Search for rare W boson decays to three charged pions using machine learning techniques in proton-proton collisions at the Large Hadron Collider

Abstract

A search is presented for the rare decay of a W boson into the fully hadronic final state of three charged pions. This search uses a combination of proton-proton collision data collected by the CMS experiment at the CERN LHC at center-of-mass energies 13 TeV and 13.6 TeV corresponding to integrated luminosities of 137.6 fb−1 and 61.9 fb−1, respectively. A ditau trigger is used to select events. This takes advantage of algorithms designed to reconstruct and identify pions from the hadronic decay of tau leptons. As a result of the very low transverse momentum distribution of the third pion, we select isolated tracks which satisfy an overall charge requirement on the three objects of plus or minus one. This search uses machine learning tools to improve the sensitivity towards signal. In addition, this thesis presents two projects completed in service of the CMS collaboration. The first quantifies radiation damage to the physical components of the detector and derives necessary corrections. The second encompasses the development and maintenance of a framework which synchronizes critical data used for alignment and calibrations of the detector.