Brake Heat Capacity Prediction by Machine Learned Friction Coefficient Model and Virtual Wheel Brake Co-Simulation

2024-01-3054

To be published on 09/08/2024

Event
Brake Colloquium & Exhibition - 42nd Annual
Authors Abstract
Content
In recent times, the increasing complexity of systems and diverse customer demands have necessitated highly efficient vehicle development processes. Accurate prediction of vehicle performance through simulation enables the determination of design specifications before building test vehicles, leading to reduced development schedules and costs. Especially with the transition to electrification in the automotive industry and the diversification of mobility businesses, the rear axle load for xEV has increased compared to conventional ICE vehicles. Detailed brake thermal performance predictions are now required not only for the front brake but also for the rear brake. Moreover, scenarios requiring verification, such as alpine, which applies braking severity to xEV with the regenerative braking system, have become more diverse. In this study, a co-simulation method with machine-learned friction coefficient prediction models is proposed to enhance the accuracy of brake thermal capacity predictions in the vehicle simulation environment. And this simulation method features the simultaneous prediction of both front and rear wheel brakes. The required brake torque for the front and rear wheels is calculated from the vehicle model and driving scenarios. The brake system model generates the necessary pressure during deceleration, and the friction coefficient acts to create brake torque, resulting in brake power for both front and rear brakes. Virtual wheel brake in simulation model calculates speed, pressure and disc temperature based on vehicle driving schedules, while the machine learned model takes these variables as inputs and returns friction coefficient. The prediction accuracy of torque and disc temperature improved significantly as the virtual wheel brake utilized the received friction coefficient from the model trained with MERF algorithm.
Meta TagsDetails
Citation
Cho, S., Baek, S., Kim, M., Hong, I. et al., "Brake Heat Capacity Prediction by Machine Learned Friction Coefficient Model and Virtual Wheel Brake Co-Simulation," SAE Technical Paper 2024-01-3054, 2024, .
Additional Details
Publisher
Published
To be published on Sep 8, 2024
Product Code
2024-01-3054
Content Type
Technical Paper
Language
English