If you are an EV battery pack assembler dealing with inconsistent cell quality in large modules — this project developed quality control systems for stacks with up to 3000 cells that improve accuracy by an order of magnitude.
AI-Driven Digital Twins to Speed Up and Lower Costs of Battery Production
Imagine having a perfect digital mirror of a battery that tells you exactly how it will perform before you even build it. This project uses high-tech microscopes and AI to spot flaws in battery materials instantly rather than waiting weeks for tests. It's like upgrading from a slow manual inspection to a high-speed digital scanner for energy cells.
What needed solving
Battery manufacturers struggle with high production waste and slow quality control cycles. Current testing methods are often too slow to keep up with agile manufacturing, leading to higher costs and lower safety reliability.
What was built
A GHz electrochemical microscope for fast material analysis and AI-linked digital twins for battery cell and module performance. They also built automated test setups for modules with 300-500 cells.
Who needs this
Who can put this to work
If you are a battery startup dealing with slow material testing for new chemistries — this project developed a GHz electrochemical microscope that increases characterisation speed by a factor of 5.
If you are an equipment provider dealing with high material waste during production ramp-up — this project developed digitally integrated process models for pilot lines that reduce CO2 footprints and waste.
Quick answers
How does this impact production costs and pricing?
Based on available project data, the project aims to reduce production costs and materials waste through AI-driven process optimization and improved quality control.
Is this technology ready for industrial scale?
Yes, the project demonstrates its tools in three pilot lines and develops test methods for integrated automotive module stacks containing up to 3000 cells.
What is the IP or licensing status of the algorithms?
The project focuses on developing a new approach for open-source algorithms and data standardization strategies.
How does it integrate with existing manufacturing lines?
It uses digitally integrated process models and calibration methods specifically designed for pilot lines and automotive field applications.
What is the timeline for implementation?
The project period runs from 2024-01-01 to 2026-12-31, indicating the development and validation phase is currently active.
Who built it
The consortium is well-balanced for commercialization, featuring 14 partners across 7 European countries. With a 36% industry ratio (5 industrial partners, including 3 SMEs), the project ensures a strong link between academic research (4 universities, 4 research centers) and market application, led by a global technology leader, Keysight Technologies.
Contact Keysight Technologies GmbH in Austria
Talk to the team behind this work.
Contact us to connect with the DigiCell consortium for pilot line integration.