No abstract

Infrared thermography (IRT) is a competitive method for nondestructive testing; yet it is susceptible to errors when testing objects with complex geometries. This work investigates the effects of regulating different thermographic testing parameters to optimize the IRT outcomes when testing complex shaped geometries, particularly cylindrical coupons. These parameters include the scanning routine, feed-rate, and heat intensity. Fine-tuning these parameters will be performed with respect to three different variables consisting of workpiece density, defect size, and defect depth. The experimental work is designed around 3D-printed cylindrical coupons, then the obtained thermal images are stitched via image processing tool to expose defects from different scans. The analysis employs a Signal-to-Noise Ratio (SNR) metric in an orthogonal tabulation following a Taguchi Design of Experiment. Moreover, test sensitivity and the best combination of factor levels are determined using Analysis of Means (ANOM) and Analysis of Variance (ANOVA). The outcomes show that the heating intensity factor is the most dominant in exposing flaws with close to 40% mean shift and up to 47% variance fluctuation. The paper introduces the tools employed in the study, and then explains the methodology followed to test one sample quadrant. The results for running the testing on all the scenarios are presented, interpreted, and their implications are recommended.