Your browser doesn't support javascript. This means that the content or functionality of our website will be limited or unavailable. If you need more information about Vinnova, please contact us.

Using Physics-Informed Machine Learning for reusing power system components

Reference number
Coordinator Kungliga Tekniska Högskolan - Skolan för elektroteknik och Skolan f elektroteknik & datavetens
Funding from Vinnova SEK 3 800 000
Project duration November 2021 - November 2025
Status Ongoing
Venture Circular and biobased economy
Call Increased resource efficiency for a circular industry

Important results from the project

Yes, we have met our goal of using PINNs to aid understanding on how power components can be used and designed more sustainably.

Expected long term effects

The project will improve understanding of insulation aging and expand research into new materials. Using Physics-Informed Neural Networks (PINNs), it aims to optimize power system components and explore sustainable production methods. Future steps include developing digital twins and simulating alternative materials, like natural ester oils, to enhance equipment upcycling, efficiency, and sustainability in the power industry.

Approach and implementation

The project was carried out according to the plan. Some difficulties have occurred due to limited data availability. But since we have also joined the CIGRE working group on a similar topic (PhD student Federica Bragone is the part of the working group) we saw that this is common issue for the international industry community. Making the final results of the project even more meaningful. Hence the non-anticipated challenge has actually led to a more important discovery.

The project description has been provided by the project members themselves and the text has not been looked at by our editors.

Last updated 21 November 2025

Reference number 2021-03748