Browse By

How live digital twins are helping power plants look ahead

Of all the challenges power plants face at present, keeping pace with innovation may be the most critical. Whether it’s incorporating renewable energy sources, fending off cyberattacks or simply updating to the latest, most efficient software, power plants are under increasing pressure to stay current in every facet of their operations.

As anyone with a smartphone already knows, updating in the digital age often means downtime — a luxury that power plants can hardly afford. But simply flipping the switch on new software or a new power source isn’t really an option for a facility pumping out 600 megawatts of power a day. To stay current, power plants need a reliable way to test their new systems in a safe, consequence-free environment that mirrors, as closely as possible, real-life operating conditions.

Simulation vs. reality
For years, power plants have relied on digital twins to create these safe testing environments. A popular tool among technology and engineering companies, a digital twin is a virtual replica of any physical mechanism or process that can be safely altered and manipulated to test how the real asset might respond. Think flight simulators, but for satellites, locomotives and offshore oil rigs. Doctors can even make digital twins of patients to see how they might respond to risky medical procedures.

But living organisms can pose a challenge for digital twins. Traditional digital twins are essentially a snapshot in time — they replicate their host object as it was at the moment the twin was created. But they don’t grow or adapt alongside their host. “There’s a slow, inevitable decline in the accuracy of that twin, because after you create it, things change,” said Bob Yeager, president of Emerson’s power & water solutions business.

Make no mistake: Power plants are living, ever-shifting organisms. The power plant you make a digital twin of today will look and operate differently than that same power plant six months or a year from now. “With power plants, a traditional digital twin can degrade to the point where it almost becomes unusable,” said Yeager. “In six months, that digital twin will practically be worse than having nothing at all, because it becomes disinformation.”

“These digital twins run parallel with the real plant. They are automatically updated, so if the plant changes, the digital twin changes. It’s always up to date and always accurate.”
BOB YEAGER,
PRESIDENT OF EMERSON’S POWER & WATER SOLUTIONS BUSINESS
A revolutionary new approach
Luckily, by taking advantage of virtual computing technology, Emerson has developed a new kind of digital twin that revolutionizes how power plants onboard new systems. Known as a live or embedded digital twin, these virtual replicas use data provided by real-world sensors to accurately mirror a power plant’s live environment. “These digital twins run parallel with the real plant,” said Yeager. “They are automatically updated, so if the plant changes, the digital twin changes. It’s always up to date and always accurate.”

Though this technology represents a leap forward in testing for many industries, it is particularly transformational for power plants. When power plants upgrade to a new system, they typically undergo a short, planned outage. It’s a brief window of time during which the plant can test the new system without fear of knocking the entire plant offline or causing some unintended malfunction.

But as critical as it is to thoroughly test a new system, power plants have obligations that make it untenable to extend these outages for long. Not only must they provide power for their local customers, but most plants also typically have purchase power agreements with other localities. “Outages are usually very short in duration because plants have already sold power in advance, and there’s usually a penalty if they fail to meet those obligations,” said Yeager.

The nightmare scenario is to discover a major problem with the new system partway through such an outage. “If you find major problems four days into a one-week outage, and you have to delay everything — that’s not good,” said Yeager. One solution is to test the new system as much as possible before the outage to avoid discovering major issues under extreme time pressure. With a live digital twin, the new system could be tested for weeks or months in an exact virtual landscape, so when the planned outage arrives, operators know what to expect. That not only saves the plant time and money, but it also saves the public from a potentially crippling power shortage.

“The live digital twin allows you to create this new world where you can implement all these software logic changes, cybersecurity responses, technology updates, in a very, very accurate environment, so when it comes time to actually load it onto the plant, you know exactly what to expect,” said Yeager.

The impact on the future workforce
Live digital twins are also changing how power plants attract and train a new generation of workers. Historically, because of the enormous public safety risks inherent to running a power plant, new technicians had to be on the job for years before getting to contribute in a meaningful way.

With the live digital twin, new technicians can rack up hours of experience running the plant without posing a danger to anyone. “They can experiment and trip the unit 10, 15 times a day and innovate,” said Yeager. “Maybe they end up really making a difference, but you would never have known that if you didn’t have this tool.”

The next 10 years will be critical for America’s power plants. As renewables and natural gas make up a greater percentage of our country’s fuel source, plants will be under enormous pressure to adapt. At the same time, much of the utility and electricity workforce is nearing retirement, making it critical for companies to capture institutional knowledge gathered through decades of working on the plant floor. Tools like the live digital twin will not only help plants stay competitive, they will keep the public safe and the lights burning.