It often takes time for power system malfunctions to be found and fixed, at times leading to larger system failures.
If operators could identify system disturbances as they happen and take action before they lead to large outages, the power grid would be more reliable and resilient. With recent support from the National Science Foundation, Meng Wang, an associate professor of electrical, computer, and systems engineering at Rensselaer Polytechnic Institute, is developing software to make that real-time analysis possible.
“If something happens in the system, we want to know as soon as possible where it is and what type of event it is so that the operator can take actions to fix that,” Wang said.
Wang is working with Joe Chow, an Institute Professor of electrical, computer, and systems engineering at Rensselaer, to develop machine learning and data analytics tools that can quickly extract information from the many measurements already being taken as operators monitor the power system.
These tools will become even more essential as the nation moves toward integrating more renewable power sources. For example, Wang said, the renewable energy from solar and wind is very volatile, which requires the power grid to be more prompt and flexible in managing the power demand and supply.
Advanced data analytics will enhance the real-time situational awareness of the operator about the power grid.
“With the increasing integration of renewables, it is really essential for us to know what happens in the power system immediately,” Wang said. “We want to be able to balance the demand and the supply continuously.”
The Rensselaer team is also working with a team from Cornell University that is focused on developing security hardware tools aimed at fortifying the future power grid as it becomes both more intelligent and more digital.
The goal of this collaboration is to create tools that can easily be used by the industry. Wang said she has already started developing data analytics tools that are being tested by power companies.