Reduction in Computational Cost of Progressive Failure Analysis

Abstract

Designing aircraft structures requires efficient modeling approaches to iterate on multiple structural configurations to achieve an optimal design. Typically, damage tolerance is not considered at the design stage because of the high computational cost in its implementation within a finite element modeling approach. Therefore, analytical or empirical approaches are often used to size critical damage-tolerant structures once an optimal design is determined. In this study, the Progressive Release eXplicit Virtual Crack-Closure Technique (PRX-VCCT) is assessed for its capability to costeffectively evaluate skin-stringer separation of a blade-stiffened panel that is subjected to seven-point bend loads. An initial verification study was performed to evaluate PRX-VCCT to accurately simulate skin-stringer separation with respect to existing cohesive element approaches. Furthermore, the influence of element size, ranging from 0.10 in. to 0.40 in., on the total computational time using the PRX-VCCT is investigated. The results indicate that the PRX-VCCT can be used to accurately simulate skin-stringer separation using large element lengths (0.40 in.). Additionally, a significant reduction in the computational time to simulate skin-stringer separation is observed using the PRX-VCCT. Large-scale progressive damage analysis using PRXVCCT can be implemented early in the design cycle of composite structures without requiring a global-to-local modeling approach.