Sensitivity analysis of subjective and objective reliability of disc brakes based on the parallel efficient global optimization algorithm

The subjective and objective reliability analysis method allows for a comprehensive consideration of both subjective and objective uncertainties in the braking process. However, challenges arise during the optimization of implicit problems in complex structures, such as excessive iterations and prolonged waiting times for finite element solutions. To solve these problems, a new method, the subjective and objective reliability sensitivity analysis method based on the parallel efficient global optimization algorithm (PEGO-SORSA), is proposed. Subsequently, the PEGO theory is integrated with the adaptive kriging method to predict subjective and objective reliability surrogate models. Finally, subjective and objective parameter intervals and reliability sensitivity rankings that significantly influence the plausibility of the brake disc's structural failure are obtained. In addition, the PEGO algorithm is used to identify the optimum within the given interval, reducing the waiting time for the finite element solution by about 41.3% .