Anomaly detection in composite elements using Lamb waves and soft computing methods

Composite materials are widely used in many important structures, which in turn entails the need to develop sensitive and reliable structural health monitoring (SHM) systems. The aim of this study was to investigate the use of guided waves and artificial neural networks as essential components of a two-stage diagnostics system. This system was designed to detect anomalies and to assess their parameters. This paper presents the first result of the application of this system for evaluation of samples made from composite materials. Defects of various origin were artificially introduced. Grids of 8 and 12 piezoelectric transducers were used. Principal components analysis was used for dimensionality reduction of measured signals. Examples of preliminary fault detection results showed that any signal anomalies are detected perfectly whereas the prediction of damage level allowed to distinguishing the defects. Successful experiments carried out on the studied specimens have already proved that this system was able to perform automatic analysis of the elastic waves and accelerate the process of structures diagnosis.