Reliability and durability-based design sensitivity analysis and optimization.

摘要

The objective of this research is to develop efficient and accurate DSA and optimization methods to support reliability and durability based design. As such, there are basically four tasks to be carried out:; First, an efficient method for component-level structural reliability analysis using the first-order reliability method (FORM) is presented. In this method, the continuum DSA and a two-point approximation (TPA) method are employed to search the most probable point (MPP) in the standard normal random variable space for structural reliability analysis.; Second, an efficient reliability-based structural DSA and optimization method is proposed. In this method, analytical reliability-based DSA is used to calculate the design sensitivity of reliability indices with respect to deterministic and probabilistic design variables. The active constraint set strategy is combined with the mean value first-order (MVFO) method and FORM to reduce the number of expensive reliability analysis steps. Furthermore, the TPA method is used to reduce the number of finite element evaluations required during design optimization.; Third, a DSA method for the fatigue life of 3-D solid structural components of mechanical systems with respect to shape design parameters is presented. The DSA method uses dynamic stress design sensitivity obtained using an analytical DSA approach to predict the dynamic stress increment due to design changes; and uses the finite difference between the new life and the original life at the same critical point to approximate the design sensitivity of the fatigue life.; Fourth, a new methodology for probabilistic crack initiation and propagation lives prediction is presented. The proposed method would allow engineers to evaluate reliability of fatigue lives of complex structures, considering uncertainties such as manufacturing tolerance, fatigue material properties, and initial crack size and direction. Furthermore, a mixed design approach is presented for design for probabilistic structural durability.