Reliability analysis of the truck frame FE model considering the statistically distributed design variables

Abstract

Reliability analysis using the Pearson System considering the variance of tolerance for the design variables of the truck frame structure is performed in this paper. First, the fatigue life of the structure is estimated by three steps. Multibody dynamic model of 9-DOF vehicle to calculate constraint forces applied on the frame is simulated. Then the quasi-static analysis of the truck frame to calculate the stress coefficient of structure is performed. Finally using the stress histories based on the linear combination of constraint forces and stress coefficients, fatigue analysis to calculate the fatigue life is done. It is assumed that the design variables related to the multibody model such as mass, spring and damping coefficients and the dimension of the section in FE model are statistically distributed. By employing the Pearson System, the reliability analysis of the structure based on the statistical moment is performed. Then the variance of the probability density function considering the tolerance size of the design variables is estimated to analyze the effect of the design variable on the variance of fatigue failure probability. From these procedures, most sensitive design variables are known and the proper tolerance size for each design variables could be determined compared to the required reliability of structure.