Dr. Bryan Bewick at PEC, as supported by Matt Barsotti and Kirk Marchand, has been awarded a Phase I Small Business Innovative Research (SBIR) project by the Defense Threat Reduction Agency (DTRA) to investigate the use of direct mapping coarse grain artificial neural network (CGANN) modeling approaches in weapon induced groundshock modeling. Dr. Bewick, in cooperation with Dr. Ian Flood of the University of Florida, will customize CGANN driving equations specifically designed to work within a coarse-grain spatial environment. As fundamental physical laws of ground shock wave propagation cannot be used directly to determine the form of these coarse-grain driving equations, empirical modeling methods (specifically ANNs) will be used for this purpose. These ANN based equations will be trained using results generated from conventional fine-grain FE simulations. Test data will be used to validate the approach. Contact Bryan Bewick at PEC for more information.