| الوصف: |
This thesis presents the results of a study of electromagnetic interactions in the ProtoDUNE-SP liquid argon time projection chamber (LArTPC) detector. The LArTPC detector technology provides high spatial resolution on the final states of charged particle interactions, which allows for different interaction modes to be distinguished based on the geometry of the ionisation energy deposition in the event. As a result, the technology has been used in a number of recent neutrino experiments, and is the technology of choice for the Deep Underground Neutrino Experiment. In order to perform high precision measurements of neutrinos in LArTPC detectors, final state particles need to be effectively identified, and their energy accurately reconstructed. This thesis focusses on these challenges with two studies on data from the ProtoDUNE-SP LArTPC: a study of track-shower classification, and a study of Michel electron energy reconstruction. A track--shower classification algorithm was developed based on the use of convolutional neural networks, and its performance was compared to the current track--shower classification algorithm in ProtoDUNE-SP. The results of this network were used to select a sample of Michel electron events, which were then used to develop a Michel electron reconstruction algorithm based on semantic segmentation with a convolutional neural network. This sample of Michel electron events were then used to estimate the energy resolution and bias for low-energy electrons in ProtoDUNE-SP based on this algorithm. |
