Validation of a Battery Management System based on AUTOSAR via FMI on a HiL platform
Leonard Janczyk, Klemens Esterle, Stephan Diehl, Michael Seibt, Arthur Gauthier, Guillaume Viry · Linköping electronic conference proceedings · 2016
In systems which are sourcing their electric energy from a battery system, such as electric or hybrid electric vehicles, it is of crucial importance to monitor the battery's condition in order to ensure its usability and longevity.The battery management system (BMS) is a control unit which supervises the physical variables in order to assess the condition of the battery.For the development and testing of control units in the automotive industry, such as the BMS, the AUTOSAR standard was introduced, which separates application code from platform-specific software.By using AUTOSAR tools and the model exchange via the Functional Mock-up Interface (FMI), this paper shows how BMS algorithms can be validated and tested in several abstraction layers.A sub-function of the algorithm is tested first in the Modelica-based system simulation tool Dymola on a personal computer and then on Hardware-in-the-Loop (HiL) platform which emulates the hardware of an automotive ECU.In order to provide realistic inputs of the physical variables, a battery model in Modelica is built using the Dymola add-on Battery Library by Dassault Systèmes.In order to run on the HiL platform the battery model is implemented such that it is real-time compliant.For both, the BMS algorithm and the battery model, it is described along the process which adjustments need to be made when switching from the simulation framework to the HiL platform.