporate Affairs Katie Jereza.

PUF: Unbelievable.

Arshad Mansoor: It is. And that is what we are seeing today. As we look forward, the depth of the canyon in PJM is projected to be around seventy gigawatts in 2035. Which is equal to the total peak load of New England. And the average hourly ramp rate is projected to be over ten gigawatts per hour.

So now, it's how you "climb down the wall." And what do you do when you're at the bottom of the wall? 

What you need to do to climb up the wall requires different thinking on markets. Different thinking on the type of resources needed.

It cannot be just more dispatchable generation and storage. It must also be dispatchable load.

So, if PJM is saying that's what PJM will face in 2035, then places where solar is going to play an even bigger role, how you navigate this canyon curve in a reliable and affordable way becomes a thought process. We knew this change was coming. The duck curve told us.

It has happened. And in some cases, it goes way beyond the duck. When the duck was happening, we said, we'll work on flattening the duck's belly by doing dispatchable load.

Well, clearly in the last thirteen years we haven't done that enough. Because it's not looking like we have any.

The sense of urgency is another message. A sense of urgency not just for technology, which is where EPRI comes in. But a sense of urgency for innovation, markets, regulations, and policies to change market rules.

Load flexibility using innovative time-of-use rates plays an important role in this transition. We have been talking about time-of-use rates for fifty years.

I think less than ten percent of residential customers have time-of-use rates. Arizona Public Service is one of the more prominent examples where time-of-use rates have been adopted.

Recently, Missouri made its time-of-use rates the default for all its customers. These time-of-use rates can help utilities to increase load during the belly of the canyon and decrease load during the time the sun is setting.

We need to realize, this decade, all the things that we have been saying about solar, wind, changing resources, electrification, are going to happen in force. Along with changing weather.

Another key finding in PJM is when they do annual reliability studies, they always had summer months as their highest risk time. They used to tell generators not to plan scheduled outages in summer.

Now, PJM is saying, beginning from 2030 onward, that ninety percent of the risk will come from the winter months. Because the sun sets earlier from five p.m. to seven p.m. That is when people are coming home, using more electricity, and now charging their electric vehicles.

Now, you've got to change the scheduling of generation outages. This is a very different way of operating the power system than engineers designed it a hundred years ago.

The canyon is not new. The canyon is a progression of the duck. The duck is not new. The fact that we'll have electric vehicles is not new. But now, EVs are seven percent of new car sales. When two years ago, it was one-and-a-half percent.

Last year, was the first year in U.S. history where more heat pumps were sold than gas furnaces.

This decade, these changes are happening at a rapid pace.

The energy sector may need to think differently and act differently, quickly, in changing the markets. For example, Texas, after some challenging events in both winter and summer, have found some of the difficulties of relying on an energy only market, and are evaluating some supplemental market products to bolster the needed support to maintain reliability.

We shouldn't be waiting for events to happen, for the system to learn. The energy sector has learned. Markets have to change, technologies have to be deployed, enabling policies and regulation has to happen, and licensing has to be streamlined. It takes a village.

PUF: We wanted high-capacity factors in all generation assets. It's not doable. At four, five, six o'clock, in Cal ISO, the solar generation is falling off a cliff. You're going to need units with one and two percent capacity factors. Otherwise, it doesn't work.

Arshad Mansoor: If you look at the broader perspective, even if you can run it reliably, you are increasing the cost of the whole system. Running generation and storage alone is not going to be the answer. The energy sector may want to look into innovative rates and pricing. Incentivizing customers so their load becomes a resource for a grid for which they're paid.

Electric vehicles cannot happen like air conditioning happened. Because with air conditioning we just added infrastructure.

What I'm saying to industry leaders and challenging them with, is we cannot do this transition affordably if we're just investing in infrastructure. We finally have reached the decade where the load is growing.

You have opportunities to invest in infrastructure. You're going to get the grid ready for extreme weather. You've got to get ready for electric vehicles and heat pumps. You've got to figure out how dispatchable generation can meet the ramp rates of the canyon. The energy industry is also looking at ways to invest in customer resources and make customers part of the journey.

But the fourth issue, which we never thought even two years ago was going to happen, is happening right now. A thirty-year trend of flat to declining industrial load is reversing. Because industries are coming back. You cannot do all these four things just by building more infrastructure.

PUF: Also, getting away from thinking about dispatching to hours. It's going to be down to minutes, with massive amounts of data about the demand side and storage.

Arshad Mansoor: Making customer resources a shared resource is going to be critical and we are thinking of a large-scale initiative, for next year, which aims to create one hundred-gigawatts of dispatchable resources that can in real time help balance the grid. This dispatchable resource would be a customer resource.

Fifty-five million U.S households, forty-three percent of the population, are considered low to moderate income, LMI. They don't use Tesla Powerwalls. They don't have solar. But most have ICE vehicles, inefficient heat, air conditioners, and water heaters.

When electric vehicles become cost-effective, they might not be able to afford a level 2 charger. So, let's deploy smart level 2 chargers, smart water heaters, and smart heat pumps as virtual power plants. If they can afford it and can adopt an EV, it will reduce their gasoline bill. Their total energy bill will be much more affordable.

PUF: Instead of having in a control area, tens or hundreds of generators on the supply side with passive demand, you're talking about a massive amount of touch points on the grid on both sides.

Arshad Mansoor: Bring in AI and massive data management. But you still will have to invest in the grid. You still need more transmission, distribution, generation. But it cannot be only that.

PUF: Rate cases have long turnaround with the focus on big ticket items, like major generating plants. But what you're talking about is a more complex set of proposals that the utility brings forward.

Arshad Mansoor: One of the things I also say to leaders, stakeholders, utility boards, and their CEOs, is clearly with more variable generation, we would initially need two- to four-hour storage. We are already seeing utility scaled storage double in 2022 from five to ten gigawatts, roughly.

It may double this year again, which is great. But if you are in your jurisdiction doing an IRP and you say, no, I need five-hundred-megawatt, four-hour storage, don't do one big project.

Do a four hundred fifty-megawatt, four-hour storage, a big project. Take another fifty megawatts. Break it into twenty-kilowatt, thirty-kilowatt chunks. Go to your police and fire stations and schools used as shelters.

Create microgrids so that battery is still your flexible resource. But that battery is the resiliency resource for those critical customers.

You are now using the same battery, but not one big five-hundred-megawatt battery. It is more complex to execute twenty projects. You've got to get together with the school district, fire, and police stations.

PUF: That means grid intelligence basically at the distribution level.

Arshad Mansoor: We have been investing in grid edge intelligence and distribution, and we need to continue to make an integrated grid that enables shared customer resources. But that investment needs to have an equal amount of investment in smart heat pumps and smart water heaters, and smart level 2 chargers that are dispatchable, especially enabling LMI customers.

The communities where utilities serve, this will make the partnership with them even more important.

Consumers now see this clean energy transition as reducing their energy costs. Yes, it will create more risk to create thousands of projects, millions of heat pumps. It will not only require smart grid and grid integration, but it will make a level playing field for all utilities, especially investor-owned utilities.

If I have the technical choice to invest in a bigger transformer or in flexible water heaters for some customers, the incentives should allow the best outcome to happen.

PUF: That gets the utility closer to the customer.

Arshad Mansoor: For those customers with low to moderate incomes, it's hard on their pocketbooks or wallets when their water heater breaks, and they must install new ones.

Imagine if we had fifty-five million water heaters that are dispatchable. Do the math. Fifty-five million times .5 kW is over twenty-five gigawatts of potential load to manage, if they are all operating at the same time.

If we can work with these customers to overcome the upfront cost problem by monetizing rebates, tax credits, and load management value, they can participate directly in the clean energy journey, and save money from day one.

Imagine if we do the same for high efficiency, grid-connected heat pumps. Imagine if we also do this for smart chargers when the costs of EVs come down and they can afford an EV, thereby reducing their gasoline costs and overall energy costs.

So that last December when power demand peaked in the southeast and North Carolina had blackouts, these millions of customers' smart resources could have made the grid more resilient and possibly avoided those blackouts. This is improving the resilience of the grid and that helps all customers that the grid serves.

It is also helping to reduce the cost of deploying these technologies. Done right, a secure, clean energy transition should reduce the cost of energy (not just electricity) for all customers, especially those who can least afford it.

PUF: The thinking must change. It's a utility asset if it's in a park with a fence around it and it has the company name. But not if it's at the customer's property behind the meter. Or right in front of the meter.

Arshad Mansoor: The customer allows the management of that asset during certain times of the year for the benefit of the grid. In return, the customer gets a significant discount and, if LMI, help with acquiring the equipment in the first place.

This is not like AMI. Because some customers will say, I don't want anybody to manage my water heater. That's ok and they don't need to participate. But getting a new hot water heater with low or no upfront payment can be a game changer.

The utility will have to work with a lot of stakeholders: states, smart-distribution grid edge providers, regulators, policymakers, and so on. It will take a village.

By doing that, you're not only creating a more flexible grid, but you're reducing the winter bill of that customer who may be using steam heating, and now they may be using a smart, grid-connected heat pump that is part of this clean energy transition.

Also, what I'm saying is that this clean energy transition will reduce the average energy cost for low- to moderate-income customers. Done right, means you cannot just do it all the way we handled air conditioning in the 1970s. We need to apply grid modernization and data analytics that we know how to do.

PUF: This is a different way to manage wholesale markets. And the distribution system. A different way to interact with customers. And a different way internally at companies to structure culture.

Arshad Mansoor: It is a big shift. And it is no longer talk only. Because if we don't do that shift, the four changes that I see the industry facing this decade are going to go forward with or without the industry.

One, electric vehicles are going to happen, whether you like it or not.

Two, unfortunately, extreme weather is going to become more frequent, especially heat, precipitation, floods, and wildfires.

Three, more and more wind and solar are going to come online.

Four, finally, utilities are getting the industrial customer back. Utilities will be asked to provide power to a thousand-megawatt semiconductor or battery plant because the CHIPS, Bipartisan Infrastructure, and Inflation Reduction Acts are creating a lot of incentives for domestic production.

You have industrial load growth, renewables, and heat pumps coming. And the climate is becoming more extreme. All four are going to happen. We cannot address all those four in the way we have done it in the past. We have to think and act differently. This means all parties, the regulators, utilities, market designers, et cetera.

Twenty years ago, when we said solar and wind are happening, you've got to get ready. The response was, I'll do it when I see it.

You're seeing it now. You're seeing electric vehicles, heat pumps, industrial load growth, extreme weather, and wind and solar.

PUF: With all these changes in culture, data, interaction with customers, and regulatory, how is EPRI going to bring this about?

Arshad Mansoor: We will be advancing technologies that could be scaled up and deployed. Whether it is a virtual power plant site at the customer site.

Or in the future, it's long duration energy storage that is providing flexibility. Or a small modular reactor. So, one EPRI role is advancing technology that could be reliably deployed and de-risked by utilities.

Another big focus by EPRI will be to "speak the facts about power to people in power." We do not influence policy. We cannot. But informing policymakers, regulators, the Department of Energy, government? We've got to do more of that.

Our foundation, which is more than seventy-five percent of our research, is existing nuclear plants, transmission lines, generation, and distribution. We've got to make sure to continue to advance technologies like AI and asset management. So, assets can be operated reliably, safely, and affordably for a long time.

That includes, for example, nuclear going from forty-year licensing to sixty. Then sixty to eighty, which requires a lot of technical work. Our foundation is the engineering foundation for existing assets.

On top of that, we are looking into how to de-risk emerging technologies that can be deployed widely. Both on the customer side as virtual power plants and on the grid side.