A systematic and comprehensive numerical investigation is carried out in order to understand linear and quasi-linear effects of the applied resonant magnetic perturbations (RMPs) on the plasma, during the initial phase of ITER operation at 5 MA plasma current and 1.8 T toroidal field, assuming a limited number of available power supplies (PSs) for controlling the type-I edge localized modes (ELMs) and taking into account uncertainties associated with predicting the plasma toroidal rotation in ITER. Utilizing the plasma surface displacement near the X-point as a proxy for the ELM control capability by RMPs, the coil phasing optimization finds optima that are not sensitive to the variation of the assumed plasma toroidal flow, nor to the choice of toroidal waveform of the coil current distribution. Several assumptions regarding the number of PSs available in this initial phase have been explored, comprising 9 and 13 for current waveforms with n = 3 symmetry: using the equatorial coil row alone, upper and lower rows of coils and the three rows of ELM control coils. The resulting effects on ELM control (through the X-point displacement proxy) and core and edge torques have been evaluated. Optimization and comparison for the three ELM control coil configurations—all three rows, upper and lower rows, and equatorial row—reveal a strong coupling between the core and edge plasma response—both linear and quasi-linear—for these ITER plasma conditions. The equatorial ELM control coil row has the strongest effects on core and edge plasma response, but provides no flexibility in its optimization; this can only be achieved by optimizing the current waveform with the upper and lower rows. A quantitative comparison of the achievable optimization of ELM control and core plasma response with the three configurations of coils has been performed, assuming that nine PSs will be available in the initial ITER operation.
Toroidal plasma response based ELM control coil design for EU DEMO