Evolutionary optimization for passive gamma tomography reconstruction of circular inclusions in nuclear waste vaults

In this study we investigate numerically the possibility of reconstructing the interior details of a nuclear waste vault using photon flux measurements. The passive gamma ray tomography method is employed in order to simultaneously retrieve the intensity and the location of strongly radioactive components. After reformulating the problem as an optimization problem the objective functional obtained is found to have a multi-modal non-smooth structure. Therefore, evolutionary algorithms are used in order to minimize the objective functional in order to determine the radioactive components. Circular inclusions are considered and detailed numerical results are presented for both exact and noisy data. Copyright q 2007 John Wiley & Sons, Ltd.