Nearest Mean Classifier for Defect Classification in CFRP Panels

The use of carbon fiber reinforced plastics (CFRP) has grown quite rapidly especially in the aircraft industry. It is well-known that CFRP laminates experience various types of defects, which if not detected early, can lead to catastrophic damage. In this paper, a parametric classification approach in conjunction with time-domain ultrasonic echo signals is developed to detect and classify three types of defects: impact damage, foreign object inclusion, and porosity. The nearest mean (NM) parametric classifier is selected because it is an optimal classifier under some conditions and also tends to give good results even if the conditions for optimality are not satisfied. Furthermore, it is quite simple to implement. In order to test the classification strategy, one hundred time-domain 5 MHz ultrasonic echo signals were acquired for each type of defect and from defectfree areas on CFRP panels. The signals for each category were randomly partitioned into equal-size training and test sets. The partitioning process was repeated so that at least 1000 signals could be tested for each category and the results could be averaged to obtain a robust estimate of the performance. The mean for each category was estimated from its corresponding training set. During testing, the test signals were start-point aligned with the means using cross-correlation prior to computing the Euclidean distances. The first set of experiments consisted of the dichotomous classification between each type of defect and defect-free signals. The results show that classification accuracies exceeding 99% are possible for each of the three cases. The next experiment focused on a 4-class problem: three defect types and defect-free. For this experiment too, it is shown that classification accuracies exceeding 99% are possible. It can, therefore, be concluded that defects in CFRP panels can be classified quite accurately using the nearest mean classifier in conjunction with ultrasonic echo signals.

References
1. Lubin, G. and S.T. Peters, “Handbook of Composites,” Chapman & Hall, London, 1998. 2. Poudel, A. and T.P. Chu, “Intelligent Nondestructive Testing Expert System for Aircraft Carbon/Carbon Composite Brakes Using Infrared Thermography and Air-coupled Ultrasound,” Materials Evaluation, 70(10), 2012. 3. Kyungmi, L. and V. Estivill-Castro, “Feature extraction and gating techniques for ultrasonic shaft signal classification,” Applied Soft Computing, 7(1), 2007. 4. Kim, J., P. Ramuhalli, L. Udpa and S. Udpa, “Automated multichannel signal classification systems for ultrasonic nondestructive evaluation,” Advanced Nondestructuve Evaluation I, Pts 1 and 2, Proceedings, 321-323, 2006. 5. Meyer, E., and T. Tuthill, “Bayesian classification of ultrasound signals using wavelet coefficients,” Proc. IEEE 1995 National Aerospace and Electronics Conference (NAECON 1995). 6. Simone, G., F.C. Morabito, R. Polikar, P. Ramuhalli, L. Udpa and S. Udpa, “Feature extraction techniques for ultrasonic signal classification,” International Journal of Applied Electromagnetics and Mechanics, 15(1-4), 2001. 7. Duda, R.O., P.E. Hart and D.G. Stork, Pattern Classification, Wiley, New York, 2001.
Metrics
Usage Shares
Total Views
12 Page Views
Total Shares
0 Tweets
12
0 PDF Downloads
0
0 Facebook Shares
Total Usage
12