Upscaling Product development Simulation Capabilities exploiting Artificial inteLligence for Electrified vehicles

Livrables

Website containing all the publishable information related to the project. This deliverable belongs to Task 7.2. In the deliverable the main website functions will be described

Corresponding load cases for the full vehicle models to be used within the project

Definition of the load cases for the previous specified vehicles models. This deliverable will be obtained in task 5.1.

Report on dissemination to end-user and stakeholders

Description of the dissemination activities performed to endusers and stakeholders will be detailed in this deliverable

Reduced order models for aerodynamic performance prediction

This delievrable will describe the steps taken to define the reduced order models that will be obtained when the task2.2 is completed.

Final report containing proposal for further use of the new methods

Full report and conclusions to provide the necessary information for the future application of the new method This deliverable is a summary of the work performed in WP5 and belongs to T54

Report on methodological approach for battery risk analysis in severe crash scenarios

This delivearble contains the description of the conclusions obtained after the performance of Task 52 related to crash impact on batteries

Requirements for setting up a reduced order model of a battery to be used in a full vehicle crash simulation

List of the requirements for the ROM of the battery that will be used in the full vehicle crash simulation. This deliverable will result from T5.1.

Requirements for setting up a reduced order model of a full vehicle model with parametrized boundary conditions

List of the requirements to define the ROM of the full vehicle according to the desired accuracy. This deliverable belongs to T5.1.

Requirements for setting up an AI model to improve parallelization of the solver

Report of the requirements for the AI model so that the solver can be parallelized his deliverable belongs to Task 53

Requirements for aerothermal simulations reduced order model

"This report will list the requirements for the variables necessary to perform the ""offline"" phase to generate the AI/ROM models. This is in terms of which variables, the size of the dataset and the characteristics to obtain the desired accuracy in every case. This deliverable results from Task 2.1. "

Potential for ML-based acceleration in finite volume

Report of the solver algorithm review and further acceleration potential assessment, focusing on the pressure correction step for the acceleration efforts. This deliverable is a result of ST1.1.1.

Assessment of reduced order models for aerodynamic performance prediction

Report on the validation process of the previously defined reduced order models used for higher fidelity data. This deliverable results from Task 2.3.

Report with the compiled requests for publications

Report listing the compiled requests for publications published during the project period This deliverable belongs to Task 73

Validated tool to handle and rationalize aerodynamic data from heterogeneous sources

This report will describe the tool implemented for dealing with all types of aerodynamic data coming from different sources and in different formats. This deliverable results from Task 2.4.

Assessment of AI/ROM based optimization performances with respect to state-of-the-art methodologies

This report will contain the comparison of the new AIROM based technique with the stateofthetechnologies to determine if it results into a better alternative This deliverable results from Task 25

Verification of the optimized model with high fidelity simulations for a fully electric SUV/city car and final report on the framework performance.

this deliverable describes a report of verification of results with higher fidelity simulations for the specified vehicles and report describing the full framework This deliverable results from Task 45

Publications

Application of Physics Informed Machine Learning model for correcting RANS modeled Reynolds Stress Anisotropy

Auteurs: Bhanu Prakash, Charalampos Tsimis, Enric Aramburu
Publié dans: 2020
Éditeur: IDIADA

Multi-fidelity data fusion through parameter space reduction with applications to automotive engineering

Auteurs: Francesco Romor, Marco Tezzele, Markus Mrosek, Carsten Othmer, Gianluigi Rozza
Publié dans: 2022
Éditeur: arxiv
DOI: 10.48550/arxiv.2110.14396

Advanced Digitalization for Development of All Types of Electrified Vehicles and Components

Auteurs: Alain Bouscayrol; Valentin Ivanov; Reinhard Tatschl; Enric Aramburu
Publié dans: 2021 IEEE Vehicle Power and Propulsion Conference (VPPC), 2022
Éditeur: IEEE
DOI: 10.1109/vppc53923.2021.9699125

AI Enhanced Methods for Virtual Prediction of Short Circuit in Full Vehicle Crash Scenarios

Auteurs: Alexandre Dumon, Michael Andres, Stefano Menegazzi, Christoph Breitfuss, Cristian Jimenez, Francisco Chinesta, Fatima Daim, Alain Tramecon
Publié dans: SAE Technical Paper Series, 2020
Éditeur: SAE International
DOI: 10.4271/2020-01-0950

Livrables

Publications

Partager cette page

Télécharger