Risk optimization: siting of nuclear power electricity generating units

Abstract

Radiological doses from nuclear power plant accidents can extend to great distances away from the site. These technical and perceived risks of nuclear power generation are addressed in an optimization of the siting of additional nuclear units primarily in the middle Atlantic and northeastern United States of America. This study relies on a constrained optimization of long-term latent cancer and early fatality effects from probable severe accident releases to the public. The results provide policy perspectives that would benefit from nuclear power, but provide an inherent safety margin in the selection of sites for future generating units of electricity. The selection of sites from among the best is aided by these calculations for minimizing fatality risks at 17 existing sites. The mathematical model minimizes a linear function represented as the number of 1000 MWe units multiplied by the estimated severe accident site risk at existing sites assuming constraints on: the allowed percentage increase in total societal latent cancer risk as well as allowed percentage increase in early fatality risk; lower bounds for existing units at the sites; and an upper bound of doubling the units and risks. The early fatality risk was also varied for several levels of increase in latent cancer risk.