Project information:
Project Partner:
Motivation:
Today, the properties of battery materials are determined at each development step by means of elaborate series tests. The heterogeneous data obtained in this process is so far difficult to link with data from manufacturing process and system characterization in order to compare them with each other. DigiBatMat creates a platform for interoperable management of battery materials data that enables predictions of quality and performance through machine learning and correlation analysis. The complexity of the "battery" system and its heterogeneous materials will be accounted for by an ontology in which the components are clearly described at different scales. In this way, material data and knowledge can be logically stringently linked and queried.
Project description:
As part of MaterialDigital and ProZell, the DigiBatMat project creates a platform that combines taxonomies, ontologies, structured data management and methods for data analysis based on four applications. The data management application "DigiBatEin" allows users to integrate their own data into the platform. Via the "DigiBatQ" application, the user accesses the ontologies developed in the project and is informed which data might be relevant for answering process and material related questions from the perspective of current knowledge and user hypotheses. The application "DigiBatML" is trained with the data collected in the project and is used to predict the quality of cells based on concrete data from the user. With the application "DigiBatKor", the user can perform linear and/or nonlinear data analyses and determine which data are evaluated in the process.
Within the sub-project, the Institute for Particle Technology (iPAT) and the Institute for Machine Tools and Manufacturing Technology (IWF) of the Technische Universität Braunschweig (TUBS) provide their competences in the field of electrode and cell manufacturing as well as process and (intermediate) product analysis, which are particularly implemented in the development of the taxonomies and ontologies. Data from production (e.g. structural parameters of raw materials and intermediate products) as well as expertise from completed and ongoing projects and from the Battery LabFactory Braunschweig (BLB) serve as the basis for the development of the DigiBatMat platform and its various applications. Accompanying the provision of data, process chains for the production of electrodes with different electrode materials are to be established. The different material combinations serve several research questions.
Contact:
Dr. rer. nat. Milena Perović
m.perovic(at)tu-braunschweig.de
Philip Gümbel
p.guembel(at)tu-braunschweig.de