Title:Vertex reconstruction in the TAO experiment

Abstract:The Taishan Antineutrino Observatory (TAO) is a tonne-scale gadolinium-doped liquid scintillator satellite experiment of the Jiangmen Underground Neutrino Observatory (JUNO). It is designed to measure the reactor antineutrino energy spectrum with unprecedented energy resolution, better than 2% at 1 MeV. To fully achieve its designed performance, precise vertex reconstruction is crucial. This work reports two distinct vertex reconstruction methods, the charge center algorithm (CCA) and the deep learning algorithm (DLA). We describe the efforts in optimizing and improving these two methods and compare their reconstruction performance. The results show that the CCA and DLA methods can achieve vertex position resolutions better than 20mm (bias<5mm) and 12mm (bias<1.3mm) at 1 MeV, respectively, fully meeting the requirements of the TAO experiment. The reconstruction algorithms developed in this study not only prepare the TAO experiment for its upcoming real data but also hold significant potential for application in other similar experiments.