Data acquisition with real-time numerical integration for COMPASS-U magnetic diagnostics

0571829 - ÚFP 2024 RIV NL eng J - Journal Article
Torres, Andre - Carvalho, B.B. - Markovič, Tomáš - Batista, A.J.N. - Havránek, Aleš - Weinzettl, Vladimír - Fernandes, H.
Data acquisition with real-time numerical integration for COMPASS-U magnetic diagnostics.
Fusion Engineering and Design. Roč. 191, June (2023), č. článku 113580. ISSN 0920-3796. E-ISSN 1873-7196
R&D Projects: GA MŠMT(CZ) LM2018117; GA MŠMT(CZ) EF16_019/0000768
EU Projects: European Commission(BE) 101052200
Institutional support: RVO:61389021
Keywords : atca * Data acquisition * Magnetic diagnostic * Magnetics * Real-time * Tokamak
OECD category: Fluids and plasma physics (including surface physics)
Impact factor: 1.7, year: 2022
Method of publishing: Open access
https://www.sciencedirect.com/science/article/pii/S0920379623001643?via%3Dihub

The COMPASS-U tokamak will feature a completely new set of magnetic diagnostics from sensors to data acquisition. In order to control and reconstruct the plasma magnetic equilibrium, the inductive magnetic sensor signals are integrated and processed. For this purpose, a modular data acquisition solution with digital integration is proposed for COMPASS-U. Based on the ATCA platform, and continuing the development path of the low-drift integrators for W7-X and ITER, this solution allows a high degree of flexibility and scalability with state-of-the-art performance that can benefit a machine with different operational stages and scientific goals over years of operation. This article details the effort to qualify the data acquisition electronics on COMPASS, how to take advantage of real-time processing to enhance the dynamic range through composition of signals sampled with different input ranges and bandwidth, and the implementation of a strategy to mitigate the loss of signal integrity when using such a composition of signals. Tests on COMPASS showed that the digital integration performs well with real probe signals for basic control purposes even in the presence of fast events. A real-time compatible algorithm to recover high-frequency components was also successfully tested, albeit with significant noise added, while simulation shows that obtaining a signal from two different sampling sources can increase the dynamic range with minimal impact on the signal integrity.
Permanent Link: https://hdl.handle.net/11104/0342738