The challenge

The Nordic high-voltage grid is extremely reliable. Despite this, undesired incidents sometimes occur. When faults occur in the power system, it is important to act quickly. In the power system of the future, characterised by more wind and solar power, higher volumes of information and stricter requirements for efficient management of society’s resources, we will need to make more use of automated decisions to operate the electricity grid. The extremely complex nature of the power system appears particularly suited to the use of artificial intelligence (AI) in decision-making.

When planning future power lines and infrastructure, it is important that Statnett can perform calculations for various situations that could arise. AI-based calculations can give quicker answers on action to be taken in fault situations and challenging operational conditions. Additional calculation capacity will also allow us to investigate a wider range of fault and operational scenarios during the same period. Together, these will provide a better decision-making basis for long-term investments.

 

The goal

The goal is to develop methods and models for handling faults in the power system within the framework of reinforcement learning. The project will create tools that provide decision-making support using an AI-based model. This will provide quicker answers on the most effective way to handle undesired incidents in the network.

 

The project

The project will help speed up simulations of various operational situations so we can make greater use of calculation-intensive, probabilistic methods when analysing security of supply. Through internal and external presentations, and publications in relevant fora, the PhD project will highlight how artificial intelligence can be part of the power system of the future and how operational decisions can be changed as a result. It could also have major transfer value for the sector as a whole.

 

Project participants

  • The Norwegian University of Science and Technology (NTNU)
  • Statnett

 

Funding

The project is being implemented as an Industrial PhD and is funded by the Research Council of Norway’s Industrial PhD Scheme (and Statnett, with guidance from NTNU).