L’uso frequente dei modelli predittivi per l’analisi di sistemi complessi, naturali o artificiali, sta cambiando il tradizionale approccio alle problematiche ambientali e di rischio. Il continuo miglioramento delle capacità di elaborazione dei computer facilita l’utilizzo e la risoluzione di metodi numerici basati su una discretizzazione spazio-temporale che permette una modellizzazione predittiva di sistemi reali complessi, riproducendo l’evoluzione dei loro patterns spaziali ed calcolando il grado di precisione della simulazione. In questa tesi presentiamo una applicazione di differenti metodi predittivi (Geomatico, Reti Neurali, Land Cover Modeler e Dinamica EGO) in un’area test del Petén, Guatemala. Durante gli ultimi decenni questa regione, inclusa nella Riserva di Biosfera Maya, ha conosciuto una rapida crescita demografica ed un’incontrollata pressione sulle sue risorse naturali. L’area test puó essere suddivisa in sotto-regioni caratterizzate da differenti dinamiche di uso del suolo. Comprendere e quantificare queste differenze permette una migliore approssimazione del sistema reale; é inoltre necessario integrare tutti i parametri fisici e socio-economici, per una rappresentazione più completa della complessità dell’impatto antropico. Data l’assenza di informazioni dettagliate sull’area di studio, quasi tutti i dati sono stati ricavati dall’elaborazione di 11 immagini ETM+, TM e SPOT; abbiamo poi realizzato un’analisi multitemporale dei cambi uso del suolo passati e costruito l’input per alimentare i modelli predittivi. I dati del 1998 e 2000 sono stati usati per la fase di calibrazione per simulare i cambiamenti nella copertura terrestre del 2003, scelta come data di riferimento per la validazione dei risultati. Quest’ultima permette di evidenziare le qualità ed i limiti per ogni modello nelle differenti sub-regioni. / The frequent use of predictive models for analyzing of complex, natural or artificial phenomena is changing the traditional approaches to environmental and hazard problems. The continuous improvement of computer performance allows for more detailed numerical methods, based on space-time discretisation, to be developed and run for a predictive modeling of complex real systems, reproducing the way their spatial patterns evolve and pointing out the degree of simulation accuracy. In this contribution we present an application of several methods (Geomatics, Neural Networks, Land Cover Modeler and Dinamica EGO) in the tropical training area of Petén, Guatemala. During the last few decades this region, included in the Biosphere Maya reserve, has seen a fast demographic raise and a subsequent uncontrolled pressure on its own geo-resources. The test area can be divided into several sub-regions characterized by different land use dynamics. Understanding and quantifying these differences permits a better approximation of a real system; moreover we have to consider all the physical, socio-economic parameters, which will be of use for representing the complex and sometimes random human impact. Because of the absence of detailed data from our test area, nearly all the information was derived from the image processing of 11 ETM+, TM and SPOT scenes; we studied the past environmental dynamics and we built the input layers for the predictive models. The data from 1998 and 2000 were used during the calibration to simulate the land cover changes in 2003, selected as reference date for the validation. The basic statistics permit to highlight the qualities or the weaknesses for each model on the different sub-regions. / L’utilisation chaque jour plus fréquent des modèles prédictifs pour décrire des systèmes complexes, naturels ou artificiels, est en train de changer progressivement les approches traditionnelles des problématiques environnementales et de gestion du risque. Le remarquable potentiel des ordinateurs actuels rend possible le développement de modèles numériques de calcul basés sur une discrétisation spatio-temporelle afin de simuler le comportement de phénomènes complexes ainsi que de prévoir des scénarios futures, avec différents dégréés d’approximation. On présent ici une application de différents modèles prédictifs (géomatique, réseaux de neurones, LCM Idrisi Andes et Dinamica), dans une région de forêt tropicale au Petén, Guatemala. Cette région, inscrite dans la réserve de biosphère Maya, a connu dans les dernières décades, une croissance démographique et une pression non contrôlée sur les ressources naturelles. Dans la région d’étude nous pouvons signaler des sous-régions avec différentes dynamiques d’utilisation du sol. Comprendre et quantifier de telles différences permet l’obtention d’une bonne approximation de la situation réelle et il est nécessaire d’intégrer les paramètres physiques, sociaux et économiques essentiels pour décrire la complexité, parfois aléatoire, de l’action anthropique. La première phase d’étude est basée exclusivement sur l’analyse d’informations satellites, car il n’existe pas d’autres données utiles disponibles sur cette étendue. Le traitement de 11 images ETM+, TM et SPOT a rendu possible l’observation des dynamiques environnementales passées ainsi que la construction de l’Input pour les différentes approches prédictives. Les résultats obtenus, en partent des informations collectées en 1998 et 2000, seront validés avec une image réelle de 2003; les statistiques de base vont permettre de souligner les points forts et les points faibles de chaque modèle sur les différentes sous-régions.

La tesi ha per obiettivo di quantificare gli effetti che la variabilità spaziale del mezzo poroso ha sull'evoluzione di un sistema geochimico. Le reazioni di dissoluzione o precipiazione di minerali modificano la struttura microscopica del mezzo, e con essa le proprietà idrodinamiche del sistema, la permeabilità in modo particolare. La variabilità spaziale iniziale del mezzo può essere causa della formazione di digitazioni o canalizzazioni? La prima parte della tesi tratta il cambiamento di scala, necessario per passare da una simulazione geostatistica su griglia fine al calcolo di trasporto su una tessellazione più grossolana. Nel caso del codice di calcolo Hytec, che implementa uno schema ai volumi finiti basato su discretizzazione in poligoni di Voronoï, sono stati confrontati diversi metodi di calcolo della permeabilità equivalente, seguendo differenti criteri. La seconda parte riguarda i calcoli di trasporto reattivo condotti su famiglie di simulazioni geostatistiche del mezzo; l'influenza della variabilità spaziale iniziale sull'evoluzione dei sistemi viene quantificata grazie ad opportune grandezze osservabili. Sono state studiate due reazioni distinte: un caso di dissoluzione, in maniera più approfondita, e più rapidamente un caso di precipitazione, il cui effetto complessivo è quello di riequilibrare il sistema. / La thèse vise à quantifier les effets de la variabilité spatiale d'un milieux poreux sur l'évolution d'un système géochimique. Les réactions de dissolution ou de précipitation des minéraux modifient la structure microscopique du milieu, et par suite les caractéristiques hydrodynamiques du système, notamment la perméabilité. La variabilité spatiale du milieu initial peut-elle alors être à l'origine de la formation de digitations ou de chenaux? La première partie traite du changement d'échelle, pour le passage d'une simulation géostatistique sur grille fine au calcul de transport sur un maillage plus grossier. Dans le cas du code Hytec, qui utilise un schéma aux volumes finis fondé sur une discrétisation en polygones de Voronoï, plusieurs méthodes de calcul de la perméabilité équivalente sont comparées, suivant différents critères. La deuxième partie aborde les calculs de transport réactif sur une famille de simulations géostatistiques du milieu. L'influence de la variabilité spatiale initiale sur l'évolution des systèmes est quantifiée à l'aide de “observables” convenables. Deux réactions distinctes sont étudiées, un cas de dissolution, de façon approfondie, et, plus rapidement, un cas de précipitation tendant au rééquilibrage du système.

The majority of carbonate reservoir is oil-wet, which is an unfavorable condition for oil production. Generally, the total oil recovery after both primary and secondary recovery in an oil-wet reservoir is low. The amount of producible oil by enhanced oil recovery techniques is still large. Alkali substances are proven to be able to reverse rock wettability from oil-wet to water-wet, which is a favorable condition for oil production. However, the wettability reversal mechanism would require a noneconomical aging period to reach the maximum reversal condition. An intermittent flow with the optimum pausing period is then combined with alkali flooding (combination technique) to increase the wettability reversal mechanism and as a consequence, oil recovery is improved. The aims of this study are to evaluate the efficiency of the combination technique and to study the parameters that affect this method. In order to implement alkali flooding, reservoir rock and fluid properties were gathered, e.g. interfacial tension of fluids, rock wettability, etc. The flooding efficiency curves are obtained from core flooding and used as a major criterion for evaluation the performance of technique. The combination technique improves oil recovery when the alkali concentration is lower than 1% wt. (where the wettability reversal mechanism is dominant). The soap plug (that appears when high alkali concentration is used) is absent in this combination as seen from no drop of production rate. Moreover, the use of low alkali concentration limits alkali loss. This combination probably improves oil recovery also in the fractured carbonate reservoirs in which oil is uneconomically produced. The results from the current study indicate that the combination technique is an option that can improve the production of carbonate reservoirs. And a less quantity of alkali is consumed in the process.

The study is aimed to calculate an innovative numerical index for bit performance evaluation called Bit Index (BI), applied on a new type of bit database named Formation Drillability Catalogue (FDC). A dedicated research programme (developed by Eni E&P and the University of Bologna) studied a drilling model for bit performance evaluation named BI, derived from data recorded while drilling (bit records, master log, wireline log, etc.) and dull bit evaluation. This index is calculated with data collected inside the FDC, a novel classification of Italian formations aimed to the geotechnical and geomechanical characterization and subdivisions of the formations, called Minimum Interval (MI). FDC was conceived and prepared at Eni E&P Div., and contains a large number of significant drilling parameters. Five wells have been identified inside the FDC and have been tested for bit performance evaluation. The values of BI are calculated for each bit run and are compared with the values of the cost per metre. The case study analyzes bits of the same type, diameters and run in the same formation. The BI methodology implemented on MI classification of FDC can improve consistently the bit performances evaluation, and it helps to identify the best performer bits. Moreover, FDC turned out to be functional to BI, since it discloses and organizes formation details that are not easily detectable or usable from bit records or master logs, allowing for targeted bit performance evaluations. At this stage of development, the BI methodology proved to be economic and reliable. The quality of bit performance analysis obtained with BI seems also more effective than the traditional “quick look” analysis, performed on bit records, or on the pure cost per metre evaluation.

L’attività di ricerca del dottorando è stata rivolta allo studio della trasmissione del rumore per via strutturale negli edifici. Questo argomento e' di notevole interesse sia fondamentale che applicativo. Il fenomeno e le problematiche ad essa connesse contribuiscono alla diminuzione delle prestazioni fonoisolanti, sia per le strutture verticali (usualmente valutate rispetto al rumore aereo), sia per quelle orizzontali (valutate anche rispetto al rumore impattivo) ed è tipico degli edifici con struttura portante a telaio e tamponatura in laterizi. Si tratta delle tipiche tipologie edilizie italiane, per le quali il problema risulta amplificato rispetto ad altre tipologie. La metodologia di studio è stata di tipo sperimentale. La scelta è dettata sia dall’insufficiente avanzamento dello stato dell’arte dei metodi di calcolo teorici, o delle loro versioni numeriche, sia dalla necessità di disporre di dati certi da confrontare con i valori forniti dai metodi previsionali semplificati indicati nelle norme UNI (modello CEN); infatti queste ultime sono un recepimento letterale di quelle europee, sebbene esse si basino su tipologie costruttive, materiali e tecniche di realizzazione differenti da quelle comunemente in uso in Italia; da qui la difformità di risultati tra formule previsionali e misurazioni sperimentali. Al fine di realizzare uno studio completo delle principali casistiche della trasmissione laterale sono state utilizzate 6 configurazioni in scala reale, con elementi edilizi diversamente collegati fra loro in modo da simulare i nodi strutturali comunemente realizzati negli edifici. La parte sperimentale della ricerca è stata svolta presso le “Camere Acustiche di Prova” del Laboratorio del Lazzaretto “R. Alessi” del DIENCA. Oltre alle usuali misurazioni di isolamento acustico, sono state eseguite numerose misurazioni di vibrazione. Infatti, dal confronto dei livelli di velocità di vibrazione dei diversi elementi di una struttura, rigidamente connessi, è possibile determinare l’indice di riduzione delle vibrazioni Kij che è la grandezza chiave per modellizzare il fenomeno della trasmissione laterale. La possibilità di determinare sperimentalmente tali valori nel contesto di un lavoro di ricerca finalizzato a determinare i meccanismi di propagazione delle vibrazioni nelle strutture, permette di valutare la precisione delle formule previsionali del modello CEN e di proporne varianti e integrazioni. I valori di Kij così determinati assumono grande importanza anche in fase di progetto, fornendo dati attendibili da utilizzare per la progettazione acustica degli edifici.

No description available.

The conversion coefficients from air kerma to ICRU operational dose equivalent quantities for ENEA’s realization of the X-radiation qualities L10-L35 of the ISO “Low Air Kerma rate” series (L), N10-N40 of the ISO “Narrow spectrum” series (N) and H10-H60 of the ISO “High Air-kerma rate” (H) series and two beams at 5 kV and 7.5 kV were determined by utilising X-ray spectrum measurements. The pulse-height spectra were measured using a planar high-purity germanium spectrometer (HPGe) and unfolded to fluence spectra using a stripping procedure then validate with using Monte Carlo generated data of the spectrometer response. HPGe portable detector has a diameter of 8.5 mm and a thickness of 5 mm. The entrance window of the crystal is collimated by a 0.5 mm thick Aluminum ring to an open diameter of 6.5 mm. The crystal is mounted at a distance of 5 mm from the Berillium window (thickness 25.4 micron). The Monte Carlo method (MCNP-4C) was used to calculate the efficiency, escape and Compton curves of a planar high-purity germanium detector (HPGe) in the 5-60 keV energy. These curves were used for the determination of photon spectra produced by the X-ray machine SEIFERT ISOVOLT 160 kV in order to allow a precise characterization of photon beams in the low energy range, according to the ISO 4037. The detector was modelled with the MCNP computer code and validated with experimental data. To verify the measuring and the stripping procedure, the first and the second half-value layers and the air kerma rate were calculated from the counts spectra and compared with the values measured using an a free-air ionization chamber. For each radiation quality, the spectrum was characterized by the parameters given in ISO 4037-1. The conversion coefficients from the air kerma to the ICRU operational quantities Hp(10), Hp(0.07), H’(0.07) and H*(10) were calculated using monoenergetic conversion coefficients. The results are discussed with respect to ISO 4037-4, and compared with published results for low-energy X-ray spectra. The main motivation for this work was the lack of a treatment of the low photon energy region (from a few keV up to about 60 keV).

L’oggetto della presente dissertazione è inerente aspetti affidabilistici e diagnostici dei componenti delle reti elettriche. Sono stati condotti studi sul ruolo svolto dai parametri presenti nei modelli affidabilistici utilizzati in presenza di regimi distorti. I risultati ottenuti nel corso della ricerca, indicano chiaramente come anche il fattore efficace Krms e, soprattutto, il fattore di forma, Kf, sotto certe condizioni, possano avere effetti considerevoli sulla degradazione degli isolanti (a volte con contributi perfino maggiori di quello dato dal ben noto fattore di picco, Kp, considerato predominante). Viene inoltre riportata un’indagine sviluppata sui principali Dispositivi Automatizzati per il Controllo degli Isolamenti (AIMS), attualmente disponibili sul mercato. Sono illustrati e discussi innovativi modelli di rischio integrati, sviluppati per integrare informazioni fornite dall’analisi affidabilistica tradizionale con misure di proprietà diagnostiche, acquisite grazie ad un monitoraggio costante dei componenti in servizio. L’impiego di tali modelli permetterebbe di ottenere una manutenzione di tipo affidabilistico-diagnostico, basata sull’effettiva condizione del componente in esame (manutenzione tipo CBM), piuttosto che su scadenze temporali fissate a priori (manutenzione di tipo TBM).

The control of a proton exchange membrane fuel cell system (PEM FC) for domestic heat and power supply requires extensive control measures to handle the complicated process. Highly dynamic and non linear behavior, increase drastically the difficulties to find the optimal design and control strategies. The objective is to design, implement and commission a controller for the entire fuel cell system. The fuel cell process and the control system are engineered simultaneously; therefore there is no access to the process hardware during the control system development. Therefore the method of choice was a model based design approach, following the rapid control prototyping (RCP) methodology. The fuel cell system is simulated using a fuel cell library which allowed thermodynamic calculations. In the course of the development the process model is continuously adapted to the real system. The controller application is designed and developed in parallel and thereby tested and verified against the process model. Furthermore, after the commissioning of the real system, the process model can be also better identified and parameterized utilizing measurement data to perform optimization procedures. The process model and the controller application are implemented in Simulink using Mathworks` Real Time Workshop (RTW) and the xPC development suite for MiL (model-in-theloop) and HiL (hardware-in-the-loop) testing. It is possible to completely develop, verify and validate the controller application without depending on the real fuel cell system, which is not available for testing during the development process. The fuel cell system can be immediately taken into operation after connecting the controller to the process.

This research, carried out during the PhD in Materials Engineering, deals with the creation of layers, with different functionality, deposited on a ceramic substrate, to obtain photovoltaic cells for electricity production. The research activities are included in the project PRRIITT, Measure 4 (Development of Networks), Action A (Research and Technology Transfer Laboratories), Thematic reference 3 (Advanced materials applications development), co-financed by the Emilia Romagna Region, for the creation of CECERBENCH laboratory, which aims to develop "Tiles with a functionalised surface”. The innovation lies in the study of materials and in the development of technologies to achieve a "photovoltaic surface", directly in the tiles production process. The goal is to preserve the technical characteristics, and to make available new surfaces, exploiting renewable energy sources. The realization of Building Integrated PhotoVoltaic (BIPV) is nowadays a more and more spread tendency. The aims of the research are essentially linked to the need to diversify the actual ceramic tile production (which is strongly present in the Emilia Romagna Region ), and to provide a higher added value to the tiles. Solar energy production is the primary objective of the functionalization, and has a relevant ecological impact, taking into account the overwhelming global energy demand. The specific activities of the PhD were carried out according to the achievement of scientific and technological objectives of CECERBENCH laboratory, and involved the collaboration in design solutions, to obtain the cells directly on the tiles surface. The author has managed personally a part of the research project. Layers with different features were made: - Electrically conductive layers, directly on the ceramic tiles surface; - Layers to obtain the photovoltaic functionality; - Electrically insulating, protective layers (double function). For each layer, the most suitable materials have been selected. Among the technical application, the screen printing was used. This technique, widely used in ceramics, has many application areas, including the electronics and photovoltaic industries. It is an inexpensive technique, easy to use in industrial production lines. The screen printing technique was therefore studied in depth by theoretical considerations, and through the use of rheological measurements.