Structural integrity monitoring of bonded composite patch repairs of aging airframes using wavelet transforms

In recent years there has been a considerable interest in the design and application of bonded composite repairs to restore cracked or corroded metallic airframes. The long-term durability in terms of disbond of composite patch, fracture at the bond-lines, delaminations in composite patches, and structural crack growth is to be thoroughly investigated before the aerospace industry adopts composite repair technology. Many researchers have investigated the use of structural modal response (natural frequencies and mode shapes) for assessing the integrity of structural systems. Although these methods have been successfully employed in some practical systems, there still exist some problems for applications in bonded repair systems such as the high sensitivity of these systems to model and measurement errors. In this paper wavelet maps and neural networks are used to detect and quantify damage in composite bonded repair systems. Two algorithms are presented, one for the location of damage in both single and multiple locations and the other for quantifying the damage. The developed algorithms have been successfully utilized for detecting and quantifying damage in bonded repair systems using simulation and experimental results. These methods can also be used to detect damage at multiple locations in a bonded repair system. The mode shapes of the test structure are obtained analytically using linear elastic finite element analysis and experimentally with a laser vibrometer. The results obtained are satisfactory.