Prediction and Detection of Failures in Laminated Composite Materials Using Neural Networks - A Review

Abstract

Reliable failure detection and prediction in laminated composite materials are critical for the utilization of these materials. Many quantitative techniques have been successfully researched and implemented for failure detection and prediction. Most of them are based on non-destructive methods for data collection and are quantified by using neural networks. In this paper a review of the subject is presented. The non-destructive techniques considered are limited to natural frequencies, electric conductivity, and lamb waves. The paper gives a brief introduction to each technique followed by summaries and evaluations for some relevant studies.