Stabilization of Burn Conditions in a Two-Temperature Fusion Reactor: Preliminary Results

In this report we show results on the stabilization of the burn conditions of a two-temperature fusion reactor using artificial neural networks. Stabilization at the nominal operating state is obtained for energy confinement times ranging from 4.5 sec to 6.5 sec, in a model with parameters taken from the EDA ITER design but with Te = 13 Kev and Ti = 12.7 Kev, as the nominal electron and ion temperatures, respectively. Control actions are implemented through the concurrent modulation of a DT refueling rate, a neutral 4He beam and independent auxiliary heating power to ions and to electrons. Examples of typical transients are used to illustrate the results