Different mechanisms can lead to disruptions in tokamaks. They are generally associated with the slowing down of MHD instabilities and their subsequent locking to the wall, which is believed to be the results of interactions between the perturbed components of the magnetic field, the error fields and currents induced in the vacuum vessel. To improve the understanding of the chain of events ending with a disruption, a statistically robust and physically based criterion has been devised to track the slowing down of modes with toroidal mode numbers n=1 and mostly poloidal mode number m=2, before their actual locking to the wall. This also provides an alternative and earlier detection compared to simple threshold-based indicators. The criterion can also help estimating the time instances at which the macroscopic MHD instabilities become stationary in the reference frame of the laboratory. A marker with high time resolution has been developed and tested on a preliminary database of 375 discharges of AUG and compared with other indicators used on AUG for the detection of the locking. The estimator is based on a weighted averaged value of the Fast Fourier Transform of the perturbed magnetic field radial n=1 filtered component, caused by the rotation of the macroscopic modes. The use of a carrier wave has also allowed improving the performances, tackling the spurious influence of other instabilities, like ELMs. The results show how the indicator constitutes a good candidate for further studies including machine learning approaches for mitigation and avoidance. Indeed, it can be deployed to systematically evaluate the time instance for the expected locking and consequently to populate multimachine databases of the type presently being built in the framework of EUROfusion. Furthermore, it can be also considered as a first feature to describe the chain of events and the dynamic of the precursors.
Alternative Detection of n=1 Modes Slowing Down on AUG
Murari Andrea;
2020
Abstract
Different mechanisms can lead to disruptions in tokamaks. They are generally associated with the slowing down of MHD instabilities and their subsequent locking to the wall, which is believed to be the results of interactions between the perturbed components of the magnetic field, the error fields and currents induced in the vacuum vessel. To improve the understanding of the chain of events ending with a disruption, a statistically robust and physically based criterion has been devised to track the slowing down of modes with toroidal mode numbers n=1 and mostly poloidal mode number m=2, before their actual locking to the wall. This also provides an alternative and earlier detection compared to simple threshold-based indicators. The criterion can also help estimating the time instances at which the macroscopic MHD instabilities become stationary in the reference frame of the laboratory. A marker with high time resolution has been developed and tested on a preliminary database of 375 discharges of AUG and compared with other indicators used on AUG for the detection of the locking. The estimator is based on a weighted averaged value of the Fast Fourier Transform of the perturbed magnetic field radial n=1 filtered component, caused by the rotation of the macroscopic modes. The use of a carrier wave has also allowed improving the performances, tackling the spurious influence of other instabilities, like ELMs. The results show how the indicator constitutes a good candidate for further studies including machine learning approaches for mitigation and avoidance. Indeed, it can be deployed to systematically evaluate the time instance for the expected locking and consequently to populate multimachine databases of the type presently being built in the framework of EUROfusion. Furthermore, it can be also considered as a first feature to describe the chain of events and the dynamic of the precursors.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
