Etude du calorilmètre hadronique semi-digital et étude du canal physique e+ e- -H nu nu ( H -> WW->qq qq) au collisionneur circulaire electron positon (CEPC) ; Study of semi-digital hadronic calorimeter and study of physics channel e+e− → νe¯νe H(H → W+W− → qq¯′qq¯′) in the Circular Electron Positrion Collider of √s = 240GeV

The discovery of the Higgs boson by the ATLAS and CMS experiments in 2012 opens a new era for particle physics. Thus the high precision measurements of the properties of the Higgs boson is one of the most important goals for future leptonic collider projects such as ILC, CEPC, CLIC, FCC in the following decades. For those future experiments, obtaining excellent Jet Energy Resolution (JER) is very important for their physics study performance. In order to achieve th is purpose, Particle Flow Algorithms (PFA) provides one of the most satisfying approaches. For the application of PFA, the high-granularity calorimeter is needed. The SDHCAL prototype which is the first technological one among the CALICE family of high granularity calorimeters, has been built and exposed to different particle beams in tests at PS and SPS of CERN. It is also one of the options of hadornic calorimeter equipping in ILD and CEPC baseline detectors. Based on the test beam data of SDHCAL, this thesis presents the detail study about the SDHCAL technological prototype including the measurements of efficiency and multiplicity of glass-based RPC detectors, the homogenization process of detector response, the particle identification and energy reconstruction. Using the collected beam muon events, the efficiency and multiplicity of GRPC is found to be around 96% ± 0.04 and 1.81 ± 0.19. Based on the test beam data of 50 GeV and 70 GeV pions taken at 2018, the relative deviation of response (nHit) of detectors improves from 4.9% and 3.9% to 1.8% and 1.4% for 50 and 70 GeV pions respectively. Based on the test beam data collected at 2015, this thesis adopts the BDT method to study the particle identification of SDHCAL technological prototype. It significantly improves the signal (pion) efficiency (>99%) and background rejection rates (>99%) comparing with the cut-based method. Based on the simulation of SDHCAL technological prototype, this thesis adopts the MVA methods including BDT and MLP to improve the reconstructed energy linearity and ...