Mainz 2022 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

HK: Fachverband Physik der Hadronen und Kerne

HK 28: Computing I

HK 28.4: Vortrag

Dienstag, 29. März 2022, 17:00–17:15, HK-H5

Space-charge distortions in the ALICE TPC II: Data-driven machine learning algorithms for the space-charge distortion calibration — •Ernst Hellbär¹, Harald Appelshäuser², Marian Ivanov¹, Matthias Kleiner², Silvia Masciocchi¹, and Jens Wiechula² for the ALICE collaboration — ¹GSI Helmholtzzentrum für Schwerionenforschung GmbH, Darmstadt, Germany — ²Institut für Kernphysik, Goethe-Universität Frankfurt, Germany

The Time Projection Chamber (TPC) plays a crucial role in tracking and particle identification for the ALICE experiment at the CERN LHC. The readout of the TPC was upgraded during the long shutdown 2 of the LHC in order to provide the capability to continuously record collision data at 50 kHz of Pb-Pb collisions. The intrinsic properties of the new readout chambers based on Gas Electron Multiplier (GEM) technology lead to a backflow of amplification ions into the drift volume of the TPC which is minimized to below 1 %. In combination with the expected particle multiplicities and high interaction rates in Pb-Pb collisions, the ion backflow (IBF) causes significant space-charge distortions and distortion fluctuations. The latter are relevant on time scales of the order of 10 ms and have to be fully corrected accordingly to restore the intrinsic space-point resolution of the TPC of the order of a few 100 µm. The calibration of the distortion fluctuations is performed using data-driven machine learning algorithms which are trained with simulated data. The calibration procedure and first result of the performance will be presented.
This contribution is supported by BMBF.