Analisi numerica della combustione di gas a basso potere calorifico per applicazioni a microturbine a gas

Il presente lavoro si occupa dell’analisi numerica di combustione di gas a basso potere calorifico (gas di sintesi derivanti da pirolisi di biomasse). L’analisi è stata condotta su due principali geometrie di camera di combustione. La prima è un bruciatore sperimentale da laboratorio adatto allo studio delle proprietà di combustione del singas. Esso è introdotto in camera separatamente rispetto ad una corrente d’aria comburente al fine di realizzare una combustione non-premiscelata diffusiva in presenza di swirl. La seconda geometria presa in considerazione è la camera di combustione anulare installata sulla microturbina a gas Elliott TA 80 per la quale si dispone di un modello installato al banco al fine dell’esecuzione di prove sperimentali. I principali obbiettivi conseguiti nello studio sono stati la determinazione numerica del campo di moto a freddo su entrambe le geometrie per poi realizzare simulazioni in combustione mediante l’utilizzo di diversi modelli di combustione. In particolare è stato approfondito lo studio dei modelli steady laminar flamelet ed unsteady flamelet con cui sono state esaminate le distribuzioni di temperatura e delle grandezze tipiche di combustione in camera, confrontando i risultati numerici ottenuti con altri modelli di combustione (Eddy Dissipation ed ED-FR) e con i dati sperimentali a disposizione. Di importanza fondamentale è stata l’analisi delle emissioni inquinanti, realizzata per entrambe le geometrie, che mostra l’entità di tali emissioni e la loro tipologia. Relativamente a questo punto, il maggior interesse si sposta sui risultati ottenuti numericamente nel caso della microturbina, per la quale sono a disposizione misure di emissione ottenute sperimentalmente. Sempre per questa geometria è stato inoltre eseguito il confronto fra microturbina alimentata con singas a confronto con le prestazioni emissive ottenute con il gas naturale. Nel corso dei tre anni, l’esecuzione delle simulazioni e l’analisi critica dei risultati ha suggerito alcuni limiti e semplificazioni eseguite sulle griglie di calcolo realizzate per lo studio numerico. Al fine di eliminare o limitare le semplificazioni o le inesattezze, le geometrie dei combustori e le griglie di calcolo sono state migliorate ed ottimizzate. In merito alle simulazioni realizzate sulla geometria del combustore della microturbina Elliott TA 80 è stata condotta dapprima l’analisi numerica di combustione a pieno carico per poi analizzare le prestazioni ai carichi parziali. Il tutto appoggiandosi a tecniche di simulazione RANS ed ipotizzando alimentazioni a gas naturale e singas derivato da biomasse. Nell’ultimo anno di dottorato è stato dedicato tempo all’approfondimento e allo studio della tecnica Large Eddy Simulation per testarne una applicazione alla geometria del bruciatore sperimentale di laboratorio. In tale simulazione è stato implementato l’SGS model di Smagorinsky-Lilly completo di combustione con modelli flamelet. Dai risultati sono stati estrapolati i profili di temperatura a confronto con i risultati sperimentali e con i risultati RANS. Il tutto in diverse simulazioni a diverso valore del time-step imposto. L’analisi LES, per quanto migliorabile, ha fornito risultati sufficientemente precisi lasciando per il futuro la possibilità di approfondire nuovi modelli adatti all’applicazione diretta sulla MTG.

Comportamento elastodinamico di meccanismi in catena chiusa bilanciati: analisi numerica e sperimentale di un quadrilatero articolato

The study presented in this work deals with the investigation of the effects produced by two common techniques of static balancing on the dynamic performances of closed-chain linkages, taking into account the compliance of the mechanism components. The long-term goal of the research consists in determining an optimal balancing strategy for parallel spatial manipulators. The present contribution is a starting point and it focuses on the planar four-bar linkage, intended as the simplest example of closed-chain mechanism. The elastodynamic behaviour of an unbalanced four-bar linkage and two balanced ones, respectively obtained by mass and elastic balancing, is investigated by means of both numerical simulations and experimental tests. The purpose of this work is to obtain preliminary results, to be refined and broadened in future developments

Fast and accurate numerical solutions in some problems of particle and radiation transport: synthetic acceleration for the method of short characteristics, Doppler-broadened scattering kernel, remote sensing of the cryosphere

The aim of this work is to present various aspects of numerical simulation of particle and radiation transport for industrial and environmental protection applications, to enable the analysis of complex physical processes in a fast, reliable, and efficient way. In the first part we deal with speed-up of numerical simulation of neutron transport for nuclear reactor core analysis. The convergence properties of the source iteration scheme of the Method of Characteristics applied to be heterogeneous structured geometries has been enhanced by means of Boundary Projection Acceleration, enabling the study of 2D and 3D geometries with transport theory without spatial homogenization. The computational performances have been verified with the C5G7 2D and 3D benchmarks, showing a sensible reduction of iterations and CPU time. The second part is devoted to the study of temperature-dependent elastic scattering of neutrons for heavy isotopes near to the thermal zone. A numerical computation of the Doppler convolution of the elastic scattering kernel based on the gas model is presented, for a general energy dependent cross section and scattering law in the center of mass system. The range of integration has been optimized employing a numerical cutoff, allowing a faster numerical evaluation of the convolution integral. Legendre moments of the transfer kernel are subsequently obtained by direct quadrature and a numerical analysis of the convergence is presented. In the third part we focus our attention to remote sensing applications of radiative transfer employed to investigate the Earth's cryosphere. The photon transport equation is applied to simulate reflectivity of glaciers varying the age of the layer of snow or ice, its thickness, the presence or not other underlying layers, the degree of dust included in the snow, creating a framework able to decipher spectral signals collected by orbiting detectors.

Iterative regularization methods for ill-posed problems

This Ph.D thesis focuses on iterative regularization methods for regularizing linear and nonlinear ill-posed problems. Regarding linear problems, three new stopping rules for the Conjugate Gradient method applied to the normal equations are proposed and tested in many numerical simulations, including some tomographic images reconstruction problems. Regarding nonlinear problems, convergence and convergence rate results are provided for a Newton-type method with a modified version of Landweber iteration as an inner iteration in a Banach space setting.

Studio di modelli matematici per la verifica e la progettazione di impianti di digestione anaerobica su scala di laboratorio e industriale

Il lavoro di Dottorato si è incentrato con successo sullo studio della possibilità di applicare il modello ADM1 per la descrizione e verifica di impianti industriali di digestione anaerobica. Dai dati sperimentali il modello e l'implementazione in software di analisi numerica si sono rivelati strumenti efficaci. Il software sviluppato è stato utilizzato come strumento di progettazione di impianti alimentati con biomasse innovative, analizzate con metodiche biochimiche (BMP) in scala di laboratorio. Lo studio è stato corredato con lo studio di fattibilità di un impianto reale con verifica di ottimo economico.

Spectral estimates and preconditioning for saddle point systems arising from optimization problems

In this thesis, we consider the problem of solving large and sparse linear systems of saddle point type stemming from optimization problems. The focus of the thesis is on iterative methods, and new preconditioning srategies are proposed, along with novel spectral estimtates for the matrices involved.

Modelling the dynamic response of rockfall protection barriers

Mountainous areas are prone to natural hazards like rockfalls. Among the many countermeasures, rockfall protection barriers represent an effective solution to mitigate the risk. They are metallic structures designed to intercept rocks falling from unstable slopes, thus dissipating the energy deriving from the impact. This study aims at providing a better understanding of the response of several rockfall barrier types, through the development of rather sophisticated three-dimensional numerical finite elements models which take into account for the highly dynamic and non-linear conditions of such events. The models are built considering the actual geometrical and mechanical properties of real systems. Particular attention is given to the connecting details between the structural components and to their interactions. The importance of the work lies in being able to support a wide experimental activity with appropriate numerical modelling. The data of several full-scale tests carried out on barrier prototypes, as well as on their structural components, are combined with results of numerical simulations. Though the models are designed with relatively simple solutions in order to obtain a low computational cost of the simulations, they are able to reproduce with great accuracy the test results, thus validating the reliability of the numerical strategy proposed for the design of these structures. The developed models have shown to be readily applied to predict the barrier performance under different possible scenarios, by varying the initial configuration of the structures and/or of the impact conditions. Furthermore, the numerical models enable to optimize the design of these structures and to evaluate the benefit of possible solutions. Finally it is shown they can be also used as a valuable supporting tool for the operators within a rockfall risk assessment procedure, to gain crucial understanding of the performance of existing barriers in working conditions.

Tensor-Train decomposition for image classification problems

In these last years a great effort has been put in the development of new techniques for automatic object classification, also due to the consequences in many applications such as medical imaging or driverless cars. To this end, several mathematical models have been developed from logistic regression to neural networks. A crucial aspect of these so called classification algorithms is the use of algebraic tools to represent and approximate the input data. In this thesis, we examine two different models for image classification based on a particular tensor decomposition named Tensor-Train (TT) decomposition. The use of tensor approaches preserves the multidimensional structure of the data and the neighboring relations among pixels. Furthermore the Tensor-Train, differently from other tensor decompositions, does not suffer from the curse of dimensionality making it an extremely powerful strategy when dealing with high-dimensional data. It also allows data compression when combined with truncation strategies that reduce memory requirements without spoiling classification performance. The first model we propose is based on a direct decomposition of the database by means of the TT decomposition to find basis vectors used to classify a new object. The second model is a tensor dictionary learning model, based on the TT decomposition where the terms of the decomposition are estimated using a proximal alternating linearized minimization algorithm with a spectral stepsize.

Variational and learning models for image and time series inverse problems

Inverse problems are at the core of many challenging applications. Variational and learning models provide estimated solutions of inverse problems as the outcome of specific reconstruction maps. In the variational approach, the result of the reconstruction map is the solution of a regularized minimization problem encoding information on the acquisition process and prior knowledge on the solution. In the learning approach, the reconstruction map is a parametric function whose parameters are identified by solving a minimization problem depending on a large set of data. In this thesis, we go beyond this apparent dichotomy between variational and learning models and we show they can be harmoniously merged in unified hybrid frameworks preserving their main advantages. We develop several highly efficient methods based on both these model-driven and data-driven strategies, for which we provide a detailed convergence analysis. The arising algorithms are applied to solve inverse problems involving images and time series. For each task, we show the proposed schemes improve the performances of many other existing methods in terms of both computational burden and quality of the solution. In the first part, we focus on gradient-based regularized variational models which are shown to be effective for segmentation purposes and thermal and medical image enhancement. We consider gradient sparsity-promoting regularized models for which we develop different strategies to estimate the regularization strength. Furthermore, we introduce a novel gradient-based Plug-and-Play convergent scheme considering a deep learning based denoiser trained on the gradient domain. In the second part, we address the tasks of natural image deblurring, image and video super resolution microscopy and positioning time series prediction, through deep learning based methods. We boost the performances of supervised, such as trained convolutional and recurrent networks, and unsupervised deep learning strategies, such as Deep Image Prior, by penalizing the losses with handcrafted regularization terms.

Data driven regularization models of non-linear ill-posed inverse problems in imaging

Imaging technologies are widely used in application fields such as natural sciences, engineering, medicine, and life sciences. A broad class of imaging problems reduces to solve ill-posed inverse problems (IPs). Traditional strategies to solve these ill-posed IPs rely on variational regularization methods, which are based on minimization of suitable energies, and make use of knowledge about the image formation model (forward operator) and prior knowledge on the solution, but lack in incorporating knowledge directly from data. On the other hand, the more recent learned approaches can easily learn the intricate statistics of images depending on a large set of data, but do not have a systematic method for incorporating prior knowledge about the image formation model. The main purpose of this thesis is to discuss data-driven image reconstruction methods which combine the benefits of these two different reconstruction strategies for the solution of highly nonlinear ill-posed inverse problems. Mathematical formulation and numerical approaches for image IPs, including linear as well as strongly nonlinear problems are described. More specifically we address the Electrical impedance Tomography (EIT) reconstruction problem by unrolling the regularized Gauss-Newton method and integrating the regularization learned by a data-adaptive neural network. Furthermore we investigate the solution of non-linear ill-posed IPs introducing a deep-PnP framework that integrates the graph convolutional denoiser into the proximal Gauss-Newton method with a practical application to the EIT, a recently introduced promising imaging technique. Efficient algorithms are then applied to the solution of the limited electrods problem in EIT, combining compressive sensing techniques and deep learning strategies. Finally, a transformer-based neural network architecture is adapted to restore the noisy solution of the Computed Tomography problem recovered using the filtered back-projection method.

Inside standard and hyperspectral low-dose CT variational imaging: parameter identification and material decomposition

The main contribution of this thesis is the proposal of novel strategies for the selection of parameters arising in variational models employed for the solution of inverse problems with data corrupted by Poisson noise. In light of the importance of using a significantly small dose of X-rays in Computed Tomography (CT), and its need of using advanced techniques to reconstruct the objects due to the high level of noise in the data, we will focus on parameter selection principles especially for low photon-counts, i.e. low dose Computed Tomography. For completeness, since such strategies can be adopted for various scenarios where the noise in the data typically follows a Poisson distribution, we will show their performance for other applications such as photography, astronomical and microscopy imaging. More specifically, in the first part of the thesis we will focus on low dose CT data corrupted only by Poisson noise by extending automatic selection strategies designed for Gaussian noise and improving the few existing ones for Poisson. The new approaches will show to outperform the state-of-the-art competitors especially in the low-counting regime. Moreover, we will propose to extend the best performing strategy to the hard task of multi-parameter selection showing promising results. Finally, in the last part of the thesis, we will introduce the problem of material decomposition for hyperspectral CT, which data encodes information of how different materials in the target attenuate X-rays in different ways according to the specific energy. We will conduct a preliminary comparative study to obtain accurate material decomposition starting from few noisy projection data.

Regularization meets GreenAI: a new framework for image reconstruction in life sciences applications

Ill-conditioned inverse problems frequently arise in life sciences, particularly in the context of image deblurring and medical image reconstruction. These problems have been addressed through iterative variational algorithms, which regularize the reconstruction by adding prior knowledge about the problem's solution. Despite the theoretical reliability of these methods, their practical utility is constrained by the time required to converge. Recently, the advent of neural networks allowed the development of reconstruction algorithms that can compute highly accurate solutions with minimal time demands. Regrettably, it is well-known that neural networks are sensitive to unexpected noise, and the quality of their reconstructions quickly deteriorates when the input is slightly perturbed. Modern efforts to address this challenge have led to the creation of massive neural network architectures, but this approach is unsustainable from both ecological and economic standpoints. The recently introduced GreenAI paradigm argues that developing sustainable neural network models is essential for practical applications. In this thesis, we aim to bridge the gap between theory and practice by introducing a novel framework that combines the reliability of model-based iterative algorithms with the speed and accuracy of end-to-end neural networks. Additionally, we demonstrate that our framework yields results comparable to state-of-the-art methods while using relatively small, sustainable models. In the first part of this thesis, we discuss the proposed framework from a theoretical perspective. We provide an extension of classical regularization theory, applicable in scenarios where neural networks are employed to solve inverse problems, and we show there exists a trade-off between accuracy and stability. Furthermore, we demonstrate the effectiveness of our methods in common life science-related scenarios. In the second part of the thesis, we initiate an exploration extending the proposed method into the probabilistic domain. We analyze some properties of deep generative models, revealing their potential applicability in addressing ill-posed inverse problems.

Analisi degli aspetti di qualità nella gestione dei sistemi di drenaggio urbano attraverso modellazione numerica e indagini di campo

In order to protect river water quality, highly affected in urban areas by continuos as intermittent immissions, it is necessary to adopt measures to intercept and treat these polluted flows. In particular during rain events, river water quality is affected by CSOs activation. Built in order to protect the sewer system and the WWTP by increased flows due to heavy rains, CSOs divert excess flows to the receiving water body. On the basis of several scientific papers, and of direct evidences as well, that demonstrate the detrimental effect of CSOs discharges, also the legislative framework moved towards a stream standard point of view. The WFD (EU/69/2000) sets new goals for receiving water quality, and groundwater as well, through an integrated immission/emissions phylosophy, in which emission limits are associated with effluent standards, based on the receiving water characteristics and their specific use. For surface waters the objective is that of a “good” ecological and chemical quality status. A surface water is defined as of good ecological quality if there is only slight departure from the biological community that would be expected in conditions of minimal anthropogenic impact. Each Member State authority is responsible for preparing and implementing a River Basin Management Plan to achieve the good ecological quality, and comply with WFD requirements. In order to cope with WFD targets, and thus to improve urban receiving water quality, a CSOs control strategy need to be implemented. Temporarily storing the overflow (or at least part of it) into tanks and treating it in the WWTP, after the end of the storm, showed good results in reducing total pollutant mass spilled into the receiving river. Italian State Authority, in order to comply with WFD statements, sets general framework, and each Region has to adopt a Water Remediation Plan (PTA, Piano Tutela Acque), setting goals, methods, and terms, to improve river water quality. Emilia Romagna PTA sets 25% reduction up to 2008, and 50% reduction up to 2015 fo total pollutants masses delivered by CSOs spills. In order to plan remediation actions, a deep insight into spills dynamics is thus of great importance. The present thesis tries to understand spills dynamics through a numerical and an experimental approach. A four months monitoring and sampling campaign was set on the Bologna sewer network, and on the Navile Channel, that is the WWTP receiving water , and that receives flows from up to 28 CSOs during rain events. On the other hand, the full model of the sewer network, was build with the commercial software InfoWorks CS. The model was either calibrated with the data from the monitoring and sampling campaign. Through further model simulations interdependencies among masses spilled, rain characteristics and basin characteristics are looked for. The thesis can be seen as a basis for further insighs and for planning remediation actions.

Travi in parete sottile con deformabilità a taglio e distorsione di sezione: formulazione teorica, modellazione numerica e applicazioni

Il presente lavoro di tesi ha riguardato una riformulazione teorica, una modellazione numerica e una serie di applicazioni della Generalized Beam Theory per lo studio dei profili in parete sottile con particolare riguardo ai profili in acciaio formati a freddo. In particolare, in questo lavoro è proposta una riscrittura della cinematica GBT che introduce in una forma originale la deformabilità a taglio della sezione. Tale formulazione consente di conservare il formato della GBT classica e introducendo uno spostamento di warping variabile lungo lo spessore della generica parete della sezione trasversale, garantisce perfetta coerenza tra la componente flessionale e tagliante della trave. E' mostrato, come tale riscrittura consente in maniera agevole di ricondursi alle teorie classiche di trave, anche deformabili a taglio. Inoltre, in tale contesto, è stata messa a punto una procedura di ricostruzione dello sforzo tridimensionale in grado ricostruire la parte reattiva delle componenti di tensioni dovuta al vincolamento interno proprio di un modello a cinematica ridotta. Sulla base di tali strumenti, è stato quindi proposto un approccio progettuale dedicato ai profili in classe 4, definito ESA (Embedded Stability Analysis), in grado di svolgere le verifiche coerentemente con quanto prescritto dalle normative vigenti. Viene infine presentata una procedura numerica per la progettazione di sistemi di copertura formati a freddo. Tale procedura permette di effettuare in pochi semplici passi il progetto dell'arcareccio e dei dettagli costruttivi relativi alla copertura.

Tetti verdi: Analisi sperimentale e Modellazione numerica

I tetti verdi rappresentano, sempre più frequentemente, una tecnologia idonea alla mitigazione alle problematiche connesse all’ urbanizzazione, tuttavia la conoscenza delle prestazioni dei GR estensivi in clima sub-Mediterraneo è ancora limitata. La presente ricerca è supportata da 15 mesi di analisi sperimentali su due GR situati presso la Scuola di Ingegneria di Bologna. Inizialmente vengono comparate, tra loro e rispetto a una superficie di riferimento (RR), le prestazioni idrologiche ed energetiche dei due GR, caratterizzati da vegetazione a Sedum (SR) e a erbe native perenni (NR). Entrambi riducono i volumi defluiti e le temperature superficiali. Il NR si dimostra migliore del SR sia in campo idrologico che termico, la fisiologia della vegetazione del NR determina l'apertura diurna degli stomi e conseguentemente una maggiore evapotraspirazione (ET). Successivamente si sono studiate la variazioni giornaliere di umidità nel substrato del SR riscontrando che la loro ampiezza è influenzata dalla temperatura, dall’umidità iniziale e dalla fase vegetativa. Queste sono state simulate mediante un modello idrologico basato sull'equazione di bilancio idrico e su due modelli convenzionali per la stima della ET potenziale combinati con una funzione di estrazione dell’ umidità dal suolo. Sono stati proposti dei coefficienti di correzione, ottenuti per calibrazione, per considerare le differenze tra la coltura di riferimento e le colture nei GR durante le fasi di crescita. Infine, con l’ausilio di un modello implementato in SWMM 5.1. 007 utilizzando il modulo Low Impact Development (LID) durante simulazioni in continuo (12 mesi) si sono valutate le prestazioni in termini di ritenzione dei plot SR e RR. Il modello, calibrato e validato, mostra di essere in grado di riprodurre in modo soddisfacente i volumi defluiti dai due plot. Il modello, a seguito di una dettagliata calibrazione, potrebbe supportare Ingegneri e Amministrazioni nella valutazioni dei vantaggi derivanti dall'utilizzo dei GR.