Enhancing Meta Model of the Brake Pad Friction Coefficient using the Explainable Machine Learning

As increasing system complexity and various customer demands result in the need for highly efficient vehicle development processes. Once the brake torque is predicted accurately during the driving scenario in the earlier stage, it will be able to prevent the changing the vehicle or brake system design to satisfy the legal regulation and customer requirement. As brake torque performance target allocate brake pad friction coefficient level and characteristic, the accurate friction coefficient prediction should be preceded for accurate prediction for brake torque. Generally, the friction coefficient of the brake pad is known to vary nonlinearly depending on the physical properties of the disc and the pad, as well as the rotational speed of the brake disc, the disc temperature, and the hydraulic pressure. Furthermore, it varies depending on the driving scenario even the same condition. Therefore, there was a need for a new method : machine learning to solve this problem. A Mixed Effect Random Forest (MERF) model was applied to reflect the fixed effect that independent variables irrespective of the object (scenario) and the random effect that different influences depending on the object to the dependent variable (coefficient of friction). Then, derivation variables that significantly affect the change in friction coefficient were explored from the raw data collected through the dynamo test. The performance test result of the MERF model trained with the derivation variable that have been discovered above as input is as follows. The overall prediction performance achieved the target MAE <0.01. In addition, it was confirmed that the offset, discontinuity, and noise vulnerability of the prediction results improved compared to previous studies. It is also possible that the trained model is integrated to total vehicle simulation model and predict the braking performance with reflecting the varying of the coefficient of friction according to the driving scenario.