InLab FIB collaboration with SEAT and Volkswagen Group Innovation

The UPC's inLab FIB collaborates with SEAT and the Volkswagen Group Innovation (Wolfsburg, Germany) in the 'Predictive eBoost' project, focusing on designing new strategies, based on machine learning algorithms and data analysis, to improve the efficiency and performance of electric vehicle engines and batteries.

Battery in electric vehicles are becoming increasingly powerful and more range, but they also need proper thermal management. Current strategies only consider the temperature of the vehicle components to decide when to cool them or heat them. The smarter development of decision-making support systems, such as machine learning models, will allows for better efficiency and consumer rules.

The focus of this project is to use vehicle route information, such as gradient of roads or circulation speed, to decide when to activate battery cooling to improve current strategies.

Why should a vehicle spend energy in cooling the battery, if in a few minutes it will find a long slope in which it can be cooled by itslef? If it could be known that an important acceleration is approaching, would it not be better to prepare the vehicle in advance to mitigate the effort it will have to do? These, among others, are some of the questions that are being asked in this project.

Considering not only the temperature of the vehicle components, but also their state, environment and route, allows optimizing the performance and the battery consumption in order to increase its autonomy and useful life, and minimise the environmental impact. The results of the 'Predictive eBoost' project will represent an important step forward in the the electric vehicle implementation, as they will allow to extend the distance travelled without stopping, while providing a satisfactory driving experience.

The research is headed by Dr. Gerhard Lux of SEAT, and Marc Duevel of Volkswagen Group Innovation, and is led by the Professor Ernest Teniente, director of inLab FIB, the Innovation and Research Laboratory of Facultat d’Informàtica de Barcelona (FIB) in Universitat Politècnica de Catalunya · BarcelonaTech (UPC).

Source: inLab FIB