Abstract
KEYWORDS - friction surface, frictional heating, measurement, modelling, temperature prediction
ABSTRACT
Effective operation of autonomous/intelligent braking within the frame of autonomous driving systems depends not only on application of advanced control systems (including both hardware and software), but also on superior behavior and characteristics of every component of the braking system. In particular, this applies to friction wheel brakes whose performance depend on many influencing factors, among which brake interface temperature is the most important, because its influence on friction variations is much higher than that of other three influencing factors – brake by itself (its kind, design and characteristics), in addition to brake sliding speed and brake control force/pressure. Therefore, it is always a particular challenge to knowing the actual brake interface temperature values under any given brake operation conditions.
Measuring of the brake sliding surface temperature during brake pad life is a specific and very difficult task to perform due to variations in service conditions of brakes, wear appearance and presence of water, corrosion, and other immersive impacts, etc. Therefore, it seems that some kind of combination of measurement data with corresponding modeling approach might enable some kind of quasi “continuous monitoring” of brake interface temperature by its prediction. With this aim a number of passenger car disk brake samples tests were performed in the laboratory by means of a single-ended full-scale inertia dynamometer. Previously defined test program was applied composed of a number of different service modes which enable duplication of real service conditions being as close as possible. Brake temperature was measured during these tests by means of thermocouples placed at different locations in the depth of the brake pad, in addition to temperature sensor which was located „at the sliding surface itself“, i.e. 0,5 mm deep from the sliding surface in the friction pad. This specific individual temperature measurements serve only as control parameters necessary to assess quality of the model. Results of all other temperature measurements made inside the brake pad are used to design analytical relation between these temperature measurement values and those for the contact surface temperature measurements. Validation of such a mathematical model was indeed made by comparison of theoretical with the experimental results in order to enable experimental and analytical estimation of brake interface temperature to be combined so as to provide information on temperature values during any section or the entire service life of the brake pads. It may be demonstrated that the estimated interface temperature follows well the character of variation of the actual temperature at any measuring point, as well as the interface temperature. Estimated values of the brake interface temperature in any real time measurement sequence offer sufficiently high accuracy requested to be used in the prediction model, and thus in any potential application of it for management of braking process at the level of any individual brake application.