Fuzzy Time-Frequency defect classifier for NDT applications

In this paper, a customized classifier is presented for the industry-practiced nondestructive evaluation (NDE) protocols using a hybrid-fuzzy inference system (FIS) to classify the and characterize the defects commonly present in the steel pipes used in the gas/petroleum industry. The presented system is hybrid in the sense that it utilizes both soft computing through fuzzy set theory, as well as conventional parametric analysis through time-frequency (TF) methods. Various TF transforms have been tested and the most suitable one for this application, multiform tiltable exponential distribution (MTED), is presented here. Four defining states are considered in the paper; slag, porosity, crack, and lack-of-fusion, representing the four most critical types of defects present in welds on the pipes. The necessary features are calculated using the TF coefficients and are then supplied to the fuzzy inference system as input to be used in the classification. The resulting system has shown excellent defect classification with very low misclassification and false alarm rates.