A model-based, multichannel, real-time capable sawtooth crash detector

Abstract

Control of the time between sawtooth crashes, necessary for ITER and DEMO, requires real-time detection of the moment of the sawtooth crash. In this paper, estimation of sawtooth crash times is demonstrated using the model-based Interacting Multiple Model (IMM) estimator, based on simplified models for the sawtooth crash. In contrast to previous detectors, this detector uses the spatial extent of the sawtooth crash as detection characteristic. The IMM estimator is tuned and applied to multiple ECE channels at once. A model for the sawtooth crash is introduced, which is used in the IMM algorithm. The IMM algorithm is applied to 7 datasets from the ASDEX Upgrade tokamak. Five crash models with different mixing radii are used. All sawtooth crashes that have been identified beforehand by visual inspection of the data, are detected by the algorithm. A few additional detections are made, which upon closer inspection are seen to be sawtooth crashes, which show a partial reconnection. A closer inspection of the detected normal crashes shows that about 42 % are not well fitted by any of the full reconnection models and show some characteristics of a partial reconnection. In some case, the measurement time is during the sawtooth crashes, which also results in an incorrect estimate of the mixing radius. For data provided at a sampling rate of 1 kHz, the run time of the IMM estimator is below 1 ms, thereby fulfilling real-time requirements.

Collective Thomson scattering measurements of fast-ion transport due to sawtooth crashes in ASDEX Upgrade

“…The observed agreement between CTS measurements on the one hand and TRANSP and EBdyna_go on the other arises despite the fact that both codes assume full reconnection of the flux surfaces during a sawtooth crash, a condition which is often not satisfied [24,25]. In fact, indications of incomplete reconnection are seen for about 40% of the crashes identified at AUG by automated methods [26]. For the two crashes discussed here, figure 8 indicates the presence of significant post-cursor activity at multiple frequencies, which might be related to incomplete reconnection.…”

Real-time plasma state monitoring and supervisory control on TCV

“…EFIT [29] and RT-LIUQE [30]), MHD analysis (e.g. [31,32]), profile estimators (e.g. the RAPTOR-observer [28,33]), ray-tracing codes (e.g.…”

An automated sawtooth detection algorithm for strongly varying plasma conditions and crash characteristics

“…The most characteristic feature of a sawtooth crash is the abrupt flattening of profiles. Previous detection algorithms therefore search for fast changes (edges) in individual signals, e.g., [5,[7][8][9], or for an abrupt change towards a centrally flat profile [6]. In TCV [9], a simple difference filter is applied to averaged central SXR signals with thresholds slowly adapted to previous sawteeth.…”

“…edge detection and does not include any discrimination for signal edges not caused by a sawtooth crash. A model-based approach is applied to ASDEX Upgrade ECE data [6]. The T e profile is compared in steps of 1ms with model predictions.…”

“…Sawtooth control requires reliable crash detection, which ideally should work for all plasma scenarios and sawtooth types. Previous sawtooth detection methods use electron cyclotron emission radiation (ECE) (e.g., [5][6][7][8]) or soft x-ray (SXR) (e.g., [9]) measurements. ECE provides a local measurement for the electron temperature, T e , which is easy to interpret, but ECE measurements can fail when n e is close to or above the cut-off density [10].…”