Contact: Dr. Alexander Thaler
Dieser Themenkreis beschäftigt sich intensiv mit Methoden zur modellbasierten und automatischen Testfallgenerierung für System- und Diagnosefunktionen unter Einbeziehung von Hardware-Komponenten. Zudem werden Methoden zur Modellverifikation und optimale Informationsgewinnung durch Sensordatenfusion behandelt.
The electrification of today's automobiles is driven by different goals, such as the reduction of CO2 emissions and a suitable range in electric driving mode. The energy storage system is one of the key factors for achieving these goals. A detailed modeling of the electrochemical system is essential for optimizing its utilization. By combining disciplines and creating a unique research network of renowned industry and scientific partners, VIRTUAL VEHICLE offers a central platform for uniting the aspects and elements necessary for future mobility. For example, its "Vehicle Electrics, Electronics and Software" Area develops novel approaches, methods and processes for model-based design, simulation and optimization, as well as system integration and validation of complex E/E systems in conventional, electric and hybrid vehicles.
- Cell deformation simulation
- Crash and crush inspections
- Functional safety (ISO26262)
- Vehicle safety
- Electrochemical (mechanistic) modeling
- Accelerated ageing
- Contamination effects simulation
- Pre & post-mortem analysis
- Thermal integration
- 1D and 3D simulation
- Real-time modeling
- Cooling and heating strategies
- Energy management
- Coupling of electrochemical, thermal and mechanical models
- Battery management strategies
- Robust SoX estimation
- Optimal design of cells
Level 1: Modeling and simulation of electro-chemical properties
From fast voltage and current simulation models, based on impedance spectroscopy and time domain data, to highly sophisticated PDE-based models that can simulate ageing, a range of different models are being deployed and constantly upgraded. Basic chemical energy storage properties can be analyzed chemically and physically to parameterise the different models.
Level 2: Thermal and mechanical cell modeling in 3D
The thermal and mechanical validation and verification of different cell designs are key factors for life cycle assurance. By supplying the inputs for an optimized cell integration process, full thermal and mechanical cell models are developed.
Level 3: Module layout and battery operation strategy
Taking battery life limiting factors into account, battery load strategies for advanced hybrid and electric vehicle controls are optimized. Module layout and design are evaluated based on geometric constraints, vehicle thermal management, load strategies, user requirements and level 2 inputs.
Level 4: Misuse and crash simulation
Combining electrochemical knowledge with mechanical expertise, a model-based verification from cell up to system-level design is performed. Validated FE models enable the assessment of the quality and security risks of battery packs in crash situations.
HYBCONS: intelligent control software for hybrid vehicles
The development of hybrid vehicle technology is most actual and an important task of modern automotive industry. The intelligent control of such vehicles is a key factor in this development. The variety of currently developed hybrid vehicles, regarding drive train topology, used components (electric motors, batteries,... ), applications (passenger car, taxi, bus, commercial vehicle etc. ) and markets is enormous. This variety is reflected in many different variations of the hybrid software architecture and its implementation. A customized software solution for each new hybrid version, however, will lead to an enormous increase in development costs and highly limited software reusability. The goal of the current HYBCONS research project (Generic Hybrid Vehicle Control Software) is to develop and implement a re-usable and scalable universal hybrid control software (generic hybrid software) for mild and full hybrid vehicles.
One of the most critical points is the systematic, data-consistent development of the hybrid software – from determining the requirements, the functional design and the implementation of embedded systems, to system integration, verification and validation on the vehicle level. One particular emphasis is on the scalability of the software for different hybrid vehicles and platform versions and the maintainability throughout the entire product life cycle.
The project team of the HYBCONS K2-research project faces the following challenges:
- Universal approach: defining a universal approach for the specification of the many different functionalities of a hybrid vehicle and the corresponding control strategy
- Scalable / expandable software: The software architecture must support a scalable and reusable approach for the implementation of the control functions of the different hybrid topologies and variants. On the other hand, the software must be easily expandable to include new hybrid functions and control algorithms.
- Easy configuration: An easy, customizable configuration of the system should be additionally possible. Mass-production functionality: The software must fulfill the "production-specific" requirements: a) it should contain functions and/or the respective interfaces for the diagnosis, process monitoring, cases of misuse, 'limp-home' operation etc., b) it should be capable to be integrated with commercially available control units and c) it should be easy to calibrate.
- Efficient integration: One additional challenge is the integration of the hybrid control software into a distributed embedded system (a network of control units, such as engine – battery – electric motor- control units) with limited run time and storage resources so that the resulting control application can remain independent from the hardware and support the efficient integration for various types of hardware platforms.
In order to successfully deal with these challenges, it is necessary to guarantee a seamless development process and to support it with the modern methods and processes.
Initially, the functional and non-functional requirements for hybrid control are created, whereby a generalized description of the hybrid functions is used. The necessary prerequisites for the derivation of a generic software architecture include the functional requirements, the analysis of co-simulation of the entire embedded hardware/middleware/software and the principles of the "Software Product Lines," which are developed in cooperation with the Institute for Technical Informatics at the Graz Technical University.
Eventually, the control software is implemented as "Proof of Concept" for a mild-hybrid with a so-called Hybrid-P2-Topology as basic hybrid system. In addition to it, the software is augmented with additional functions for diagnosis, process monitoring, cases of misuse, limp-home operation etc., integrated in a standard control unit and verified and validated in MiL-, SiL- and HiL- configurations. Particularly for this part of the project, the support and close cooperation with the industrial partner AVL List GmbH is indispensible, since this partner contributes the hybrid vehicle, including the control units and the test tools (e.g. AVL InMotion, AVL Cruise).
In order to investigate the scalability, the existing software is enhanced with the functions for a full hybrid. Except for the pure electric driving function, the full hybrid poses much higher demands on the control in terms of energy management and start-and-stop operation.