Würzburg 2018 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
T: Fachverband Teilchenphysik
T 20: Neutrinophysik I
T 20.3: Vortrag
Montag, 19. März 2018, 16:35–16:50, Z6 - SR 2.012
Charge-only energy reconstruction with Convolutional Neural Networks for the EXO-200 experiment — •Tobias Ziegler1, Michael Jewell2, Sebastian Schmidt1, Jürgen Hößl1, Gisela Anton1, and Thilo Michel1 — 1Friedrich-Alexander-Universität Erlangen-Nürnberg, ECAP — 2Stanford University, California, USA
The EXO-200 experiment searches for the neutrinoless double beta (0νββ) decay of 136Xe with a single-phase liquid xenon (LXe) time projection chamber (TPC) filled with enriched LXe. The TPC provides the position (X,Y,Z) of events and the deposited energy in LXe by simultaneously detecting the xenon scintillation light and the amount of secondary electrons. For charge collection, electrons drift in the electric field towards the anode, where they induce currents in a first plane of wires and are collected by a second plane of wires. In this study, we investigate the energy reconstruction of events with single or multiple charge deposits using all available collection wires. We apply Deep Learning methods, esp. Convolutional Neural Networks, to reconstruct the charge-only energy deposition in the EXO-200 experiment and compare its performance to the conventional approach.