Author(s): Treesa Rose Joseph, Sylvain Gouttebroze, Daniel Marchand (SINTEF)

In the ever-evolving landscape of battery manufacturing, there is a need to facilitate deeper collaboration between different entities of battery research and manufacturing plants. As the demand for energy storage surges, particularly driven by the electrification of transportation, the battery manufacturing industry faces significant challenges. The rapid establishment of numerous Gigafactories to meet this demand intensifies competition, especially against China’s cost-effective and efficient production methods. To address these challenges effectively, there’s a critical need for intelligent machinery capable of fast adaptation and optimisation, leveraging AI technology. In the EU-funded project: BATMACHINE [1][2], we aim to create intelligent machinery to improve battery production.

Imagine this as the blueprint for an optimised battery manufacturing hub. We envision a collaborative and FAIR [3] (Findable, Accessible, Interoperable, Reusable) data space (i.e. data repository with all the time series data collected from machinery used in the battery production line). Consequently, we use this data in the realm of process control to optimise the manufacturing process.
To facilitate efficient communication and data sharing, we will implement digital interfaces that seamlessly connect battery manufacturing equipment, process control systems, and engineers. These digital highways will lay the foundation for improved collaboration and secure information exchange.

Dataspace: FAIR data repository

Our dataspace comprises all data gathered from the equipment in the battery production line. To ensure the data is reusable as part of the FAIR objective, we will integrate application ontologies—a common and structured vocabulary to describe battery materials, measurements, and manufacturing equipment. As a result, this standardised language will enable our digital systems to communicate effectively. In turn, this will ensure a smooth understanding between different components of the battery manufacturing process. The data, stored in our databases, is thoroughly documented by linking each parameter to semantic concepts defined in the ontology. Consequentially, this will ensure it’s easily findable and understandable by different machines.

Building Bridges: Development of Digital Interfaces

We will be developing two types of digital interfaces. The first type will register manufacturing equipment and measurements, populating automatically our dataspace. The second type will connect the dataspace to the process control, providing information about historical equipment behaviour. In addition, this interface will feature equipment state in the previous steps to optimise in real-time, and ensure high quality and low scrap in the manufacturing process. These interfaces are designed to harmonise data from different types of equipment, ensuring that they speak the same language. In fact, our goal extends beyond mere connectivity. Indeed, we aim for semantic interoperability, that is, making sure that machines and control systems understand each other.

Unlocking the Potential: Exploration and exploitation of data

Once the system is in place, our dataspace will be a rich data repository that can be explored and exploited by both humans and artificial intelligence alike. Hence, we will be developing a set of tools that will enable us to search and connect to the data in the dataspace. Moreover, Machine Learning packages can make use of data from this dataspace to develop new data-based models.

Fig.1: Components of the Deployment kit interacting with Battery manufacturing hub

All the above parts will be bundled together as a configurable deployment kit that can then be installed on-premises in battery manufacturing plants. This is particularly important, as most industrial equipment may not allow interaction with cloud solutions for security reasons. By building robust digital foundations, connecting manufacturing elements seamlessly, and effective utilization of collected/generated data, we are contributing to the improvement of the battery manufacturing process.

References

[1]: Website: https://batmachineproject.eu/
[2]: BATMACHINE project, funded by HORIZON-CL5-2022-D2-01, the European Union’s Research, and Innovation Programme (Grant Agreement no 101104246).
[3]: Wilkinson et al., The FAIR Guiding Principles for scientific data management and stewardship, Scientific Data volume 3, 160018, 2016.