An adaptive algorithm to accelerate the critical plane identification for multiaxial fatigue criteria

authored by

M. Wentingmann, P. Noever-Castelos, C. Balzani

Abstract

For the fatigue analysis of structures undergoing non-proportional stress histories, the critical plane approach has proven a physically meaningful and thus comprehensive method. However, procedures that accurately identify the critical plane
are computationally very costly. In order to reduce computation times and to make full use of the critical plane approach an adaptive algorithm for the identification of the critical plane is presented in this work. The algorithm is based on the segmentation of a half sphere in segments of equal surface areas. Starting with a coarse mesh the algorithm refines only those segments that probably include the actual critical plane. This simple yet very effective approach refers only to the accumulated damages of the segments and is hence suitable for every critical plane failure criterion. Depending on the discretisation level and used failure criterion reductions of up to 82 % in computational time can be expected without loss of accuracy, which is demonstrated by a fatigue analysis of a wind turbine’s trailing edge adhesive joint.

Organisation(s)

Institute of Wind Energy Systems

Type

Conference contribution

Pages

3745-3754

No. of pages

10

Publication date

2018

Publication status

Published

ASJC Scopus subject areas

Mechanics of Materials, Mechanical Engineering, Computer Science Applications, Computational Theory and Mathematics

Sustainable Development Goals

SDG 7 - Affordable and Clean Energy