OFFLINE AND ONLINE INVESTIGATION OF DROP IMPACT DAMAGE ON GFRP COMPOSITE USING NON-DESTRUCTIVE DATA BY ARTIFICIAL NEURAL NETWORK

Document Type: Research Paper

Authors

1 Dept. of Mechanical Engineering,Sri Balaji Chockalingam Enginering College, Arni, Tamilnadu, India-632317

2 Dept. of Mechanical Engineering, College of Engineering, Guindy, Anna University, Chennai-600 025, Tamilnadu, India

Abstract

The objective of this experimental work was to assess the drop impact damage on
Woven Glass Fibre Reinforced Polymer composite laminate through online method and offline
method. Online monitoring of drop impact damage was carried out by Acoustic Emission (AE)
technique and AE signals during the drop impact test were captured. From the analysis of AE
signals, it was observed that as the impact energy increases the AE parameters such as counts,
counts to peak, signal strength and root mean square (RMS) values also increase. Offline
assessment of impact damage on composite laminate was also observed by ultrasonic technique
and it was inferred that ultrasonic parameters, namely amplitude and attenuation ratio were
decreased with increase in impact energy of test. But attenuation coefficient had an indirect
relationship with impact energy. During online/offline monitoring of composite laminate the
AE/UT parameters which were obtained from real time monitoring are used to predict Impact
Damage Tolerance (IDT) using a separate trained artificial neural network model. Based on the
IDT value of composite, the component should be continued in-service or replaced.

Keywords