Modular and Multiscale Model Integration

Our physics-based battery simulation software, BEST, provides in-depth insights into electrochemical processes and enables robust predictions based on physical models. With our latest enhancements, BEST now supports the Functional Mock-up Interface (FMI) standard, which is used by more than 250 commercial software products. This allows BEST to be integrated as a modular component in holistic multiscale simulations of battery-electric systems – such as Battery-Electric Vehicles (BEVs).

The major advantage of this approach is that, during system simulation, the detailed, spatially resolved physical states from BEST are available at every time step. This provides a comprehensive understanding of battery behavior within the overall system.

Our implementation runs in a container environment and exchanges data via network communication. This allows BEST to be seamlessly integrated into various simulation environments – even when running on remote servers or in the cloud.

Showcase: BEV Simulation with BEST

As a possible application example, we demonstrate here how BEST can be integrated with MATLAB^®/Simulink^® via the FMI standard to predict the system behavior of a BEV under realistic operating conditions.

To do this, we implement a simple energy balance model of a BEV in MATLAB^®/Simulink^®with four blocks for the key components:

Powertrain: Based on a time-dependent target speed, this block calculates the power requirement for vehicle acceleration using the vehicle parameters.
Battery: The battery block provides the required power using the underlying cell model from BEST. For a given power requirement, BEST calculates the battery’s electrical and thermal response.
Heating, Ventilation, and Air Conditioning (HVAC): This block models the thermal system. It keeps the components within specified temperature limits and acts as an additional electrical load.
Driver’s Cabin: Climate control of the vehicle interior is another major energy consumer. It is coupled with the HVAC system and is influenced by the ambient temperature.

© Fraunhofer ITWM
BEV simulation block diagram in MATLAB®/Simulink®

Integration of Best Into the System Model and Driving Scenario

While the powertrain, HVAC, and driver's cabin are modeled directly in MATLAB^®/Simulink^®, the Digital Twin of the battery cell is created by embedding BEST into a Functional Mockup Unit (FMU) that complies with the FMI standard. Simulink^® continuously transmits electrical and thermal boundary conditions to BEST and, in return, receives simulation results for the battery, such as cell voltage, state of charge, and heat generation.

For our application example, we consider a driving scenario with three phases:

Standardized driving cycle in accordance with the Worldwide Harmonized Light Vehicles Test Cycle (WLTC) for Class 3 vehicles
Highway driving at a constant high speed
Charging the vehicle

Throughout the entire simulation, BEST saves its regular simulation results. This allows for a detailed analysis of internal battery processes within the context of system dynamics.

© Fraunhofer ITWM
Zeitliche Systemdynamik aus der BEV-Simulation: (a) Geschwindigkeitsprofil und Ladezustand (SoC), (b) Zellspannung und minimales Lithium-Plating-Potenzial, (c) Temperatur verschiedener Systemkomponenten und (d) thermische Leistungen der Batterie and des HVAC-Systems.

Räumlich aufgelöste Lithium-Ionen-Konzentration im Elektrolyten entlang der Dickenrichtung, generiert mit BEST.. — © Fraunhofer ITWM
Räumlich aufgelöste Lithium-Ionen-Konzentration im Elektrolyten entlang der Dickenrichtung, generiert mit BEST. Die verschieden farbigen Graphen wurden zu unterschiedlichen Zeitpunkten aufgenommen, siehe vertikale Linien im Geschwindigkeitsplot der zeitlichen Systemdynamik.

Virtual Measurement Campaigns for BEVs

Early versions of this BEST coupling interface were already being used in conjunction with our Virtual Measurement Campaign (VMC^®) simulation software to simulate battery electric vehicles. This approach combines:

the extensive, georeferenced VMC^® database of real-world environmental data
an efficient vehicle model with VMC^® Simulation
BEST’s high-resolution electrochemical and thermal battery models

As a result virtual test drives with BEVs can be performed significantly faster than in real time. This allows for rapid evaluation of vehicle and battery concepts across a wide range of realistic usage scenarios.