Interaction of stress-raising features in aerostructures
It is widely accepted that fatigue is responsible for 90% of all unscheduled component failures, and is therefore a key research area for the aerospace industry. The basis of successful design against fatigue relies on the accurate and consistent determination of stress concentration factors around stress-raising features. Within aerostructures, design problems often occur where two or more geometric features are placed in close proximity such that their associated stress-fields interact. It is difficult to assess potential susceptibility to fatigue failure in such situations as no universally-accepted method has been available by which a design engineer could estimate accurately the stress levels produced. This research has focused on the development of standardised evaluation techniques by which fatigue analysts can predict the stress concentration factors associated with problems involving interacting stress-raising features. These methods have been designed as flexible and accurate predictive tools to be incorporated into the fatigue analysis process as time-efficient alternatives, or pre-cursors to the application of techniques such as finite element analysis. Fundamental research was carried out into the understanding of stress concentration factors associated with a series of commonly occurring geometries of interacting features. The approaches employed for the development and validation of the methods have included numerical solutions (using the finite element method) and experimental studies (using photoelastic analysis and strain gauge measurements). Through these research techniques, new insights have been gained into the manner in which two or more stress raisers interact when placed in proximity. Based on these understandings, two categories of evaluation tool have been developed: (i) a generic methodology for those stress-raisers that lie on a common cross-section, and (ii) individually-tailored prediction techniques for those features of alternative alignment. In all cases, these techniques were found to provide highly accurate stress concentration predictions, as compared to appropriate analytical and numerical treatments. These methodologies are currently being implemented as standard practice within the fatigue analysis process at BAE SYSTEMS through incorporation into the SCONES knowledge-based system (Robinson et al. 2001) and the BAE SYSTEMS Technical Standards Manual for stress concentration analysis (BAE SYSTEMS 2001 (2)).