Abstract: To tackle the problem of reliability decline and braking noise caused by the wear of disc brake, an analysis / optimization method for the timevarying stability of vehicle disc brake system is proposed. With the method, the random process parameters are introduced to describe the brake structure with wear during operation. A parametric model covering random parameters and random process parameters for the brake system with uncertainty is constructed to perform the dynamic prediction of stable reliability during brake operation. On this basis and aiming at the reliability decline in the later stage of brake operation, an optimization on the timevarying stability of brake system is conducted by using the analysis theory for timevarying reliability, response surface technique and genetic algorithm, with minimizing the negative complex eigenvalue of damping ratio as objective, to ensure its reliability meeting requirements in the later stage of operation. Finally the method is effectively applied to the prediction on the timevarying reliability of a brake with floating caliper as an example, with its stable reliability in the later period of operation effectively enhanced by an optimization on the density and thickness of supporting back plate of the brake.

Key words: disk brake, random parameters, time-varying parameters, brake noise, stability, reliability optimization