Multiphysical MF-based tyre modelling and parametrisation for vehicle setup and control strategies optimisation. Issue 10 (3rd October 2022)

• Record Type: Journal Article
• Authors: Sakhnevych, Aleksandr
• Is Part Of: Vehicle system dynamics. Volume 60:Issue 10(2022)
• Page Start: 3462
• Page End: 3483
• Publication Date: 2022-10-03
• Subjects: Multiphysical tyre -- tyre data analysis -- nonlinear system modelling -- multiphysical parametrisation -- thermodynamic and wear analysis; Motor vehicles -- Dynamics -- Periodicals; Electronic journals; 629.231
• DOI: 10.1080/00423114.2021.1977833
• Languages: English
• ISSNs: 0042-3114

Abstract

ABSTRACT: Starting from the earliest phases of design of the vehicle and its control systems, the understanding of tyres is of fundamental importance to govern the overall vehicle dynamics. A properly characterised tyre–road interaction model is essential to achieve a reliable vehicle dynamics model on which more design variations can be studied directly in simulation environment optimising both cost and time. The possibility to count on computationally efficient and reliable formulations represents nowadays a great advantage, and the multiphysical Pacejka's Magic Formula (MF-evo) tyre model presented is one of the best trade-off solutions to meet the strict real-time requirements and to reproduce multiphysical variations of the tyre dynamic behaviour towards temperature, pressure and wear effects. A specific methodology has been developed to characterise and to identify the MF-evo parameters with a high grade of accuracy and reliability directly from experimental data. The proposed technique is based on a pre-processing procedure to remove non-physical outliers and to cluster the data, which allows to optimise the multidimensional parameterisation process. To the purpose of validation of the parametrisation routine, data from a motorsport case, exceptionally difficult to reproduce in simulation due particularly significant variations of the tyre dynamics during a single test, have been employed demonstrating the MF-evo model potential and robustness.