AI harnessed to help prevent power outages
Dr Jie Zhang (c.) with co-researchers Dr. Yulia Gel and Roshni Anna Jacob. Image: UT Dallas
University of Texas at Dallas researchers have developed an AI model that could help prevent power outages by automatic re-routing of electricity in real time.
The approach, offering an advance on an early example of a ‘self healing grid’, is aimed to use AI to speed up this process from the minutes to hours that a human operator could take to a few milliseconds.
While the model is yet to be implemented in the field, various scenarios in test networks have demonstrated that it can automatically identify alternative routes to transfer electricity to users before an outage occurs.
“Our goal is to find the optimal path to send power to the majority of users as quickly as possible,” says Dr Jie Zhang, associate professor of mechanical engineering in the School of Engineering and Computer Science, who led the research.
“But more research is needed before this system can be implemented,” he adds.
Have you read?
AI in the energy transition: problem or solution?
AI challenges in Europe
The research, which is published in the journal Nature Communications, was undertaken in collaboration with engineers at the University at Buffalo in New York.
The distinctive characteristic of the approach is that it explicitly accounts for the underlying network topology and its variations with switching control, while also capturing the complex interdependencies between the variables by modelling the task as a graph learning problem.
Application of a machine learning neural network model to the graphs enables mapping of these complex relationships between the entities that make up a power distribution network and how the electricity moves through the system.
With this, the approach relies on reinforcement learning that makes the best decisions to achieve optimal results.
So for example, if electricity is blocked due to line faults, the system is able to reconfigure using switches and draw power from available sources in close proximity, such as from large-scale solar panels or batteries on a university campus or business.
