Acoustic emission source location in unidirectional carbon-fiber-reinforced plastic plates with virtually trained artificial neural networks

Acoustic emission (AE) source location in a unidirectional carbon-fiber-reinforced plastic plate was attempted with artificial neural network (ANN) technology. The AE events were produced by a lead break, and the response wave was received by piezoelectric sensors. The time of arrival, determined through the conventional threshold-crossing technique, was used to measure the dependence of the wave velocity on the fiber orientation. A simple empirical formula, relying on classical lamination and suggested by wave-propagation theory, was able to accurately model the experimental trend. On the basis of the formula, virtual training and testing data sets were generated for the case of a plate monitored by three transducers and were adopted to select two potentially effective ANN architectures. For final validation, experimental tests were carried out, with the source positioned at predetermined points evenly distributed within the plate area. A very satisfactory correlation was found between the actual source locations and those predicted by the virtually trained ANNs.