The project reduces the model calibration and validation time by 60%, enabling more efficient electric vehicles.
The HiFi-ELEMENTS project is coming to an end after more than three years of research and testing in which CIDETEC Energy Storage has contributed to the development of a battery cell model that was used in the simulation of a full vehicle and has proposed a modelling methodology that reduces the time and means to calibrate and validate this model. The project focused on the creation of a standard system architecture for Battery Electric Vehicle (BEV) development activities. This will reduce and facilitate the total development and calibration effort, which will accelerate the introduction of electric vehicles to the market.
During the project, four different cases were analysed in which the developed models were tested for verification and validation. The diversity of the four selected cases allows the conclusions to contribute to the improvement of the total powertrain system of the electric vehicles of the future.
Cases analysed in HiFi_ELEMENTS:
Replacement of the electric motor:
This case focused on the efficient integration of the electric motor of a BEV on the xMOD co-simulation platform, which was successfully verified. Variants were made between radial and axial flux electric motor models. The total effort for the development of real-time simulation component models for HiL (Hardware in the Loop) applications was reduced by 71%, while the total effort for model calibration was reduced by 60%.
Battery development (relevant participation of CIDETEC Energy Storage):
The validity of the battery virtualisation was tested using comprehensive testing and validation trials. For this, the process was carried out step by step in order to cover all stages of battery development: cell selection, module assembly, battery concept design, battery testing and calibration. Each of these steps was developed with its own virtualisation-simulation. For the battery testing procedure, two approaches were carried out and compared at the cell (Cell in the Loop) and module (Module in the Loop) level. This analysis was used to verify that the high-fidelity battery model was correctly calibrated and met the design objectives.
In this particular case, CIDETEC led the cell characterisation and modelling methodology as well as the Cell in the Loop activity, where different BMS functions and the behaviour of the cell model were tested.
Figure 1 – Conceptualisation of virtual cell testing (Cell in the Loop)
Figure 2 – Cell testing in climatic chamber
Co-optimisation of components:
Another objective of HiFi-ELEMENTS was to demonstrate the rapid integration and interchangeability of models, between different suppliers and levels of fidelity, thanks to the standardisation of the modelling process and the built-in co-simulation tools. This objective was achieved, under the requirement of reducing consumption, using 2 different powertrain architectures and multiple combinations and variants among their components, using the co-simulation environment proposed by Hi-Fi.
Electric powertrain concepts
A conceptual design of an electric vehicle considering a wide range of electric powertrain concepts has demonstrated that the use of real-time reusable models (higher complexity) in the conceptual phase can aid decision making, thus contributing to greater efficiency in the final product. The electric propulsion concepts with higher power offered better operational efficiency compared to the concept with only the necessary rated power for basic operation. In the case of this electric motor, a key reason for improving operating efficiency with “overpower” was the increase in base speed and, therefore, a reduced possibility of inefficiency with weakened field. In addition, it has been shown that centralising the system’s power on one axis, as opposed to two axes, also improves the final efficiency. On the other hand, the motor concept implemented on the wheels compromises energy efficiency, so it has been concluded that its strengths should be found in other domains.