WHY CAN'T WE BE FRIENDS?

How Renewables Can Be Integrated Into the Grid and Continue to Provide Safe and Reliable Service

As promised, I am devoting this post to an explanation of how renewables can be integrated into the grid. This is a complicated subject, and I could write a book about it. In fact, about half of the book I am in the process of writing will address the subject. I am not putting half of my book into this post, however, but instead will provide just a few relatively short thoughts.

Let me start with a discussion of the arguments I have seen about why renewables can never be integrated into the grid. They are all similar and point to experiences where a rapid buildup of renewables is pushed in a region at the same time that a concerted effort is made to shut down fossil and nuclear generation. The result is the premature retirement of large baseload units, resulting in decreased reliability and an increase in the cost of electricity. Variations on this argument are repeatedly made not only by people who support fossil fuels or otherwise don’t like renewables but also by some advocates of nuclear power who see renewables as nuclear power’s enemy.

These types of arguments seem somewhat exaggerated to me, but whether they are or not does not matter. The arguments all suffer from the same fundamental flaw. They assume that, because there have been problems in integrating renewable resources into the grid in the past, integration can never happen in the future. The history of technology, however, has repeatedly demonstrated that past failures do not necessarily mean success in the future is impossible.

This seems especially true for renewables and energy storage. These technologies are still in their relative infancy, especially energy storage. It would be surprising if they cannot be improved in the future. Advances in renewable and energy storage technologies will not give renewables a cost advantage over fossil generation any time soon, as I explained in my post What Does the Explosive Growth in Computing Power Tell Us About the Pace of a Clean Energy Transition? But these advances nevertheless can dramatically reduce the cost of providing reliable service with renewables paired with energy storage.

The real problem illustrated by past experience is not that integration of renewables into the grid is impossible, but that it cannot be forced too quickly. You could not rip a gasoline engine out of your car, slap in an electric engine, and expect the car to start. Similarly, we cannot simply build renewables, turn off fossil generation, and expect the lights to stay on. Forcing a transition to renewables before the grid is ready is a recipe for disaster. And today, the grid is not ready.

This does not mean we cannot start integrating renewables into the grid today. But currently, the primary value provided by renewable generation is as a good source of cheap energy. Once they are constructed, renewables have no fuel costs and, when the sun is shining and the wind is blowing, can produce electricity more cheaply than fossil units. But, because wind and solar do not provide much capacity value, we still need nuclear, hydro, and fossil generators to produce electricity at those times when the sunlight is weak or nonexistent and the wind is calm.

Today, most Regional Transmission Organizations (RTOs) allow renewables to sell unlimited amounts of electricity into their energy markets. But they limit the amount of capacity they purchase from renewables through their capacity accreditation programs, something I wrote about in more detail in my post Should Solar and Wind Be Paid for Capacity? Through these capacity accreditation programs, RTOs determine the total amount of solar and wind capacity they can acquire without imposing a risk to the reliability of their systems. Currently, these analyses show that renewables make up only a relatively small percentage of the total amount of capacity RTOs purchase. RTOs obtain the rest of their capacity from more conventional generation resources, thereby ensuring that the capacity they acquire will be available to provide the electricity needed to serve their consumer loads. This approach of buying energy, but only limited capacity, from renewables is pretty much all that can be done to integrate renewables into the grid as it currently is configured.

The RTO capacity accreditation programs, however, are not a complete answer to the question of how renewables should be integrated into the grid today. In the past, large baseload nuclear and fossil generators were the low-cost producers of electricity, and these generators depended on the energy markets for a large part of their revenues. But today, when the variable cost incurred by renewables to generate electricity is almost zero, renewables can offer their electricity into the energy markets at a very low price well below the fuel costs of baseload units, as I explain in my post A Race to the Bottom. In that same post, I explain how some renewables can offer into the energy markets at negative prices and still make a profit because of the subsidies they receive. The low prices renewables offer into RTO energy markets have reduced energy prices in those markets, with a corresponding reduction in the energy revenues once earned by large baseload units.

In the absence of significant energy revenues, the sale of capacity represents the primary alternative source of revenues for baseload units. It therefore is necessary for RTOs to employ market mechanisms that provide for capacity revenues that not only sustain baseload units in the present. RTO capacity markets must also encourage the necessary additional capital investment to ensure that baseload units can remain in service until renewables are able to provide the capacity value necessary to replace them. I discuss this in more detail in my post What’s Going on in the PJM Capacity Market? Otherwise, construction of large numbers of renewables too quickly will threaten the reliability of the grid by reducing market revenues and forcing the premature retirement of baseload units. Inadequate capacity revenues can drive baseload units to retire even if most renewables are not being relied on to provide capacity. This is already starting to happen, as described in my post What is Happening to Coal-Fired Generation?

Now let’s turn from the present to the future. Increased integration of renewables in the future will depend on the necessary energy storage technologies being developed and placed in service. Without adequate energy storage there is a point at which the addition of more renewables provides no further energy or capacity benefit, and their deployment can only threaten the reliability of the grid. This is already starting to happen in California, as I explain in my post How to Understand the Benefits and Detriments of Renewable Generation.

But with plentiful energy storage, baseload generators and renewables will be able to work in a symbiotic fashion to provide safe, reliable, and cost-efficient service. Let me use three versions of my cycle graph that I have used in some of my other posts to help explain what I mean. To summarize how the graphs work, the red line running from left to right on the graphs represents the total amount of electricity consumption (consumer loads) that occurs during different hours of the day. The line shows that consumption peaks in the mid afternoon and is lowest in the night hours, something that is true in all electric systems. The blocks below the red line show the type of generation being used to meet consumer loads.

This first graph shows the optimal conditions for renewable resources. It is a hot sunny, cloudless day and there is ample wind as well. The combined wind and solar generation supplies the entire consumer load during the day and strong winds at night allow the wind turbines to supply the entire reduced consumer loads at night. Baseload units are not needed to serve loads, but continue running to fill the energy storage units, supplemented by solar and wind energy generated in excess of loads.

Of course, on many days the sun may not be as strong because the temperature is not as hot and/or it is cloudy. And at the same time the wind may not be as strong. The following graph shows how the grid would be operated differently to supply loads under such conditions.

As this second graph shows, under these less than optimal conditions, some amount of the baseload units are needed to serve consumer loads, and some amount of electricity must be discharged from energy storage to serve consumer loads when they are their highest in the midafternoon. But the full capacity of the baseload units is not required to serve loads for the entire 24-hour period, and there is some excess renewable electricity during some part of the day. This excess capacity again can be used to charge energy storage.

Just as my first graph shows optimal conditions for renewables, my third graph shows how the grid could operate during suboptimal conditions that could occur on a very cold, cloudy, snowy day that essentially knocks all solar generation off line, and greatly reduces the amount of electricity generated by wind turbines.

Under these conditions, all of the baseload generation capacity is needed to serve loads. In addition, significant discharges from energy storage are needed during daytime, when consumer loads are higher, to supplement the relatively meager wind output. There is no excess generation during these suboptimal conditions, so the energy storage is not being charged.

I do not suggest that my graphs represent the future generation mix in any region. I am using them solely to illustrate how baseload generation, renewables, and energy storage can be integrated in the grid to provide safe and reliable service. Each of these technologies has its strengths and weaknesses. Baseload units can provide large amounts of electricity around the clock, but they are not very flexible operationally. Furthermore, large baseload units can be very expensive, costing in the billions of dollars, and they provide large lumpy amounts of capacity that may not be needed to serve relatively small loads. Renewables can produce electricity at almost no variable cost and can be constructed and operated in small amounts to meet small increases in loads. But renewables are dependent on the sun and the weather to produce electricity. Energy storage is very flexible operationally and can store electricity for relatively long periods regardless of the weather or the availability of fuel. Of course, however, energy storage needs to be charged by generation facilities.

By combining renewables, baseload generation and energy storage, it should be possible to rely on the strengths of each technology to mitigate each technology’s weaknesses. In this way it should be possible to lower the total cost of energy by relying on renewables to produce as much energy as possible, while also maintaining system reliability through the use of baseload units and energy storage. And because each technology mitigates the weaknesses of the others, it will be possible to employ each technology at cost efficient levels to avoid overbuilding of any of them.

Developing the appropriate mix of renewables, baseload units, and energy storage will vary from region to region. More solar could be included in the mix in regions where there is plentiful sun, like California, the Southwest, and the Southeast. More wind could be included in the Great Plains and in Texas, states with plentiful wind. Colder, less windy regions, such as New England may need more baseload and energy storage capacity. And all regions may also need, at least for some period of time, some amount of flexible natural gas combined cycle and turbine capacity to supply electricity when necessary.

Finally, I’m sure that most of you have noticed that my graphs show both fossil and nuclear baseload units. I have done this to show that the integration of renewables into the grid does not depend on the type of baseload unit involved. If, however, the goal is to eventually phase out fossil generation to create a low carbon grid, there are two ways this can be done. The first is to replace fossil baseload generators entirely with renewables and energy storage. To do that, however, would require a massive overbuilding of renewables and energy storage, as I explain in more detail in my post What Does the Explosive Growth in Computing Power Tell Us About the Pace of a Clean Energy Transition?

A second way would be to replace fossil baseload generation with nuclear power plants. In many ways, nuclear power and renewables are the perfect compliments to each other. Both generate electricity without emitting carbon or other harmful gases. Nuclear power can generate large amounts of electricity nonstop for well over a year, but it is inflexible operationally, extremely expensive to construct, and when nuclear is taken offline for refueling or other maintenance, it can take weeks or months before it can be restarted. Renewables operate only intermittently, but they can produce electricity cheaply, and they can instantly come online when the sun starts shining or the wind starts blowing. And renewable capacity is relatively inexpensive when compared to nuclear power. It makes perfect sense to pair these technologies, along with energy storage, to the greatest extent possible.

And so, I say to pro-nuclear advocates, renewables can be your friend, not your enemy.

***

I hope you enjoyed this post. I enjoy writing them and will never charge for subscriptions or ask for donations. I only ask that, if you did enjoy it, you press the “like” button below. Doing so will help me evaluate interest in the book I am writing on grid operations. Of course, if you have a reaction to, or question about, this post, please leave a comment and I will be happy to respond.