Response of CMS High Granularity Calorimeter Prototype to Charged Pions

Abstract

This thesis presents the calibration and performance study of CMS High Granularity Calorimeter (HGCAL) based on the data collected during the beam tests conducted in 2018. The muon run data was used to calibrate the silicon sensors of HGCAL which record the energy deposits of the incoming high energy particles. The calibration values thus found were then used to study the signal-to-noise ratio of these sensors. The pion run data was later used to check the energy response and resolution of the detector. The simulation of the passage of charged pions through HGCAL test-beam setup was used to understand the leakage and absorbed energy profile. With multiple active layers available, the shower start algorithm was used to predict the approximate location of the first hadronic interaction of the pion with the detector material. In order to improve the pion energy resolution, a simple machine learning regression model was considered which gave better results than the conventional methods, suggesting the possibility for improvement with more complex machine learning models.