I don’t often get an email from a colleague in Scotland asking how I’m weathering the power cuts in Texas. But I did last Friday, after the worst was over.
For much of last week, millions of Texans had to endure the loss of electric power, and all that entails, during some of the coldest weather on record.
Early Monday morning, Feb. 15, the Electric Reliability Council of Texas (ERCOT), the nonprofit organisation that operates the Texas power grid, ran out of power. That is to say, the soaring demand due to millions of heating units working overtime exceeded the dropping supply due to equipment failures caused by the same cold weather.
In order to avoid an uncontrolled system crash that would take weeks to recover from, ERCOT commanded its grid customers (operating companies that distribute the power and collect power bills) to implement rolling blackouts, handing out percentages of their load they had to shed. How they shed it was up to them, but shed they must.
This put increasing numbers of electric customers in the dark, both physically and informationally. Although many people could access news and websites through their phones, there wasn’t that much to learn about how the power cuts were being decided or how long they would last.
Here in San Marcos, our house had power all day Monday. That evening I was on a Zoom conference with some people in Austin, who’d been told that the rolling blackouts would last only an hour or so. Instead, it seemed that almost everybody in Austin lost power and stayed in the dark except for a lucky few who happened to be on a “critical feeder” that powered a hospital or fire station.
In the middle of the Zoom call, our power went out. My laptop battery kept my computer going, though, and I managed to power our modem and wireless network with an emergency battery powerpack and inverter, and I got online long enough to complete the call. But then I went to bed at 8:30, as it was getting cold in the house.
We had an hour or two of power every so often for the next two days, but it was mostly off until Wednesday afternoon. Many people in Austin and other parts of Central Texas fared much worse, losing power for two or three days straight, and when their pipes froze they had to seek out an emergency shelter or stay with friends.
Like many engineering failures, this one had multiple causes. On Friday, the IEEE (Institute of Electrical and Electronics Engineers) Smart Grid Initiative sponsored a webinar panel discussion on the crisis, and much of what follows is taken from the information presented in that webinar.
As you may know, Texas has its own power grid that operates independently of the grids in the rest of the country. A “grid” is defined by a region where all the power is synchronised to 60 Hz and can be fairly easily shipped back and forth as the need arises. For historical reasons having to do with the criticality of Texas infrastructure during World War II (and a bit of Texas stubborn independence), most of Texas is covered by the grid supervised by ERCOT. There are a few power interties between Texas and other grids, but they require special equipment and cannot provide significant amounts of power transfer.
So when an extraordinarily cold air mass charged through Texas beginning on Sunday, ERCOT was going to have to handle it by itself. On paper, they were prepared. About half of the 80+ gigawatts (billions of watts, abbreviated GW) of nominal capacity for the grid is natural-gas generation. ERCOT has over 500 generating stations of various kinds to draw from, and some are more reliable than others.
Coal and nuclear plants are the most reliable kind, but coal plants are being shut down these days and no new nuclear plants are being built. A quarter of Texas’ installed capacity is wind, but wind generators are notoriously unreliable, and once the cold air arrived it stopped moving, idling most of the wind turbines and freezing up many of them.
The usual alternative in such a situation is to turn on gas-fired turbine generators, which in principle can be started up in minutes to augment the grid’s energy sources. But the same weather that caused demand to soar also crippled the natural-gas infrastructure, freezing well heads and control valves and causing other problems that eventually eliminated some 26 GW of gas-fired generators that could have otherwise been used.
Although at the peak of the crisis, the grid was producing some 62 GW of power, ERCOT would have had to come up with another 10 GW to meet the extraordinary demand produced by single-digit temperatures in Houston and other normally balmy parts of central Texas. Hence the rolling blackouts that quit rolling and just sat there until pipes froze and people had to seek warmer shelter.
How could this have been avoided? The panel experts proposed a number of solutions.
One was much better interconnections to other grids, both from Texas to other grids and across North America. This is an expensive long-term solution, costing many billions of dollars over many years, and it would increase the robustness of electric power nationwide. But as one expert pointed out, the grid covering much of the Midwest, just north of ERCOT’s grid, was experiencing its own rolling blackouts of less severity, and had no power to spare. So even if Texas had not been electrically independent, we still would have had blackouts, but perhaps not as severe.
Another idea that has been widely adopted in places like Italy is “demand response.” This is a smart-grid technology that allows the power company to adjust the demand from individual consumers. For example, in exchange for a discount on your power bill, you might allow your electric utility the right to reduce your thermostat setting or cut off your electric dryer in power emergencies such as the one Texas just experienced. If demand response had been tested and widely deployed in Texas, the blackouts might have been less severe, but they would probably still have been needed.
Fortunately, there were not many fatalities directly attributable to the power outages, and so as crises go, this one was greatly inconvenient but not deadly. Better enforcement of winterisation protection for natural-gas plants is the most urgent thing to do that will keep something like this from happening again, but the final call is up to Mother Nature.
This article has been republished with permission from the Engineering Ethics blog.