Severe accident research has been the main focus for reactor safety research over the past two decades. This largely started with the TMI-2 accident with phenomenological research, and the need to reduce severe accident risk further was re-enforced by the Chernobyl accident in 1986. For existing plants (Krugmann, 2001), measures have been introduced to reduce severe accident vulnerabilities, such as primary and secondary feed and bleed, filtered containment venting, hydrogen control by recombiners, igniters or by inerting, and filtration of control room air intake. For new designs, the IAEA has set more restrictive technical safety objectives (IAEA, 1999) such as severe core damage frequency less than 10 5 per plant operating year, elimination of sequences that could give rise to large early releases, and prevention of containment failure, thus limiting the need for off­site protection measures. These objectives have led to greater emphasis in reducing severe accident risk in the newer evolutionary designs.

There have been numerous research programmes over the past few decades to develop understanding of severe accident-related phenomena and also to develop guidelines for the prevention and mitigation of severe accidents. Much knowledge has been gained and at the present time, there is a reduction of effort on severe accidents R&D worldwide. Some research workers (Krugmann, 2001) believe that sufficient knowledge of severe accident phenomenology now exists. Confirmatory research, however, is still in progress in some areas. Also, clearly further work may be required to support a particular design in the event of new building. Recent research programmes are summarised in Table 15.5.