In-Vehicle Brake System Temperature Model

Abstract

The brakes system is critical with respect to vehicle safety. One situation during which the brake system is put to the test is an Alpine descent. Such a descent causes very high brake system temperatures and may even induce brake fluid vaporization. In following report an In-Vehicle Brake System Temperature Model is developed and tested. This model makes use of the information that is available on the vehicle CAN-bus in order to estimate the temperature of the brake system and detect the risk of brake fluid vaporization. Implementing such a model in production vehicles would improve vehicle safety and in the long run allow downsizing of the brake system without giving in on any safety margins. Firstly, possible approaches for estimating the amount of kinetic energy that is converted into heat by the brake system are investigated. A Feed-Forward model is developed that uses the pressure in the primary hydraulic circuit as input. The problem with such a Feed-Forward Model is that the friction coefficient is variable. A Feed-Back model that uses the vehicle deceleration as input gives better results. The challenge in this approach lies in the fact that not all required inputs for the Feed-Back Model are known. Secondly, the Temperature Estimation Model is developed. This is composed of models of different parts of the brake system which are combined and matched to the measurements. The brake disc is modeled as a lumped mass. The brake pad is modeled as a Finite Difference Thermal Model. The fluid and surrounding caliper are modeled as one lumped mass which receives its heat through conduction from the brake pad and spreads the heat into the surroundings by means of convection. Finally, for the Human Machine Interface, several active and passive intervention concepts are proposed. One important challenge to take into account is the uncertainty in the brake fluid boiling temperature, which decreases over the course of time due to brake hygroscopic effects.