To The Director of Radioprotection and Nuclear Security
Paris, August 12th 2003
In your letter, you ask me to examine the EPR reactor capacity of resistance to/to withstand an potential commercial plane crash, and then to make any necessary suggestions. Very quickly after the September 11th attacks in the USA , the EDF made a point of analysing the problem and in particular with regards to the conception/design of the EPR.
As you note in your letter, the new project takes into account resistance to a military plane crash, which is already a heavy charge. For this, the designers have chosen functional and geographical building plans taking account of such accidents. The project has 4 trains which are completely separate, and a part of the construction is "bunkerised", in particular the buildings containing the reactor and used nuclear fuel, and one building containing 2 of the 4 safeguards trains (electrical and mechanical parts).
The "bunkerised" part, designed to resist to the impact of a military plane, presents a high resistance and especially with regards to perforation : a military plane is considered to be the equivalent of a perforating missile.
All this gives to the EPR an important capacity to resist to the impact of a commercial plane, so no change has been made in the construction plans.
Despite this capacity for resistance to plane accidents, it is nevertheless necessary to note that EDF is not in a position to ensure resistance to eventual war or terrorist action. Prevention or limitation of such action and its possible results involve State responsibility In this case
- The controls concerning resistance to such accidents and any necessary supplementary measures are to be considered as outside the (normal )design basis of the building, and I am obliged to place this situation amongst the " Risk Reduction Categories "
- The study of different possibilities concerning an impact should induce a reasonable response to the risk incurred and will not be able to take into consideration/cover each and every possibility . Furthermore, the measures should, in my opinion, be in complete coherence with the measures adopted internationally, and should not be too different from the measures adopted for other industrial risks. I also consider that the different scenarios studied, the rules and analysis used to do so should not appear in the security reports immediately available or which could become available to the public.
Precisions concerning this general logic are to be found in the joined annex. And, added to this, in order to decide or control the design basis of the protective construction/shear wall of the "bunkerised" part of the building, it will be necessary to define a reference impact load. This reference, whilst generally covering the case of the sort of planes which could crash in the event of an intentional action, should not be associated directly to a particular plane nor to a particular speed of impact. It should correspond to a general hypothesis based on criteria and calculation of a general and conventional nature.
For this reason I propose to retain as the reference the impact charge given in the annex which represents the risks reasonably considered possible by the sorts of planes in European skies.