Applied Edge AI for Batteries

Important results from the project

The purpose of the project was to test distributed Edge AI locally on batteries. The project focused on forecasting Remaining Useful Life (RUL) in individual batteries. Overall, the project has resulted in expertise on how batteries RUL can be forecasted and modeled with the help of modern Data Science-metoder. The project has also built up an ability to run an edge platform that communicates with the battery´s BMS. These skills and insights form a good basis for further work on developing intelligent battery solutions.

Expected long term effects

The project has established an ability to run an edge platform for batteries, which in itself enables a variety of applications. In addition, the project has developed a basic ability to forecast the remaining life of batteries. In the next steps in the work, these capabilities will be further developed so that they can later be implemented in products on the market, and thereby contribute to the shift from internal combustion engines to battery operation.

Approach and implementation

The project´s planned layout followed a sequential process, where the implementation of the edge platform would be followed by data collection and modeling. In the light of hardware challenges related to the battery, the actual implementation had to take place instead in parallel currents. The establishment of the edge platform was therefore conducted in parallel with the modeling work.