Electromobility, heat pumps, etc. are increasing the demand for electricity. But in many countries, the power grids are not designed for this. Researchers from the USA see artificial intelligence as a solution to avoid blackouts.

Solar, wind and hydropower are intended to drive the energy transition. But in many countries, the power grids are currently unable to cope with the rapid growth of renewable energy sources.

According to a study by the International Energy Agency (IEA), around 80 million kilometers of electricity grids will have to be newly built or modernized worldwide by 2040. This is due, among other things, to the fact that electric cars and heat pumps will cause electricity demand to rise enormously in the coming years.

This could endanger the security of supply and lead to blackouts. But researchers at the University of Texas at Dallas and the University at Buffalo want to counteract this with a “self-healing power grid” using artificial intelligence.

Self-healing power grid – how artificial intelligence should help

In their publication in the journal Nature Communications, they explain a first approach to a self-healing power grid. It uses artificial intelligence that can work autonomously in the event of an impending blackout.

The AI ​​can detect problems such as an impending power outage or damaged wires. After analyzing the problem, it can reroute power within milliseconds to avoid blackouts.

“Our goal is to find the optimal way to send power to the majority of users as quickly as possible. But more research is needed before this system can be implemented,” explains Jie Zhang of the University of Texas at Dallas.

At the moment, the AI ​​cannot work independently. The processes still have to be controlled by humans. Redirecting the power can currently take a long time.

AI will also be able to carry out repairs in the future

The researchers are using machine learning for the self-healing power grid. This allows the AI ​​to understand the power grid and the complex relationships between the power lines.

Once the AI ​​system has understood these relationships, it can use reinforcement learning to make the best possible decision in an emergency. If a power outage is imminent, the system can divert power from another source nearby. These could be large solar systems or battery storage systems, for example.

In the future, however, the researchers do not just want to prevent power outages. Their “self-healing power grid” should be so automated that it can also help repair power outages.

Also interesting:

Source: https://www.basicthinking.de/blog/2024/08/13/stromnetz-kuenstliche-intelligenz/

Leave a Reply

Your email address will not be published. Required fields are marked *