Powerfull Advanced N-Level Digitalization Architecture for models of electrified vehicles and their components

Informations projet

PANDA

N° de convention de subvention: 824256

DOI

10.3030/824256

Projet clôturé

Date de signature de la CE 23 Octobre 2018

Date de début 1 Decembre 2018

Date de fin 31 Mai 2022

Financé au titre de

• SOCIETAL CHALLENGES - Smart, Green And Integrated Transport

Coût total

€ 3 488 671,25

Contribution de l’UE

€ 3 488 671,25

Coordonné par

UNIVERSITE DE LILLE
France

Periodic Reporting for period 2 - PANDA (Powerfull Advanced N-Level Digitalization Architecture for models of electrified vehicles and their components)

Période du rapport: 2020-06-01 au 2022-05-31

Résumé du contexte et des objectifs généraux du projet

The automotive industry needs to adopt a new approach for developing cars, because the automotive market is undergoing disruptive changes. Whilst electrified vehicles represented only 0.1% of the market in 2015, the number of electrified vehicles sale has doubled from 2014 to 2015 (from 600,000 to 1.2 Million). EV sales increased nearly 200 percent between the second quarter 2020 and the second quarter 2021, and are expected to continue to grow massively in the coming decade. Traditionally manufacturers of internal combustion engines (ICEs) develop and assemble engines and transmissions (to a certain extent) independently form car manufacturers. The development is fundamentally different in electrified powertrains, as it is more complex to integrate all electrified systems in the vehicle design.

The PANDA project has made this fundamental change easier by developing a method to organise and interconnect models for all electrical vehicle components. A common framework based on the Energetic Macroscopic Representation (EMR) formalism has solved the problems of incompatibility between different models from different organisations, physical domains and levels of accuracy. PANDA is fully compatible with existing state-of-the-art methods such as Hardware-in-the-Loop (HiL) testing. HiL testing is used to test a real component by simulating the other vehicle components in a virtual environment (in real time). A dedicated interface is required to couple the real components with the virtual components. If accurate simulation models are developed, they can replace part of the experimental tests and thus reduce the development time.

The objectives were to:
• Develop an open organisation methodology for virtual & real testing of EVs
• Develop a multi-power open platform for Stand-Alone and Cloud-computing testing
• Perform virtual tests of reference electrified vehicles and real testing of selected subsystems

All work being completed, PANDA will play a leading role in the development of software tools and methods for improving the virtual generation of new products and new technologies.

Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent

At the start of the project, the rules for organizing the inputs and outputs of the models have been defined. A multi-level electro-thermal model of a battery has been developed, together with a dedicated battery pack. The model has been validated using (among others) this battery pack. A method to develop EMR-models using machine learning has been developed, together with EMR models of (permanent magnet) synchronous motors, and validated on a test bench.

In parallel, the commercial simulation environment Simcenter Amesim © has been extended with an EMR library of all relevant components (batteries, e-motors, etc). Cloud facilities have been set up, to enable cloud-based simulations. An EMR library has also been developed for a HiL environment, together with dedicated HiL equipment.

The methodology has been validated by simulating 3 references vehicles (BEV, FCV and p-HEV), and shows an accuracy of at least 95% with experiment test on the real vehicles. Moreover, different real subsystems of the P-HEV has been tested using HIL testing using real-time simulation with the cloud of models.

Finally, a framework has been developed for a forward life cycle assessment (LCA), that can be used to include all LCA aspects already during the design phase. The results show that the energy requirement – its source and demand - for cell and battery manufacture is an important parameter to further improve production impacts. Likewise, the battery lifetime significantly affects the impact.

Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent)

The partners have achieved all PANDA targets, and achieved a number of results beyond expectation. Based on the partners' experience with designing vehicles for mass production, a realistic scenario has been established for developing a completely new vehicle. When applying the knowledge from the PANDA industrial partners and the results from the PANDA project, the analysis shows that the lead-time of a new vehicle development can be reduced by up to 25%. The actual gain will vary for each use case, but the analysis shows that the potential benefit of using the PANDA approach can very high.

Please note that the 25% reduction in the lead-time of a new vehicle includes the extra time that is needed for training in the PANDA methodology. So whilst the additional education costs time, the benefits greatly outweigh the costs. However, as the PANDA methodology is a disruptive model organization, training programmes are needed to educate future engineers. Since 2000, University of Lille has developed training programmes at Master and PhD level. More and more universities worldwide are now teaching EMR. During the project, PANDA material has been presented to 250+ MSc and PhD students as part of regular academic teaching material and a Student version has been added to the Simcenter Amesim commercial platform, with an entry-level module for the EMR library.

The partners have reached a large audience worldwide, with many website visitors from e.g. the US and Indonesia, common workshops with related projects (attracting 80+ attendees from 10+ different countries for each workshop) and 200+ attendees at the PANDA final event. At the time of writing, the PANDA results have been presented at 22 international conferences, in 7 workshops, in 7 scientific articles and in 11 conference proceedings.

On top of that, the partners have achieved beyond expectation: (1) a spin-off company that will offer the PANDA methodology as (part of) a commercial service to companies, institutes and universities and (2) an evaluation group, to give 5 companies and universities a low-threshold opportunity to test the PANDA methodology on their use case. And finally, (3) the flexibility and portability of the PANDA method has been demonstrated. Firstly, an e-drive has been tested on a HiL setup (using reduced and full scale HiL) in which three different platforms were interacting seamlessly, remotely and in real-time (dSPACE, HiL Typhoon and Simcenter Amesim). Secondly, simulations in HiL Typhoon Control Center ©, with a direct compilation from Simecenter Amesim, and simulations in MATALAB Simulink ©.

In conclusion. based on the demonstrated cost and time savings, the PANDA integrated virtual development will support the complete generation of a new electrified vehicles.

homepage-concept-picture-2.png