Have you ever been on a sailboat when the wind kicked up and a big gust tossed you into the water? (Not that I have, of course. My friend maintains that he is an “expert captain.”) Or have you experienced the wind dying down and it’s like you’re going nowhere fast? (He’ll deny that one too.) Wind has a similar effect on the power grid – but if you know when the gusts are coming, you can adjust accordingly.
Wind is one of the most difficult weather variables to forecast (as my friend can attest). So many factors affect it – topography, ground cover, temperature inversions – even the number of leaves on nearby trees.
Keeping energy supply and demand in close balance is a complex and constant challenge for utilities. When hundreds of wind turbines ramp up and down at nearly the same time, system operators have to quickly adjust other power generation resources to accommodate the fluctuations.
Xcel Energy relies on wind energy a lot more than the average utility. Fortunately for us and our customers, the weather experts at the National Center for Atmospheric Research are cracking the code to reliable wind forecasting.
You’ve heard of Windtalkers; these folks are Windreaders.
A High-tech Crystal Ball
“One of the major obstacles that has prevented more widespread use of wind energy is the difficulty in predicting when and how strongly the wind will blow at the wind farms,” says William Mahoney, the NCAR program director overseeing the project.
NCAR’s state-of-the-art forecasting system combines real-time, turbine-level operating data with weather-prediction models and sophisticated algorithms to forecast wind energy out for 72 hours. The forecasts help operators make critical decisions about powering down coal- and gas-fired power plants when sufficient winds are predicted.
The task is especially challenging because researchers are actually pinpointing the breezes near the turbines, which are typically 200 to 400 feet above the ground. Winds at these heights tend to be much stronger than those measured by the ground-level weather stations that top out at 33 feet.
The Results
The NCAR folks have honed their forecasting system so accurately that Xcel has been able to idle coal plants for up to three days at a time – when the plants would otherwise be running.
That’s huge. It’s one thing to power down natural gas plants, which are designed to come on- and offline quickly. It’s another to idle a coal plant that can take more than a day to restart. The stakes are high, and accurate forecasting is crucial.
Once the NCAR researchers are satisfied that their forecasting system is ready for prime time, they plan to apply it more broadly. As unpredictable as the wind may seem to the rest of us (especially to that “expert captain”), it’s great to see that there are people who can crack the forecasting code and help us use less coal.
Now if only my friend can apply that forecasting technology to his boating…
Great thoughts and very insightful! Keep up the good work Xcel
That’s pretty cool. I had no idea that it was possible to idle a coal plant for 3 days based on wind capacity! I’m curious about a couple things:
I think wind turbines can also be idled to avoid overproduction, yes? Although I like the fossil-fuel idling approach better!
What are the emissions costs of idling & restarting a coal plant? I assume steady-state has lower emissions than start-up, I wonder what the break-even downtime is for lower emissions…
Hi Eric,
You’re right that turbines can be shut down to avoid overproduction, but we prefer to keep them running (for obvious reasons). They also can be shut down to avoid damage if the wind is too strong.
You’re also right that when a coal plant is ramping up, its emissions are slightly higher because it’s not running at its highest efficiency. The ramp-up emissions don’t represent a huge spike though; they’re just slightly higher than when the plant is running at optimal efficiency. Plus, our plant operators are trained to get the plants to high-efficiency operating levels as quickly as possible.
Our environmental compliance experts said that any time we can get wind on our system, it’s worth it to ramp down the plants not running on renewable energy. The non-optimal operating time and the associated emissions are nominal. When I asked them if five minutes of wind produced a net decrease in emissions, they said, “Absolutely.”
Thanks for the great questions.
Sheila
Nice entry. How often does this happen, and where?
Thanks Christof! Great to see you on the blog.
I’m told we can’t share production details at that level because it could have an impact on the price of energy we buy from and sell to other utilities. (They wouldn’t even tell me “off the record” due to FERC regulations. They said something about having to kill me if they told me…)
I posed your questions to Beth Chacon, environmental policy manager. Here’s her reply:
“The idling was just one example. Every day we ramp down our gas and coal plants to accommodate wind energy. We are averaging nearly ten percent wind energy on our system on an annual basis, and we expect that, over the next decade, wind will account for about a third of our expected CO2 reductions. The remainder of our CO2 reductions are achieved through customer energy efficiency programs and improvement in plant efficiencies, including the retirement of older coal plants.”
Thanks for asking!
Sheila