Uvais Qidwai and Mohamed Shakir, Qatar University, Qatar

This paper describes a classification method for raw sensor data using a Fuzzy Inference System to detect the defects in large LNG tanks. The data is obtained from a Magnetic Flux Leakage (MFL) sensing system which is usually used in the industry to located defects in metallic surfaces, such as tank floors. A robotic inspection system has been developed in conjunction with the presented work which performs the same inspection tasks at much lower temperatures than human operators would thus reducing the shutdown time significantly which is typically of the order of 15-20 million Dollars per day. The main challenge was to come up with an algorithm that can map the human heuristics used by the MFL inspectors in field to locate the defects into an automated system and yet keep the algorithm simple enough to be deployed in near real-time applications. Unlike the human operation of the MFL equipment, the proposed technique is not very sensitive to the sensor distance from the test surface and the calibration requirements are also very minimal which are usually a big impediment in speedy inspections of the floor by human operator. The use of wavelet decomposition with Coiflet waves has been utilized here for deconvolving the essential features of the signal before calculating the classification features. This wavelet was selected to its canny resemblance with the actual MFL signals that makes these wavelets very natural basis function for decomposition.

Nondestructive Testing (NDT), Fuzzy Inference System, Defect Detection, Classification, filters, Wavelet based Deconvolution, Coiflet Transform.