Quality controls for the Mu2e electromagnetic calorimeter and R&D studies for the muon capture rate determination

According to the SM, while Lepton Flavour Violation is allowed in the neutral sector, Charged Lepton Flavour Violation (CLFV) processes are forbidden. The Mu2e Experiment at Fermilab will search for the CLFV process of neutrinoless conversion of a muon into an electron within the field of an Al nucleus. The Mu2e detectors and its state-of-the-art superconducting magnetic system are presented, with special focus put to the electromagnetic crystal calorimeter. The calorimeter is composed by two annular disks, each one hosting pure CsI crystals read-out by custom silicon photomultipliers (SiPMs). The SiPMs are amplified by custom electronics (FEE) and are glued to copper holders in group of 2 SiPMs and 2 FEE boards thus forming a crystal Readout Unit. These Readout Units are being tested at the Quality Control (QC) Station, whose design, realization and operations are presented in this work. The QC Station allows to determine the gain, the response and the photon detection efficiency of each unit and to evaluate the dependence of these parameters from the supply voltage and temperature. The station is powered by two remotely-controlled power supplies and monitored thanks to a Slow Control system which is also illustrated in this work. In this thesis, we also demonstrated that the calorimeter can perform its own measurement of the Mu2e normalization factor, i.e. the counting of the 1.8 MeV photon line produced in nuclear muon captures. A specific calorimeter sub-system called CAPHRI, composed by four LYSO crystals with SiPM readout, has been designed and tested. We simulated the capability of this system on performing this task showing that it can get a faster and more reliable measurement of the muon capture rates with respect to the current Mu2e detector dedicated to this measurement. The characterization of energy resolution and response uniformity of the four procured LYSO crystals are llustrated.

ICE Emissions Estimation through Machine-learning techniques to support real-world R&D experiments

The emissions estimation, both during homologation and standard driving, is one of the new challenges that automotive industries have to face. The new European and American regulation will allow a lower and lower quantity of Carbon Monoxide emission and will require that all the vehicles have to be able to monitor their own pollutants production. Since numerical models are too computationally expensive and approximated, new solutions based on Machine Learning are replacing standard techniques. In this project we considered a real V12 Internal Combustion Engine to propose a novel approach pushing Random Forests to generate meaningful prediction also in extreme cases (extrapolation, very high frequency peaks, noisy instrumentation etc.). The present work proposes also a data preprocessing pipeline for strongly unbalanced datasets and a reinterpretation of the regression problem as a classification problem in a logarithmic quantized domain. Results have been evaluated for two different models representing a pure interpolation scenario (more standard) and an extrapolation scenario, to test the out of bounds robustness of the model. The employed metrics take into account different aspects which can affect the homologation procedure, so the final analysis will focus on combining all the specific performances together to obtain the overall conclusions.

Experimental analysis on the use of CCC systems to analyze the in-situ bearing capacity of C&D materials

Compaction is one of the most important processes in roadway construction. It is needed to achieve high quality and uniformity of pavement materials, which in turn better ensure long lasting performance.