Lessons From Texas for the Grid of the Future
What will it take to create a truly reliable, net-zero-carbon energy supply?
The story of the tragic 2021 blackouts in Texas isn’t about wind power versus natural gas. It’s not about ‘greedy utilities’ versus the people. It’s not even about the wonky details of how electricity markets are designed by the Texas power grid operator (the Electric Reliability Coordinating Council of Texas, or ERCOT).
The real story is about the value of electric reliability, and the difficulty of securing it in a changing energy system. Texas has shone a spotlight on the acute challenge of ensuring reliabity through a major energy transition, but not in the simplistic way that certain media outlets and politicians have implied. The solution is not to turn away from wind & solar energy; instead, the solution is to embrace the unique features of these energy sources, along with complementary technology like battery storage, to create an even more reliable grid alongside a cleaner one.
Reliability is getting more complicated, and simultaneously becoming even more critical. We need to rethink how we plan our energy system.
Two lessons from Texas for our net-zero-carbon future
I. Electric reliability is undervalued. In economics parlance, energy is one of the highest “consumer surplus” goods money can buy: its value exceeds its cost by leaps and bounds. The longer you go without electricity, the more you realize just how much extra you’d be willing to pay for a few watts of it. Moreover, as we’ve seen in Texas, there are externalities associated with large, regional power outages that exceed the cost to any individual consumer — for example, threats to public health, safety, and supply chains — which are difficult for a competitive market to fully ‘price in’.
The US electricity sector already scores incredibly well on reliability, all things considered. On average across the US, the lights turn on when you flip the switch more than 997 times out of 1,000. And yet, it’s not enough. Today, electricity delivers about 20% of the energy we consume in the United States (as in other industrialized countries). By 2050, most credible net-zero-carbon scenarios peg electricity’s share of the ‘final energy’ pie at 50% or higher.
We know how debilitating & dangerous blackouts can be for a 20% electric society; the recent tragedy in Texas is evidence enough. Sustained outages will become even more intolerable in a 50%+ electric society. Hence, the industry needs to do even better, and we need to do so in the face of two forces that are making the job harder…
II. Reliability is also getting more complicated.
The first force is climate change, which is already causing more extreme weather ‘events’ (including, most likely, the exceptionally frigid February temperatures we just saw in Texas). We’re already beginning to see the impact of more extreme weather play out in reliability data from US utilities.
This chart illustrates why 99.7% electric reliability isn’t good enough. Electric utilities have done a laudable job keeping the lights on during normal conditions. It’s abnormal conditions that are increasingly the driver of power outages. These are, for intuitive reasons, the most dangerous types of outages; they’re longer and more regionally concentrated. See: Texas, February 2021.
The second force making reliability more complicated is the rise of wind & solar power. Today, there is plenty of firm, on-demand power generation on the grid to manage the intermittency of renewables. Contrary to messaging from various politicians, the recent events in Texas had nothing to do with the natural variability of wind & solar energy resources. In fact, the Texas outages had much more to do with the failure of natural gas infrastructure, which Texans were counting on as a “firm” source of energy.
Yet, it’s important to note that neither wind power nor natural gas infrastructure in Texas was sufficiently prepared for extreme weather. A grid fully reliant on either resource, absent investment to prepare for such freezing temperatures, would have failed.
Many studies suggest we ought to be able to achieve very high levels of renewable energy on the grid—upwards of 80–90% of total annual power generation . During ‘normal’ conditions, we can rely on a combination of energy storage and gas-fired power plants to fill in the gaps left by the variability of wind & sun. However, there’s a few reasons to believe our power system planners are going to need to sharpen their pencils. As the situation in Texas highlights, we can’t just prepare for ‘normal’ conditions.
Obviously, we’ll need to install cold-weather gear on wind turbines and gas pipelines — but that’s just a start. Studies of high renewable energy penetration generally employ some approximation of ‘typical meteorological year’ weather data to model the hourly power generation of wind turbines & solar panels. These profiles don’t adequately account for extreme weather events — i.e. unusually extended periods of very low wind or sunshine. As we begin to rely more & more on weather-driven power generation, we need to make sure we’re prepared for really strange weather.
Second, most high-penetration renewable energy plans lean heavily on energy storage technology to balance out daily variations in wind & solar energy. But storage facilities — like big Lithium-ion battery plants, for example—are much more difficult to plan & operate effectively than conventional power plants.
Today, many electricity providers are already aiming to satisfy a portion of their peak demand needs with storage. Market operators & regulators are allowing them to do so with battery systems that have just four hours of ‘duration’ — meaning they can only provide their full power output for four hours without stopping to recharge. Four-hours has become a kind of ‘rule of thumb’ in the industry for the amount of storage one needs in order to serve as a form of ‘firm’ peaking capacity.
If we’re going to depend on storage for electric grid reliability, rules of thumb obviously aren’t going to cut it. We’ll need to know exactly what duration of energy storage the grid will need to make it through periods of low wind & sun. That will depend on a host of tough-to-pin-down variables, including: the precise shape of regional wind & solar generation curves (especially the very low-probability, but increasingly high-consequence tails); the extent to which electricity demand can be made more flexible; and the precision with which storage system operators can predict the optimal time to dispatch their storage plants.
We’ll need to add storage in longer-and-longer duration tranches, eventually using storage to balance all of the intra-day variability in renewable supply — and potentially even seasonal variability (with a much, much cheaper, multi-day storage technology):
In short: planning for a high renenewable energy and storage based system — with even better reliability than our current system — is going to require some new strategies.
Two complementary strategies
There’s no shortage of good ideas for eking out marginal gains in grid reliability while reducing the cost of the system to boot. Leading, next-generation solutions range from drone-based power line inspections to vegetation management employing satellite data. These types of solutions can make a big difference in the near-term, but we’ll also require a long-term game plan — a plan that focuses on two polar opposite, yet highly complementary strategies. One I’ll call “Big Transmission”; the other, “Distributed Resilience”.
Big Transmission
It is what it sounds like: a much bigger, more interconnected transmission system that essentially makes the entire country —perhaps the entire continent —behave like a single, tightly integrated grid. Big Transmission is most manifestly needed to access remote wind & solar resources many miles away from big cities; however, it also has a subtler role to play in blending the generation profiles of renewable resources across regions. As we hook up increasingly disparate wind & solar resources, we also ought to be weaving together regional transmission systems that have historically been only weakly interconnected.
Take Texas as an example. Texas is famous [infamous?] for refusing to interconnect the ERCOT grid outside of the state. But, take a look at the seasonal generation profiles for wind & solar resources along the Texas coast, where most of the energy demand is centered:
Relying solely on these renewable resources, Texas would pretty quickly encounter an overgeneration problem in the spring and summer months, and shortfalls in the winter. There are emerging solutions to this seasonality problem, ranging from ultra-long-duration storage to natural gas with carbon capture & sequestration. But, transmission is an even more tried & true answer.
Texas has even more abundant wind resources in the sparesly-populated Western half of the state than it does on the coast (especially the panhandle). Perhaps more importantly, the seasonal generation profile of panhandle wind is rather inverseley correlated with that of coastal wind:
If Texans were willing to set aside their aversion to inter-regional cooperation, they could do even better. For example, here’s a seasonal Iowa wind resource profile layered on top of Texas wind:
A Big Continental Transmission strategy wouldn’t solve all of our reliability challenges as we transition to net-zero-carbon. But studies show that it would be one of the lowest-cost mechanisms for balancing the grid in normal conditions, as well as guarding against extreme weather in any given region.
The challenge for a Big Transmission strategy in the United States is that it’s proven to be incredibly difficult to plan, site, permit, and build new long-distance, high-voltage transmission lines in the United States. Inter-regional transmission lines are especially tough, because they require coordination and approval from multiple regulatory bodies in every state they touch. So far, even highly promising transmission projects that have lined up willing renewable energy sellers on one end, and willing buyers on the other, have been stymied by objections from state gatekeepers in the middle.
Clearly, we need a stronger national transmission policy. But fortunately, technology can also help, even without a complete policy & planning overhaul. For example, the Brattle Group just released a study demonstrating how three specific “grid enhancing technologies” — advanced power flow control, dynamic line ratings, and topology optimization — can collectively double renewable energy integration capacity with no additional ‘poles & wires’ investments:
In addition to these ‘soft’ technology solutions, we’re also probably going to need to upgrade the wires themselves. The easiest way to add more transmission capacity is to use the corridors we already have, which are already sited and permitted. Each year for the next two decades, 4–6 thousand miles of transmission lines in the US will be turning 50–80 years old; those aged lines are ripe for replacement with conductors capable of carrying more energy — without expanding the size of the clearing or the towers (which would trigger a new environmental review). Eventually there ought to be a big market for new lighter-weight, higher-conductivity cables, or even superconducting cables.
Distributed Resilience
There’s no way we can cost-effectively prepare the macro-grid for every hypothetical threat, which range from extreme cold snaps, to weeks of low wind & sun, to cyberattacks.
What we can do is make our system more resilient so that we can make it through unforseen challenges. And the best place to do that is out at the edge of the grid, close to where people actually consume energy. In short: microgrids.
“Microgrids” are portions of the grid — or even simply individual buildings — which can ‘island’ themselves from the macro-grid for periods of time with their own power supply. Historically, this was done primarily with backup diesel generators, which are too inefficient and dirty to run except during emergency circumstances. The nice thing about renewables and storage is they can contribute to the macro-grid during normal conditions as well.
Today, these renewable microgrids work best when complemented by natural gas gensets, in order to provide as close to 100% reliability as money can buy. One of my firm’s portfolio companies, Enchanted Rock, is a specialist in low-cost, low-emissions natural gas systems, which performed valiantly during the recent crisis in Texas. Distributed natural gas systems are not, of course, zero-carbon today. But eventually they can be transitioned to run on zero-carbon fuels like hydrogen or RNG. Because of their critical contribution to reliability, they could very well be the last form of carbon-emitting generation pushed out of the system.
Microgrids that can provide both regular service to the energy system as well as backup power supply to homes & businesses require a unique blend of technology and business model innovation. Entrepreneurs are finally beginning to figure out the right recipes — taking slightly different approaches at the commercial building level, for individual homeowners, and for entire neighborhoods. Meanwhile, advanced system modeling & control software is making the dream of a ‘grid of microgrids’ feasible. (Disclosure: My firm is also an investor in Opus One Solutions, which makes some of that advanced software.)
Today, about 25% of households pay for some sort of security service focused on crime, but only about 5% of buildings have backup power. This is clearly a market with room to grow — and one that will need to grow as the grid faces greater threats of extended outages, and we become ever more dependent on electricity.
Getting to 100% clean…AND 100% reliable
We have the tools to begin preparing for both. With a combination of investment in Big Transmission and Distributed Resilience, our energy sector can deliver on the promise of zero-carbon energy, with even better reliability than we enjoy today, and without breaking the bank.
