NEURAL NETWORKS FOR DETECTING DEFECTS IN AIRCRAFT STRUCTURES

There are various nondestructive testing (NDT) inspection methods, such as, vision, eddy currdent, and ultrasonic, used for crack or corrosion detection of the skin or the structures of aircraft around rivets and fasteners. These methods require a skilled technician to identify the existence of the cracks. Despite the training that a technician goes through, human error is identified to be one of the major contributing factors to problems in the determination of the safety of aircraft. There has been some effort to develop expert systems that can be used by technicians. However, there is currently no expert system developed that can learn and improve its capability as it encounters new situations. In this paper, the authors review the neural network and its possibility for aiding the technician in detecting defects.