Nondestructive Evaluation of Advanced Material Substances

The manufacture of composite materials for use as high confidence components in aircraft has resulted in a requirement for rigorous defect analysis. This proposal researches the use of neural network technology combined with Laser Ultrasonics to provide the quality assurance inspector with the capability of identifying potential defects in composite structures. Neural network technology is used to pre-process the raw ultrasonics data, determine potential defects, classify defect types, and assist the inspector in determining the suitability of a composite component or structure.