41199 research outputs found
Sort by
Distributed complex event recognition
No full textThe goal of this thesis document can be divided in three main steps. First, summarize the current state of the art in complex event recognition (CER). Second, motivate the need for a new approach overcoming the issues posed by current solutions. Lastly propose
our own solution for distributing CER systems and analyze the results against current solutions to evaluate possible gains in performances. The ultimate goal when distributing CER is to augment the capacity of the system to deal with the ingestion of more events and/or patterns while maintaining a steady through-
put, in other words to scale. We specically focus on scale-out settings. In such cases, a cluster of interconnected machines (i.e., processing units) over the network distribute the workload so that each deals with only a fraction of it, thus overall a higher erformance
is achieved
Gestione termo-economica del sistema energetico di un ospedale: analisi e miglioramento
Get PDFLa tesi parte dall'analisi dell'attuale gestione energetica dell'impianto e a valle di questa operazione si verifica se sia conveniente eseguire dei miglioramenti complessivi dal punto di vista energetico ed economico tramite un modello di ottimizzazione. Il problema generale è un problema di ottimizzazione non lineare che viene risolto utilizzando il metodo di programmazione mista lineare intera, in cui si linearizzano le curve caratteristiche
Modello di crescita economica: scelte maladattive, degrado ambientale e indeterminatezza
No full textIn particolare, l’obbiettivo della tesi e' quello di studiare gli effetti negativi delle scelte maladattive degli individui dell’economia e di come reagiscono di fronte alla riduzione di guadagno dovuta al degrado creato dalle proprie scelte, scegliendo di lavorare più o meno, muovendo l’economia verso traiettorie che mostreranno grande indeterminatezza locale e global
Modeling the 21cm global signal from first stars and black holes.
Get PDFThe aim of this project is to predict the 21-centimeter
global signal generated by the transition between two hyperfine levels
of the atomic hydrogen coming from the high-redshift universe. This
signal is supposed to be generated during the epoch of formation of
first structures because, once luminous objects are formed, they will
emit UV radiation that penetrates primordial hydrogen and that is able
to alter its spin temperature. This process ends up in a global signal
in the 21-centimeter band that begins when the first structures start
to form and so the spin temperature decouples from the photon
temperature. The signal saturates when reionization completes since
there is no more atomic hydrogen that can emit or absorb in the 21-
centimeter band. The 21 centimeter signal thus, depends on the
ionization and thermal histories of the intergalactic medium. Three
relevant processes determine these two histories: X-ray heating
responsible for the increase of the kinetic temperature of the gas,
Lyman-alpha photons that couple the spin temperature of the gas to its
kinetic temperature and UV ionizing photons that drive the cosmic
reionization. A crucial role is thus played by the sources that
firstly formed in the universe (stars and black holes) since they can
emit photons at all the different frequencies of our interest. In the
first part of the thesis, we adopted a standard analytical model for
structure formation based on the Press-Schechter formalism in order to
obtain thermal and ionization histories (and thus the 21-centimeter
global signal) consistent with already published results. In the
second part of the thesis, we used the semi-numerical code Cosmic
Archaeology Tool (shortly CAT) developed within our research group
(Trinca et al., 2021). CAT is able to follow the evolution of dark
matter halos tracking merger history using the extended PressSchechter formalism and provides an ab.initio description of their
baryonic evolution, starting from the formation of the first stars and
black holes in mini-halos at z=20-30. The model is well anchored to
observations of galaxies and AGN at z<6 and it predicts a reionization
history consistent with observations. We then estimated the 21-
centimeter global signal using the same formalism described above but
with the rate of formation and emission properties of the sources
(stars and accreting black holes) provided by CAT. We obtained a 21-
centimeter global signal with an absorption feature between z=23 and
z=19 and with a depth of 150 mK. The timing and the depth of this
feature is consistent with many other semi-numerical models that
account for star formation in mini-halos, but it is not consistent
with the detection claimed by the EDGES collaboration (Bowman et al.,
2018) since it has a depth three times stronger. We tried to reproduce
this depth considering an additional radio background produced by the
emission of early accreting black hole seeds adopting the same
formalism of Ewall-Wice et al. (2018). We found that considering only
black holes which are accreting with an Eddington ratio larger than
0.01 we may reproduce the observed depth of 500 mK
Design and characterization of the neutron-gamma detection module of the DRAGON project
Get PDFThe DRAGoN (Drone for Radiation detection of Gammas and Neutrons) project aims to design and develop a detection system that is placed on an unmanned aerial vehicle (UAV). It represents an innovative solution for the detection and identification of radioactive materials in a specific area, thanks to the simultaneous detection and discrimination of gamma and neutrons (fast or thermal). In this work, a study and a full characterization of the detection system of the DRAGoN project are presented. In particular, it showed the characterization of the radioactivity counter mode of the detection system, with the plastic scintillator EJ-276 (size 3”x5”). The parameters studied for the characterization are the energy resolution, the time resolution, the gamma efficiency, the pulse shape discrimination between photon and neutrons, and finally, the dead time and the neutron detection response are determined when the system is under a high gamma rate. The last parameters are very important for the objective of the DRAGoN project, which is, identify SNM when the system is under a high gamma background. For a complete assessment of the characterization, Monte Carlos simulations were combined with the experimental results. Finally, the results obtained in the characterization suggest that the EJ-276 3”x5” detector perfectly suits the requirements of the DRAGoN project.
Another challenging study carried out in this work was to test the PSD performance of different inorganic scintillators when they are coupled to a SiPM matrix. The replacement of a conventional PMT by a SiPM matrix in the DRAGoN project will guarantee compactness, lightness, and lower power consumption, increasing in that way the flight time of the UAV. From all the detectors tested, the small plastic scintillators with a SiPM matrix showed a complete discrimination, and with the large detectors a discrimination was visible but not a good result is achieved, nevertheless, it could be improved in the future using different experimental techniques
