DSpace Collection:
http://www.openstarts.units.it:80/dspace/handle/10077/93
2015-05-30T00:23:25ZHigh resolution ship hydrodynamics simulations in opens source environment
http://www.openstarts.units.it:80/dspace/handle/10077/10983
Title: High resolution ship hydrodynamics simulations in opens source environment
Authors: Del Puppo, Norman
Abstract: The numerical simulation of wake and free-surface flow around ships is a complex topic that involves multiple tasks: the generation of an optimal computational grid and the development of numerical algorithms capable to predict the flow field around a hull. In this work, a numerical framework is developed aimed at high-resolution CFD simulations of turbulent, free-surface flows around ship hulls. The framework consists in the concatenation of “tools” in the open-source finite volume library OpenFOAM®. A novel, flexible mesh-generation algorithm is presented, capable of producing high-quality computational grids for free-surface ship hydrodynamics. The numerical framework is used to solve some benchmark problems, providing results that are in excellent agreement with the experimental measures.
Description: 2013/2014
Type: Tesi di dottorato; Doctoral Thesis2015-04-16T00:00:00ZPersonalized setup of high frequency percussive ventilator by estimation of respiratory system viscoelastic parameters
http://www.openstarts.units.it:80/dspace/handle/10077/10976
Title: Personalized setup of high frequency percussive ventilator by estimation of respiratory system viscoelastic parameters
Authors: Ajčević, Miloš
Abstract: High Frequency Percussive Ventilation (HFPV) is a non-conventional ventilatory modality which has proven highly effective in patients with severe gas exchange impairment. However, at the present time, HFPV ventilator provides only airway pressure measurement. The airway pressure measurements and gas exchange analysis are currently the only parameters that guide the physician during the HFPV ventilator setup and treatment monitoring. The evaluation of respiratory system resistance and compliance parameters in patients undergoing mechanical ventilation is used for lung dysfunctions detection, ventilation setup and treatment effect evaluation. Furthermore, the pressure measured by ventilator represents the sum of the endotracheal tube pressure drop and the tracheal pressure. From the clinical point of view, it is very important to take into account the real amount of pressure dissipated by endotracheal tube to avoid lung injury. HFPV is pressure controlled logic ventilation, thus hypoventilation and hyperventilation cases are possible because of tidal volume variations in function of pulmonary and endotracheal tube impedance.
This thesis offers a new approach for HFPV ventilator setup in accordance with protective ventilatory strategy and optimization of alveolar recruitment using estimation of the respiratory mechanics parameters and endotracheal pressure drop. Respiratory system resistance and compliance parameters were estimated, firstly in vitro and successively in patients undergoing HFPV, applying least squares regression on Dorkin high frequency model starting from measured respiratory signals. The Blasius model was identified as the most adequate to estimate pressure drop across the endotracheal tube during HFPV. Beside measurement device was developed in order to measure respiratory parameters in patients undergoing HFPV.
The possibility to tailor HFPV ventilator setup, using respiratory signals measurement and estimation of respiratory system resistance, compliance and endotracheal tube pressure drop, provided by this thesis, opens a new prospective to this particular ventilatory strategy, improving its beneficial effects and minimizing ventilator-induced lung damage.
Description: 2013/2014
Type: Tesi di dottorato; Doctoral Thesis2015-04-13T00:00:00ZDistributed Discrete Consensus Algorithms: Theory and Applications for the Task Assignment Problem
http://www.openstarts.units.it:80/dspace/handle/10077/10975
Title: Distributed Discrete Consensus Algorithms: Theory and Applications for the Task Assignment Problem
Authors: Pedroncelli, Giovanni
Abstract: Distributed computation paradigms belong to a research field of increasing interest. Using these algorithms will allow to exploit the capabilities of large scale networks and systems in the near future. Relevant information for the resolution of a problem are distributed among a network of agents with limited memory and computation capability; the problem is solved only by means of local computation and message exchange between neighbour agents.
In this thesis we consider the multi-agent assignment problem dealt with distributed computation: a network of agents has to cooperatively negotiate the assignment of a number of tasks by applying a distributed discrete consensus algorithm which defines how the agents exchange information. Consensus algorithms are dealt with always more frequently in the related scientific literature.
Therefore, in the first chapter of this thesis we present a related literature review containing some of the most interesting works concerning distributed computation and, in particular, distributed consensus algorithms: some of these works deal with the theory of consensus algorithms, in particular convergence properties, others deal with applications of these algorithms.
In the second chapter the main contribution of this thesis is presented: aniterative distributed discrete consensus algorithm based on the resolution of local linear integer optimization problems (L-ILPs) to be used for the multi-agent assignment problem. The algorithm is characterized by theorems proving convergence to a final solution and the value of the convergence time expressed in terms of number of iterations. The chapter is concluded by a performance analysis by means of the results of simulations performed with Matlab software. All the results are presented considering two different network topologies in order to model two different real life scenarios for the connection among agents.
The third chapter presents an interesting application of the proposed algorithm: a network of charging stations (considered as agents) has to reach a consensus on the assignment of a number of Electric Vehicles (EVs) requiring to be recharged. In this application the algorithm proposed in the previous chapter undergoes several modifications in order to model effectively this case: considering the inter-arrival times of vehicles to a charging station, a non-linear element appears in the objective function and therefore a novel algorithm to be performed before the assignment algorithm is presented; this algorithm defines the order in which the assigned vehicles have to reach a charging station. Moreover, a communication protocol is proposed by which charging stations and vehicles can communicate and exchange information also allowing charging stations to send to each assigned vehicle the maximum waiting time which can pass before a vehicle loses its right to be recharged. The chapter ends with an example of application of the rivisited assignment algorithm.
In the fourth and last chapter, we present an application in an industrial environment: a network of Autonomous Guided Vehicles (AGVs) in a warehouse modeled as a graph has to perform the distributed discrete consensus algorithm in order to assign themselves a set of destinations in which some tasks are located. This application deals not only with the task assignment problem but also with the following destination reaching problem: therefore a distributed coordination algorithm is proposed which allows the AGVs to move into the warehouse avoiding collisions and deadlock. An example of the control strategy application involving both the assignment and coordination algorithms concludes this chapter.
Description: 2013/2014
Type: Tesi di dottorato; Doctoral Thesis2015-04-13T00:00:00ZApplication of linear and nonlinear methods for processing HRV and EEG signals
http://www.openstarts.units.it:80/dspace/handle/10077/10974
Title: Application of linear and nonlinear methods for processing HRV and EEG signals
Authors: Fornasa, Elisa
Abstract: L'elaborazione dei segnali biomedici è fondamentale per l'interpretazione oggettiva dei sistemi fisiologici, infatti, permette di estrarre e quantificare le informazioni contenute nei segnali che sono generati dai sistemi oggetto di studio. Per analizzare i segnali biomedici, sono stati introdotti un gran numero di algoritmi inizialmente nati in ambiti di ricerca differenti. Negli ultimi decenni, il classico approccio lineare, basato principalmente sull'analisi spettrale, è stato affiancato con successo da metodi e tecniche derivanti dalla teoria della dinamica nonlineare e, in particolare, da quella del caos deterministico.
L'obiettivo di questa tesi è quello di valutare i risultati dell'applicazione di diversi metodi di elaborazione, lineari e non lineari, a specifici studi clinici basati sul segnale di variabilità cardiaca (Heart Rate Variability, HRV) e sul segnale elettroencefalografico (EEG). Questi segnali, infatti, mostrano comportamenti attribuibili a sistemi la cui natura può essere alternativamente di tipo lineare o non, a seconda delle condizioni nelle quali i sistemi vengono analizzati.
Nella prima parte della tesi, sono presentati i due segnali oggetto di studio (HRV ed EEG) e le tecniche di analisi utilizzate. Nel capitolo 1 vengono descritti il significato fisiologico, i requisiti necessari per l'acquisizione dei dati e i metodi di pre-elaborazione dei segnali. Nel capitolo 2 sono presentati i metodi e gli algoritmi utilizzati in questa tesi per la caratterizzazione delle diverse condizioni sperimentali in cui HRV e EEG sono stati studiati, prestando particolare attenzione alle tecniche di analisi non lineare.
Nei capitoli seguenti (capitoli 3-7), sono presentate le cinque applicazioni dell'analisi dei segnali HRV ed EEG esaminate durante il dottorato. Più precisamente, le prime tre riguardano la variabilità cardiaca, le altre due il segnale EEG. Per quanto riguarda il segnale HRV, il primo studio analizza le variazioni delle proprietà spettrali e frattali in soggetti sani di diversa età; il secondo è focalizzatosull'importanza dell'approccio nonlineare nell'analisi del segnale HRV ricavato da registrazioni polisonnografiche di pazienti affetti da gravi apnee notturne; il terzo presenta le differenze nelle caratteristiche spettrali e nonlineari della variabilità cardiaca in pazienti con scompenso cardiaco determinato da diverse eziologie. Invece, per il segnale EEG, il primo studio analizza le alterazioni negli indici spettrali e nonlineari in pazienti con deficit cognitivi soggettivi e lievi, mentre il secondo valuta l'efficacia di un nuovo protocollo per la riabilitazione della malattia di Parkinson, attraverso la quantificazione dei parametri spettrali dell'EEG.
Description: 2013/2014
Type: Tesi di dottorato; Doctoral Thesis2015-04-13T00:00:00Z