After having set the course in November at "1 or even 1,5 (EPR2 reactor) per year" in Europe from the next decade, CEO Luc Rémont revised upwards the group's ambition to "two reactors per year" in a interview given to Context, published Friday.
A pace far from the current level of “one or two (reactors) per decade”.
The industrial challenge is colossal for the group, weighed down by an abysmal debt (54,4 billion euros) and criticized for the setbacks of its EPR projects. Especially since EDF must also respond to the relaunch of a nuclear program in France of up to 18 EPR2 reactors - an improved version of the EPR - and carry out its two English programs, Hinkley Point and Sizewell.
“We have already done four per year”, in the 1970s and 80s, “that’s because it’s possible”, reassured the CEO in November.
Today, only two power plants equipped with EDF EPRs are operating in the world, in Finland and China, while awaiting the commercial start-up of the Normandy Flamanville EPR announced for mid-2024, 12 years late . Fuel loading, a crucial step before connection, is still officially planned by March 31, but the safety authority has warned that the schedule is "tense"...
Launched in 1992 as the pinnacle of nuclear technology, based on an initial Franco-German collaboration, the European pressurized reactor (EPR) was designed to relaunch the atom in Europe, after the Chernobyl disaster of 1986, by promising safety and increased power. But the flagship has continued to accumulate difficulties against a backdrop of loss of skills in the sector while construction of the last reactor commissioned in France began in 1991, 16 years before Flamanville.
“Keep your costs and schedule”
As recently as January, EDF announced that its Hinkley Point project could experience up to six years of delay and an almost doubling of the cost. And a similar scenario is emerging for the six EPR2 program supported by the French government, for which the provisional bill has already increased by 30% (67,4 billion instead of the 51,7 billion announced), according to the newspaper Les Echos.
An inflation of the program that Luc Rémont refuses to confirm at this stage: “For the moment, there is no estimated cost,” he declared to Contexte, arguing that this work is “in progress”.
Nevertheless, this new costing has revived doubts about the electrician's ability to deliver its projects within budgetary and time limits, to the point of annoying the Minister of the Economy Bruno Le Maire: "EDF must learn to keep its costs in check and its calendar.
The group says it is fully mobilized to achieve an optimized and standardized EPR2 model, taking into account the pitfalls of the past. “We must simplify construction (...) standardize the number and type of equipment,” explained Joël Barre, interministerial delegate for new nuclear power, in the Senate on February 8. For EDF, the objective is to achieve a “series effect”, therefore building reactors industrially to be more competitive.
The group is in discussions with the Netherlands, Slovenia, Poland, Finland and Sweden, at a time when the atom is regaining popularity, thanks to the climate imperative to move away from fossil fuels and greater independence vis-à-vis Russia, the largest exporter of power plants. In the Czech Republic, EDF is pre-selected with its South Korean competitor Kepco in a call for tenders for four reactors.
"But for consultant Mycle Schneider, coordinator of a critical annual report on the atom, the World Nuclear Industry Status Report, the very "feasibility" of the EPR is in doubt, while all of Europe is already facing to the “extraordinary” challenge of maintaining and dismantling aging power plants.
According to him, "it is not enough to launch a political appeal and simplify administrative procedures to rebuild a nuclear industry", which is facing a recruitment wall with, in France for example, 100.000 people to find and train over 10 years ".
According to the interministerial delegate for new nuclear power, it is at the end of the "5th and 6th reactor" that the French program will achieve a model that is "competitive in terms of time", i.e. by 2042-2043.