Reliability-Based Design Optimization of Full-Floating Automobile Semiaxle Using Evidence Theory

A method of reliability-based design optimization is presented. Uncertainties such random variable and fuzzy variable are remodeled as the formulism of evidence theory. By means of the combination of evidence, the upper and lower bounds of reliability are obtained, and then a substituting model of constraint conditions of reliability is proposed based on the obtained bounds of reliability. As an example, the reliability-based design optimization of full-floating automobile semiaxle is modeled by using the proposed substituting model of reliability constraint. Under the constraint conditions of allowable reliability of torsional strength and torsional rigidity, the optimized solution is achieved to minimize the weight of the automobile semiaxle. The example shows that the proposed method is practical and effective.