Structural fatigue life calculation methodology for random loading

Abstract

Fatigue life calculation has become a mandatory step while designing structural components in any type of industry. Some of the loading conditions for these components have a random behavior to some extent, and even though a simplifi cation to static loads is not the most accurate way, it is one of the most used methodologies given the direct application of the fatigue theories and the knowledge and experience of the engineers. There are a couple of reported approaches to consider the randomness of the load and come up with an expected value for fatigue life, such as Steinberg, Lalanne, Tunna, Dirlik, etcetera; however, these procedures are not familiar and straightforward for most of the structural engineers in the industry. There are commercial software packages available that can help close the knowledge gap, but those are expensive and require additional training for the engineers. This thesis work defi nes a straightforward methodology and develop a cost and training free application that considers load randomness to calculate fatigue life for structural components. To achieve this, a review of the available approaches was done, and by selecting the most recommended techniques throughout literature, an application was developed in Matlab to perform all the necessary calculations, whose GUI is simple enough to be included as part of the virtual validation procedures. With this methodology and application, calculating fatigue life for random loading conditions is simpli fied, making it available for every engineer nevertheless of their fatigue experience and knowledge. This document guides the reader through a general understanding on what is a random load, the mathematical and physical concepts on the fields of mechanical vibrations and fatigue theory necessary to understand random vibration and how it connects with fatigue, a proposed methodology with a series of steps to follow to calculate random loading fatigue, and finally a demonstration of the methodology applied to a specimen and how the calculations correlate successfully with testing.