The UPSCALE project received EUR 4 million from the Horizon 2020 programme (grant agreement no. 824306) with the aim of improving electric vehicle development and increasing performance.
The consortium has been working for the past four years on the application of artificial intelligence methods to reduce development time and increase the performance of electric vehicles required by the automotive industry to reduce global emission levels.
Barcelona, 30 April 2022. The UPSCALE Project (Upscaling Product development Simulation Capabilities exploiting Artificial intelligence for Electrified vehicles) was born, under the framework of Horizon 2020 (grant agreement no. 824306), with the aim of replacing the most expensive automotive simulation cases with data-driven computational methods. To this end, the most difficult simulation load cases were identified, and alternative methodologies based on machine learning techniques were designed and fully implemented.
The project has been developed by a consortium of 11 European companies and organisations from the automotive, software, engineering and research sectors.
The consortium is led by the Spanish company Applus+ IDIADA (Spain) and among the entities that form part of this consortium are Algorithmica Technologies GmbH (Germany), Centro Ricerche Fiat SCPA (Italy), École Nationale Supérieure d’arts et métiers (France), ENGYS Ltd (UK), ESI Group (France), FI Group (Spain), Kompetenzzentrum- Das Virtuelle Fahrzeug Forschungsgesellschaft mbH (Austria), Volkswagen AG (Germany), Volvo Personvagnar AB (Sweden) y Vrije Universiteit Brussel (Belgium).
Four years of project:
Since 2018, UPSCALE has sought to reduce the time it takes to develop electric vehicles, as well as in increasing specific performance. Enric Aramburu, Fluid Engineering Product Manager at Applus+ IDIADA, leader of the project, explains that "the UPSCALE project will save 20% in the development period of electric vehicles by using artificial intelligence and machine learning applied to simulation in crash, aerodynamic and battery tests".
He adds that "UPSCALE has developed machine learning tools to replace finite element stacks with reduced order models for structural calculation, as well as for short-cut risk calculation, and full wind tunnel simulations using aerodynamic ROMS that provide the aerodynamic forces and airflow around the vehicle in real time."
The European Union, with the funding to the UPSCALE project through the Horizon 2020 programme, seeks to foster the electric vehicle industry in Europe and at the same time, promote compliance with new laws and pollutant emission standards in the region. "This is something that the European Union has been promoting a lot in recent years, it aims to be a climate neutral continent by 2050, so this type of project is essential to pursue that goal," says Cristina Nappi, Senior European Consultant at FI Group.
Results:
In these four years, the UPSCALE project has demonstrated the feasibility of replacing FEM and CFD solutions with ML algorithms through a successful proof of concept and paves the way for further deployment of AI in vehicle development programmes.
