Investigators and developers for the AI-Grid Platform (from left to right): Kamini Shahare, Rong Zhao, Yifan Zhou, Xin Wang, Khezr Sanjani, Zachary Lerman, Zimin Jiang, Peng Zhang and Tianyun Ling inside the Power Lab located in the CEWIT Building. Photos by John Griffin.

Peng Zhang, a professor in Stony Brook University’s Department of Electrical and Computer Engineering, is working on an AI-grid project to make power grids more resilient.

Self-sufficient small energy systems, known as microgrids, can help upgrade power reliability. Microgrids are especially useful during disasters such as hurricanes where power outages are more likely to occur. Networked microgrids (NMs) that work together have the ability to raise these benefits; however, networked microgrids pose some challenges. The NMs are complex, contain a lot of data and pose cybersecurity risks.

AI grids help tackle these issues. AI grids are an AI-powered system that make NMs more reliable. “We use AI to control and to make the system way more resilient,” said Zhang.

“We use AI to learn the model out of the data. Then we use the learned model to control the microgrid, and it has worked really well.”

Professor Peng Zhang

AI-Grid offers several benefits for its consumers, which range from civilian to military. These benefits include lower costs, improved reliability and better protection against cyber threats.

“We are helping utilities, we are helping the renewable energy vendors,” said Zhang. “We also put in a significant amount of effort to help with the Department of Defense to help with our naval military bases to make sure their systems are resilient against different attacks.”

This project is supported by the New York State Center of Excellence in Wireless and Information Technology (CEWIT) at Stony Brook University. Working closely with the research group led by Zhang and co-investigators Profs. Xin Wang, Yifan Zhou, Scott Smolka and Scott Stoller,  CEWIT engineers are developing the data management and visualization capabilities for AI-Grid and supporting the testing and evaluation performed with Hitachi, ConEd, and other partners in the utility industry.

The project collaborates with numerous organizations to test and bring the product to market. The AI grids will be tested in partnership with multiple companies, including Hitachi and ConEd.

“Our mission at the Energy Solutions Lab at Hitachi America R&D is to democratize the grid to accelerate the adoption of renewable energy sources,” said Bo Yang, vice president and head of the Energy Solutions Lab at Hitachi America R&D. “Our collaboration with Stony Brook University proved that utility companies can leverage the cloud to bolster the use of micro-grids and deliver efficient, low-cost, clean energy to the communities they serve.”

In collaboration with Stony Brook University, Hitachi has developed a cost-efficient edge device for the microgrid. With the device, costs are reduced by approximately 50 percent compared to commercial alternatives.

Angelina Livigni

More on SBU News