Abstract
Techniques to evaluate corrosion damage to bottom and annular plates (floor conditions) in above-ground tanks is strongly needed in order to protect the environment from contamination caused by the leakage of hazardous products due to corrosion. Source location provides important information for the evaluation of acoustic emission (AE) data. AE signals were generated by pencil lead breaks at arbitrary locations on the bottom of a tank (300 kL capacity) and the P-waves traveling in water were used to investigate the accuracy of AE source location using neural networks (NN). The arrival times of wavefronts were determined by both threshold crossing, using AE waves normalized by the peak amplitude, and threshold crossing using a wavelet transformed waveform. The results were compared in order to discuss source location errors.
