23. Internationales Stuttgarter Symposium

Automobil- und Motorentechnik

4. - 5. Juli 2023

Session: Battery I | | 10:05 - 10:35

Distributed Co-Simulation for effective Development of Battery Management Functions

Matthias Scharrer, Virtual Vehicle Research GmbH

Electrification calls for a range of system components, that need to be developed and tested. On one hand, these system components include hardware components, like battery cells, modules, and packs, and battery management systems (BMS). On the other hand, software components are used for testing, e.g. algorithms for BMS or simulation models for batteries. Execution of tests on real batteries is typically time- and cost-intense, and includes considerable risks, leading to safety hazards. In this paper, we introduce a novel development and test approach for battery systems, that is driven by a unified, standardized interface between hardware- and software components and physical devices alike. Whereas established Hardware-in-the-Loop (HiL) systems are built on proprietary systems and environments, our approach is based on both open-source and industrial simulation software solutions. The Distributed Co-Simulation Protocol (DCP) is used to encapsulate and virtualize these components, as shown in a demonstrator use case. A "DCP master" is used for effective configuration and re-configuration of these "DCP slaves". It allows to seamlessly exchange real hardware components of the system (e.g. Battery modules, vehicle power train, Balancing circuits) with software components (e.g. Simulation models, Vehicle behaviour/electric motors) and vice versa. The proposed demonstrator use case is composed of several "DCP slaves", i.e., the BMS hardware, communication gateways and simulation models, connected via an easily configurable "DCP master". The test setup can be installed in an office environment without additional safety requirements and validated using a continous integration approach for automatic BMS software deployment and testing, as the simulation models may replace the battery cells and electric loads. Using these models streamlines the testing of BMS functions, as it eliminates the test preparation times. The concept is promising to simplify the distributed development of components with many different test cases.