Linking quantum computing to AI leads to a smarter electricity grid

Searching for flashlights during blackouts may soon be a distant memory, as quantum computers and artificial intelligence could learn to decipher the problematic quirks of a power grid and fix system failures so quickly that people might not notice.

Instead of power grid failures becoming gigantic problems — such as voltage variations or widespread blackouts — blazing-fast computation mixed with artificial intelligence can quickly diagnose problems and find solutions in tiny fractions of seconds, according to research from Cornell in Applied Energy (Dec. 1). . , 2021).

“Errors in energy systems are an old problem and we still use classical computational methods to solve them,” said Fengqi You, the Roxanne E. and Michael J. Zak Professor of Energy Systems Engineering in the College of Engineering. “Today’s energy systems can take advantage of AI and the computing power of quantum computing so that energy systems can be stable and reliable.”

You, along with PhD student Akshay Ajagekar, are co-authors of “Quantum Computing-based Hybrid Deep Learning for Fault Diagnosis in Electrical Power Systems.”

U.S. utilities generated about 4 trillion kilowatt hours in 2020, according to the federal U.S. Energy Information Administration (USEIA). This electricity is transported through regional grids, but disruptions occur due to storms, fallen trees, old transmission lines and other setbacks.

For example, in 2016, U.S. customers averaged more than four hours of power outage, while that average rose to nearly eight hours in 2017, according to USEIA. Consumers experienced approximately six hours of power outage in 2018.

The scientists propose a first, new hybrid solution by creating a quantum computer-based “intelligent system” approach to build a fault diagnosis framework to accurately find problems in electrical power systems.

In the paper, the researchers demonstrated efficacy and scalability in a large-scale IEEE test flow system. In it, they found that a quantum computer-based deep learning approach can be efficiently scaled for rapid diagnosis in larger power systems without performance loss.

You and Ajagekar believe that quantum computing and artificial intelligence can prevent most system failures. “Integrating quantum computing with intelligence – although it is not yet a mature technology – will now solve real problems,” Ajagekar said. “It works very well.

“We can’t afford to lose the nets,” Ajagekar said. “That’s why rapid fault diagnosis in electrical power systems is very important. Today’s systems have sensors, but even those aren’t good enough right now. We need efficiency. It is very expensive to wait minutes, hours or days.”

As society moves towards a greener environmental future, the ubiquity of electricity will become more important. “Electric power systems are the backbone of our modern world,” said You, a faculty member at the Cornell Atkinson Center for Sustainability. “The marriage of quantum technologies and AI can make a difference in our daily lives.”

This study used resources from the Oak Ridge Leadership Computing Facility, which is part of the US Department of Energy. In December 2020, You and Ajagekar published a paper on quantum computer-based deep learning for flaw detection and diagnosis in industrial manufacturing, following a US patent application.


Leave a Reply

Your email address will not be published.

Back to top