According to their research, blazing fast computation blended with AI could rapidly diagnose trouble in energy grids and find solutions in tiny splits of seconds, preventing problems such as voltage variations or widespread blackouts.
“Energy power system failures are an old problem and we are still using classic computational methods to resolve them,” says Fengqi You, the Roxanne E. and Michael J. Zak Professor in Energy Systems Engineering in the College of Engineering. “Today’s power systems can benefit from AI and the computational power of quantum computing, so power systems can be stable and reliable.”
Power grid outages commonly occur due to storms, downed trees, ancient transmission lines, and other misfortunes, with U.S. customers experiencing on average more than four hours of electric energy interruption in 2016, rising to nearly eight hours in 2017, according to the federal U.S. Energy Information Administration (USEIA). Consumers reportedly suffered about six hours of energy interruption in 2018.
To solve these issues, the researchers propose a first-time, novel hybrid solution by creating a quantum-computing-based “intelligent system” approach to build a fault-diagnosis framework to accurately find problems in electrical power systems. In their paper, the researchers demonstrated the efficacy and scalability in a large-scale IEEE test electric power system.
In it, they found that a quantum computing-based deep-learning approach can be scaled efficiently for a quick diagnosis in larger power systems without loss of performance. The researchers say they believe that quantum computing and artificial intelligence can save most of the system failure.
“Integrating quantum computing with intelligence – even though it is not yet a mature technology – will solve real problems now,” Ajagekar said. “It works very well.
“We cannot afford for grids to go down,” says doctoral student Akshay Ajagekar, who co-authored the paper. “That’s why fast fault diagnosis is in electrical power systems is very important. Today’s systems have sensors, but even they’re not good enough now. We need efficiency. It’s