PATTERN RECOGNITION TECHNIQUES APPLIED TO THEDETECTION AND CLASSIFICATION OF WELDING DEFECTSBY MAGNETIC TESTING

摘要

The importance of nondestructive tests cannot be denied nowadays, and their use in the pet-roleum industry is extremely important to ensure safe operation of the equipment. The mag-netic flux leakage (MFL) technique in instrumented pig is frequently used for pipelineinspection to detect weld defects and internal or external corrosion; however, as every non-destructive technique, it may present report errors, caused by several human factors (likelack of theoretical knowledge or experience in the technique applied, tiredness due to longhours of work, and so on) and by the inspection technique itself. Because pattern recog-nition techniques are already well known and applied in other technological research areas,such as fuzzy logic and artificial neural networks, the possibility to develop an automaticsystem to detect and classify defects through some nondestructive techniques aroused,mainly for those techniques that actually operate with signal or image interpretation. Inthe present work, pattern nonlinear classifiers were applied using artificial neural networks,to check the possibility to detect and classify defects in pipelines inspected throughmagnetic pigs (MFL technique). Several tests were performed on specimens with defectsartificially inserted (internal and external corrosion and lack of penetration).