offline and online investigation of drop impact damage on gfrp composite using non-destructive data by artificial neural network

the objective of this experimental work was to assess the drop impact damage onwoven glass fibre reinforced polymer composite laminate through online method and offlinemethod. 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 aesignals, 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. offlineassessment of impact damage on composite laminate was also observed by ultrasonic techniqueand it was inferred that ultrasonic parameters, namely amplitude and attenuation ratio weredecreased with increase in impact energy of test. but attenuation coefficient had an indirectrelationship with impact energy. during online/offline monitoring of composite laminate theae/ut parameters which were obtained from real time monitoring are used to predict impactdamage tolerance (idt) using a separate trained artificial neural network model. based on theidt value of composite, the component should be continued in-service or replaced.