Investigation of intrinsic errors fields in MAST-U device (Oxford, UK)

In magnetic fusion devices, undesired non-axisymmetric magnetic field perturbations, typically called error fields, have been observed to have a detrimental effect on plasma stability and confinement. These spurious perturbations can arise from many sources, namely misalignment introduced in the construction of the device, imperfections in the manufacture of the field coils, 3D structures in the wall surrounding the plasma, current feeds and the presence of ferromagnetic materials near the plasma surface. Error fields with toroidal mode number n = 1 can destabilize magnetic islands in otherwise tearing-stable plasmas, as the m=2, n=1 mode (being m the poloidal mode number), leading to plasma termination. Moreover, error fields can inhibit the exploration of some operational regimes, such as at low density and at high-pressure, and within a plasma scenario, they can be also responsible of fast ion losses, rotation braking, thus causing plasma performance degradation. The main strategies that can be adopted to reduce n=1 error fields are: a careful alignment of the divertor, poloidal field coils, i.e. applying coil shift and tilt, when assembling a fusion device, and the installation of error field correction coils capable of inducing a magnetic field pattern which counteract the error field source. In this Thesis work, a database of 90 discharges performed in the MAST-U device, upgrade of the previous MAST tokamak, located at the Culham Centre for Fusion Energy, Oxfordshire, England, has been analyzed to achieve the following objectives: i) identify the n=1 error field and ii) investigate the dependence of the m=2, n=1 mode onset, i.e. the locked mode, with plasma density. During the assembly of MAST-U, to minimize the n=1 intrinsic error field due to the coil manufacturing, a careful alignment of the magnetic field coils has been applied. To assess the presence of a residual error field, the compass scan method has been executed. This method consists in triggering a locked mode by applying a n=1 probing error field with different phases and relies on an accurate detection of the locked mode onset. In this Thesis work, a robust and reliable tool able to detect such a triggering mechanism has been developed which allows to reach the objectives above mentioned. Thanks to its portability, such a tool can be also exploited for real-time applications, such as disruption avoidance, as proposed in the next MAST-U campaign.