A fast prediction of plant behaviour in the steam generator tube rupture accident at mihama unit 2 using a similar case

It is important to predict fast and accurately future trend of behaviour of a nuclear power plant in an emergency situation. The case-based reasoning is a strong tool for this purpose because it solves a problem by effectively using past similar cases. This study investigates the applicability of the case-based reasoning as a fast prediction technique of plant behaviour. This paper discusses a prediction of initial plant behaviour in the steam generator tube rupture accident happened at the Mihama nuclear power plant unit 2 by using the behaviour data of an accident of the same type happened at Prairie Island nuclear power plant unit 1. The prediction results coincide well with the reported plant behaviour although there are several important differences in the detailed plant specifications and operator actions between the two SGTR accidents.