A statistical, nonparametric regression analysis utilizing a conditional average and a smoothed empirical probability distribution forms the basis of an automatic modeler that resembles the operation of a neural-like network. Coupled with a small receiver array, the modeler can determine both the distance and direction to a source of acoustic emission. The use of such a small receiver array to locate sources of emission on a thick plate is demonstrated. The characteristic property of such a system is its capability for predicting the locations of various sources in a structure from the emitted waveforms. Using the modeler involves two steps. First, the modeler is trained with the computed Green’s functions for the structure. In the examples shown, these correspond to a normal force and the corresponding displacement signal for a broad range of source/receiver separations. The database of these responses forms the basis of the modeler memory. After training, the modeler can be used to locate sources of emission on the plate. The operation of the system is demonstrated with both synthetic as well as measured waveforms corresponding to impact and step-loading forces on a thick plate. [Work supported by AFOSR.]
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.