Application of Neural Networks to Active Damage Detection Technique of Composites

摘要

Cost-effective and reliable damage detection is critical for the utilization of composite materials. This paper applies a new combined method of Kohonen self-organizing feature neural networks and active damage detection technique to composite health monitoring. The proposed method features a simple structure algorithm, supervision free self-study and lateral association, etc. According to such method and virtual instrument technique, an active damage detection system for composites is developed. This effective method for damage detection has been verified in the course of its practical use.