Digital Tools and Services

In the field of “Digital Tools and Services,” we focus on using our broad and structured base of curated data to develop innovative digital tools and services with our domain knowledge. These serve to automate, facilitate, and accelerate the analysis of battery data. 

Our work includes developing tools for easily visualizing and quickly evaluating complex data. By linking data points along the entire value chain, we can incorporate the entire data history into our analyses. In addition, we use our experimental data to develop empirical models. 

When designing our software, we draw on the logical description of our processes and experiments. By consistently describing the laboratory and simulation processes, we ensure interoperability and flexibility when using our tools. This enables us to develop customized solutions for our customers' specific needs. 

Our goal is to see digitalization as an opportunity to make battery research and production more efficient and effective. We strive to open up new possibilities and solutions for industrial contract research and collaborative projects by using digital tools and services.

We use the battery and cell models developed at Fraunhofer ISE and their parameterization to support battery cell, module, and system manufacturers in the realistic simulation of various operating scenarios. In general, these are equivalent circuit models that represent the electrical and thermal behavior with high accuracy and map the aging behavior according to the current state of science. In addition to empirical models, physically motivated and theoretical models (e.g. pseudo-2D model) can also be created. By using the Functional Mock-up Interface, our models can be used flexibly.  

The high accuracy and validity of our models allows us to generate synthetic data that can be used for machine learning. These can also be used in addition to laboratory measurement data or field data. In turn, model-based state estimators (e.g. Kalman filters) are used to generate the labels necessary for supervised learning, which the synthetic data already contains.