The past few years have made one thing clear: The Texas power grid needs to get stronger and smarter if it’s going to keep up with the state’s needs.

The Texas grid presently is capable of producing and delivering about 9 gigawatts of electricity. Most days, people use that power in their homes and businesses without a second thought.

Yet the simplicity of flipping a switch or throwing a breaker masks an immensely complex system in which inputs must match outputs, prices matter nearly as much as physics, and the harmony of power generation and power transmission is in constant flux.

Existing technology solutions already are helping to fortify the grid. Batteries, microgrids and innovative artificial intelligence software all assist in making Texans’ electricity generation and delivery cheaper, more efficient and more resilient.

But those innovations, while promising, may ultimately amount to half measures that can’t offset the deeper economic and political issues that make the grid a source of profound worry for Texans.

Ed Hirs, the inaugural University of Houston Energy Fellow and a lecturer in the Department of Economics in the College of Liberal Arts and Social Sciences, has argued for years that Texas’ biggest hurdle to a reliable grid is its economic model. The state government has created a marketplace that fails to encourage new producers to enter or to reward existing producers for maintenance and upkeep.

During eight of the 10 years prior to 2021, Hirs says, the average price paid for electricity in the Texas market was less than what a utility would need to make to invest in a new plant. Without offering an expected rate of return, Texas is not seeing power plants come online fast enough to meet future needs.

“It’s a market design issue much more than it’s a technical issue,” Hirs says. “If you don’t allow anybody to make a rate of return, they’re not going to be there. ... Wall Street won’t let you build power plants just to lose money.”

Hirs is passionate about the issue, in part because two immutable factors are putting increasing pressure on the grid.

The first is simply rapid growth in demand. Texas has added almost 1 million more residents (and their economic activities) since 2021, and the state continues to grow. To keep up, the Electric Reliability Council of Texas, or ERCOT, may need to double its supply capacity as soon as 2030.

The second factor is the increasing frequency and danger of weather disasters. Statewide heat domes and polar vortexes strain the infrastructure across Texas. The average coal power plant is now 55 years old, while natural gas power plants have been around more than 30 years on average. Their rate of breakdowns is also the highest in the nation.

If a statewide winter storm were to knock out both wind and solar farms, there would not be enough coal, natural gas and nuclear power plants to meet winter demand. The National Centers for Environmental Information at the National Oceanic and Atmospheric Administration counted about 1.4 such disasters per year during the 1980s. That number has increased about tenfold during the 2020s.

A plan for a more resilient grid must account for such extreme events. When those hurricanes and winter storms and wildfires strike — simultaneously knocking out power transmission and raising the demand for energy — the grid risks breaking.

During the winter storm of February 2021, the grid was less than five minutes from a complete and catastrophic collapse before ERCOT shut down parts of it. Hundreds of Texans died of hypothermia, energy consumers were hit with billions in excess charges, and the state suffered more than $100 billion in losses.

Hirs points out that Texas has thus far avoided implementing the simplest fix: a demand response program, whereby ERCOT would offer all of its consumers the opportunity to opt out of peak demand periods in exchange for payment.

“Texans have had smart meters in their homes and businesses for more than 10 years. The technology is proven. During the summer of 2022, California sent out texts requesting that consumers opt out on peak afternoons. More that 1 million did — and without getting anything in exchange.”

In the meantime, the prospect of a full collapse has worried Hirs for years. Imagine, he says, “26 million people without fuel, without clean water, without sewage, without anything that electricity provides. How many Berlin Airlifts would it take to keep Texas alive? We’ve got natural gas, but you need electricity to make sure there’s integrity to the natural gas network itself. It’s the Achilles’ heel of this nation’s national security.”

To stave off that scenario, and to buy time for Texas to make structural changes, these new technologies may become indispensable.

Batteries

They’re not yet big enough, cheap enough or reliable enough to fully make up for the shortcomings of the Texas grid. But batteries are a significant and fast-growing component of the state’s power supply. In May, Texas saw a single-day record of 3.2 gigawatts of batteries simultaneously discharging to the grid, accounting for about 5% of the state’s electricity load.

That capacity is only projected to increase as the state brings ever more solar and wind capacity online. Fluctuations in sunlight and wind make batteries an essential complement to renewables in particular.

“Texas is uniquely equipped for rapid growth for utility-scale battery storage,” says Yan Yao, Hugh Roy and Lillie Cranz Cullen Distinguished Professor in the Department of Electrical and Computer Engineering at the Cullen College of Engineering and principal investigator of the Texas Center for Superconductivity at UH. Yao and his research team are globally known for their work to create next-generation batteries.

“This shift in energy storage would coincide with volatile weather conditions and rising demands on the energy infrastructure. So that is definitely a positive and fast-moving field.”

Protecting people from volatile weather is a key driver of battery adoption. A Harris County initiative is equipping community buildings, such as senior centers and libraries, with solar panels and batteries to ensure these gathering spaces can be used reliably during emergencies, including heat waves, when people need cooling centers.

But batteries still face a few daunting obstacles on the way to wider adoption.

Most of the current battery technology used in home or building construction is similar to what electric vehicles use: lithium-ion batteries capable of only a couple of hours’ worth of storage. While their prices have come down steeply in the past decade, Yao says those batteries need to get significantly cheaper for people to adopt them more broadly.

Lithium-ion batteries also pose a notable fire risk — no small consideration when weighing the best options for a home. And, vexingly, they offer the best performance at moderate, pleasant temperatures and falter somewhat in very cold and very hot conditions — exactly the weather that makes outages more likely. The very slow process of mining lithium adds another drag to batteries’ production and advancement until other emerging technologies, such as sodium ion batteries, can be improved and manufactured.

So, the drawbacks are significant. But Texas will be counting on greater battery capacity to keep pace with the demands on the grid, especially as drivers transition to electric vehicles and more renewable production comes online.

“We need a hundred times more storage, looking at the next 10 years,” Yao says. “The cost of the battery really has to go down further to enable the bigger, full-scale transition.”

Microgrids

Think of your traditional grid as a one-way system in which a producer generates energy at a remote location and then a complex network delivers energy to your home or business. That system is vulnerable to extreme events — storms, heat waves — because it must constantly balance supply and demand. When a producer or some portion of the delivery network breaks, you can lose power.

To prevent a loss of power, many people around the world are turning to smaller, more contained power-generating and transmission networks, known as microgrids. When connected to the grid, microgrids can operate in harmony with that wider network.

And when disconnected from the grid, microgrids can stay up on their own, powering a specific user — a university, say, or maybe a warehouse or even a neighborhood — for two or three days. Communities in California, Chicago and Detroit have invested in microgrids in recent years.

Jian Shi, an associate professor of electrical power engineering technology in the Cullen College of Engineering, was part of a team that sent a proposal to the U.S. Department of Energy for what’s called a virtual power plant for Harris County. If funded and implemented, the project would link the county’s distributed energy sources — to be located at different community buildings, such as libraries and community centers — and help them work in concert.

“A virtual power plant provides a framework that we can integrate all these energy sources — batteries, solar panels and the building management system located in different neighborhoods,” Shi says. “And we’re able to operate all of them simultaneously under a unified platform as a combined source of power generation.”

The benefits to a microgrid that runs on renewables are multifold, Shi says. On a normal day, that microgrid can generate its own electricity, thus keeping its power bills to a minimum. On a very hot, sunny day, when demand is at its peak, the microgrid may produce enough electricity that it can contribute — that is, sell — some back to the larger grid, further offsetting its costs. And in the event of an outage on the grid, the microgrid can sustain itself for hours or even days, potentially saving lives along the way.

“That is really a win-win situation if we can make this kind of ‘exchange’ smart enough,” Shi says. “We want to make sure we take advantage of the different energy prices or different electricity prices in ERCOT so that we can maximize the benefits of renewable energy to power the community while supporting the broader grid.”

AI solutions

As the state’s electricity demand increases rapidly due to industrialization and population growth, and as renewables replace less volatile sources of power, utilities and consumers alike will need to optimize the power flowing through the grid or a microgrid. The complexity and speed necessary for such a task makes it perfect for a machine to handle.

An AI tool that balances inputs from batteries and other clean backup resources (such as renewable natural gas generators or hydrogen fuel cells), weighing the moment-to-moment price fluctuations from different primary energy sources, could help the grid run smoothly at full capacity. And it could do so while ensuring that people who connect their power storage and solar panels to the grid can get a rate of return on their investments.

“In our lab, we are working on AI-based energy routing solutions that can understand when to use the energy storage and other resources carefully,” says Harish Krishnamoorthy, an associate professor in the Department of Electrical and Computer Engineering at the Cullen College of Engineering and associate director of the Power Electronics, Microgrids & Subsea Electrical Systems Center at UH.

“We have to be smart about when to charge and discharge the energy storage systems, for example, depending upon the real-time energy prices so that the owners of these systems can maximize profit by providing services to the grid,” Krishnamoorthy says. “At the same time, how can they ensure that there will be sufficient resources available to satisfy the daily local energy needs and during potential power outages?”

The answers to that equation may not make up for large-scale, storm-damaged hardware or a lack of power plants in the state from the grid’s perspective. But on very hot days, or very cold days, optimizing every potential input may mean a big difference in price and reliability to Texas power customers.