THE CONTROL ARCHITECTURE OF LARGE SCIENTIFIC FACILITIES: ITER AND LHC LESSONS FOR IFMIF

The development of an intense source of neutrons with the spectrum of deuterium-tritium (DT) fusion reactions is indispensable to qualify suitable materials for the blanket of the nuclear vessel in fusion power plants. An overlap of different layers will absorb the 14 MeV of fusion neutrons that will be converted to thermal energy and generate tritium to feed the DT reactions. IFMIF will reproduce those irradiation conditions with two parallel 40 MeV deuteron linacs, each at 125 mA continuous wave (CW) beam current, colliding on a 25 mm thick liquid Li screen flowing at 15 m/s. A neutron flux of 10 18 m -2 s -1 with a broad peak energy at 14 MeV will be generated in the forward direction through Li(d,xn) nuclear reactions irradiating 500 cm 3 volume capable to house around 1000 small specimens. An availability of the facility above 70% is expected to maximize the irradiation time. The design of the control architecture of a large scientific facility is dependent on the particularities of the processes in place or the volume of data generated; but it is also closely tied to project management issues. The LHC and ITER are two complex facilities with ~10 6 process variables and with different control systems strategies: from the modular approach of CODAC, to the more integrated implementation of CERNs Technical Network. This paper analyses both solutions, and extracts conclusions that shall be applied to the future control architecture of IFMIF.