Development of effective Structural Health Monitoring strategies using Self Organizing Map and Index Based Reasoning

Abstract

Structural Health Monitoring (SHM) systems are expected to reduce the mishap rate of the Unmanned Aerial Vehicles (UAV). In this study, frequency responses of structures were analyzed to evaluate the health of the tubes of a small 4 rotor helicopter and propellers separately. For inspection of the integrity of the tubes, two piezoelectric actuators were installed at two ends of a tube. One of them was excited by using the swept sine wave while the frequency domain response characteristics of the other one were encoded. Self Organizing Map (SOM) and Index Based Reasoning (IBR) were used for analysis and interpretation of the response respectively. The combination of Fast Fourier Transformation (FFT) and IBR was used for detection of the material loss of the propellers. To evaluate the performance of the proposed methods, additional data was generated around the experimental results. The proposed methods worked effectively on the experimental and synthetic data.