10081 research outputs found
Sort by
Low-power heterogeneous architectures for efficient and predictable autonomous cyber-physical systems
No full textIn today’s rapidly evolving technological landscape, cyber-physical systems have pervaded various aspects of our daily lives, from autonomous vehicles and healthcare to industrial automation and smart cities. Such applications span a wide range in criticality, performance, and memory footprint, under tight cost and power constraints.
High-end applications rely on power-hungry Systems-on-Chip (SoCs) featuring powerful processors, large LPDDR/DDR3/4/5 memories, and supporting full-fledged Operating Systems (OS). On the contrary, low-end applications typically rely on Ultra-Low-Power µcontrollers with a "close to metal" software environment and simple real-time micro-kernel-based runtimes.
Emerging applications and trends of cyber-physical systems require the "best of both worlds": cheap and low-power SoC systems able to (i) run increasingly complex multi-tasking workloads with large memory footprints within a few hundred mW power budget, (ii) offer a well-known and agile software environment based on full-fledged OS while (iii) providing extreme energy efficient processing capabilities and (iv) not compromising on time-predictability.
In this context, this work presents a threefold contribution. First, the thesis introduces Shaheen, a 22nm low-power (<200mW) heterogeneous SoC designed for autonomous nano-unmanned aerial vehicles, an emerging class of cyber-physical systems. Shaheen features an application-class RV64 host processor with hardware virtualization support, enabling the secure consolidation of a real-time and a full-blown OS onto the same platform. Furthermore, it integrates a flexible cluster of eight RV32 cores, providing state-of-the-art energy-efficient performance for low-power artificial intelligence algorithms. Secondly, this work discusses enhancements to Shaheen’s memory hierarchy, demonstrating the trade-off between low-power and high-end off-chip memories. Lastly, it focuses on the time predictability of AXI-based architectures in safety-critical applications. Namely, it introduces (i) a novel fine-grained methodology for modeling the typical resources composing modern heterogeneous SoCs and (ii) a complete mathematical analysis to upper bound the response time of the interactions between the agents in the system
The measurement of economic insecurity
No full textThe interest in the study of economic insecurity has grown in recent years. However, the ongoing debate about its measurement remains unresolved. This thesis aims to investigate the measurement of economic insecurity by proposing classes of individual objective retrospective indices. The first proposal introduces relative versions of the Bossert et al. (2022) measures. This modification allows for the evaluation of economic insecurity in terms of the proportion of outcomes, favoring the comparison between individuals and between countries adopting different currencies or having different purchasing power. Moreover, the class of measures is applied to assess the decision to change jobs. Results indicate that individuals' economic insecurity significantly increases the probability of changing jobs. The estimated models demonstrate a better fit to the data when measured by the proposed index compared to indices previously suggested in the literature. The second proposal assumes that economic insecurity depends not only on personal economic conditions but also on the comparison with peers' financial situations. A new definition of economic insecurity and a composite inter-temporal class of indices are proposed. This approach combines a longitudinal component and a cross-sectional component that gauges individuals' relative positions compared to their peers. The components' weights are determined through a novel approach incorporating a subjective economic insecurity indicator. The class of composite inter-temporal indices is applied to compare a set of European countries. Results highlight that the index provides new insights into individual perceptions of well-being, not captured by other poverty and inequality measures. Lastly, an empirical study on the effect of economic insecurity on political instability is proposed, comparing different insecurity indicators. The findings show that economic insecurity has a positive relationship with anti-incumbency, which results in unstable government at the aggregate level
Study and development of BaCe0.65Zr0.20Y0.1503-Δ (BCZY) – Gd0.2Ce0.8O2-Δ (GDC) dense ceramic membranes sysyem for high temperature H2 purification
No full textThe first main conclusion drawn from this dissertation concerns the amount of Pt deposited on the asymmetric layer of membrane produced by tape casting porosity shaping method. Three different amounts were investigated (0.15, 1.5 and 4.5 mg cm-2 ). The most optimal performance, based on H2 permeation performances, was attained when 1.5 mg cm-2 of Pt was deposited on the porous layer, resulting in a 0.642 mL min-1 cm-2 permeated H2 when 80% H2 in He was employed as the feed. Pt deposition method is influenced by the concentration of the Pt precursor, which results in different morphology of the catalyst. The second development focused on further optimization on tape casting membranes concerning the solvent employed for the Pt catalyst deposition. The same concentration of Pt was employed, depositing 1.5 mg cm-2 on the porous side of the membrane, but a mixture of acetone and water was employed as solvent. This mixture allowed the suppression of effects leading to poorly dispersed particles. As a result, it was possible to achieve 0.74 mL min-1 cm-2 at 750°C with 50% H2 in He. Lastly, first-ever permeation performance measurements into an innovative ceramic membrane type for hydrogen separation was investigated. In-depth research was done on a group of hierarchically-structured BaCe0.65Zr0.20Y0.15O3-δ(BCZY) - Gd0.2Ce0.8O2-δ(GDC) membranes produced by freeze casting porosity shaping method. Membranes were investigated observing the effect of deposition solvent and the effect of porous layer thickness. Employing a mixture of Acetone and water resulted in better hydrogen permeation at temperatures (T > 650°C), reaching 0.26 mL min-1 cm-2 at 750°C with 50% H2 in He. The reduction of porous layer thickness led to a hydrogen flow of 0.33 mL min-1 cm-2 , at 750°C with 50% H2 in He
Organic electrochemistry: mechanisms and synthetic applications
No full textOrganic synthesis is essential for developing economical, sustainable, and environmentally compatible reaction pathways, which are crucial in medicinal chemistry, pharmaceutical development, and materials science. Electrochemistry, introduced by Faraday in the 19th century, played a pivotal role in organic chemistry but lost prominence in the 20th century, largely replaced by dipolar chemistry characterized by complex and less sustainable processes. With the growing focus on sustainability, electrochemistry has experienced a resurgence, enabled by recent technological advancements that offer more efficient, selective, and sustainable synthetic methods. This PhD thesis investigates electrochemistry as a tool for achieving organic syntheses that are challenging with traditional methods, encompassing three main projects.
Project I: The study of NHC-Ag(I) complexes as catalysts in the Borono-Minisci reaction focuses on the functionalization of N-heterocycles, essential in pharmaceuticals. These complexes enhance catalytic efficiency and are easily recyclable. Using cyclic voltammetry, the catalysts were characterized, and the Ag(I)/Ag(II) redox potentials were analyzed, revealing the relationship between complex structure and reaction yield.
Project II: A novel electrochemical methodology for the side-chain decarboxylation of Asp and Glu was developed, enabling the synthesis of unnatural amino acids (UAAs) with electron-rich heteroaromatics. These UAAs are critical for protein research and pharmaceutical applications. The optimized method demonstrated high versatility in synthesizing complex UAAs.
Project III: An innovative electrophotocatalytic method for direct amide bond formation utilizes FeCl3 as a catalyst, activated through hydrogen atom transfer (HAT) photocatalysis and radical-polar crossover mechanisms. This process eliminates the need for sacrificial oxidants, resulting in a sustainable, selective, and efficient approach for late-stage functionalization of complex molecules. Its scalability in flow systems enhances productivity, making it highly promising for industrial applications, particularly in pharmaceutical synthesis
Genomics, oncology, one-health approach, and precision medicine using next generation sequencing technologies
No full textTraditional diagnostic methods in veterinary medicine often lack the ability to detect genetic variations and fully elucidate the pathogenesis of complex diseases, especially those involving rapidly evolving pathogens and genetic mutations. Next-Generation Sequencing (NGS) technologies have revolutionized the field by providing precise, high-throughput capabilities for disease diagnosis, pathogen identification, and genetic research. This thesis explores the application of NGS in improving diagnostic accuracy and developing precision medicine strategies across various veterinary conditions. Key studies include the analysis of gastrointestinal nematodes and microbiome diversity through metagenomics (16S rRNA sequencing), the identification of genetic mutations responsible for limb-girdle muscular dystrophy in Lagotto Romagnolo dogs, and, finally, the investigation of neoangiogenesis and biomarkers in canine urothelial carcinoma. Using metagenomic sequencing, we identified a diverse range of microbial taxa, shedding light on the complex interactions between gut microbiota and parasite load in equine gastrointestinal health. Whole Exome Sequencing (WES) uncovered genetic mutations linked to muscular dystrophy in dogs, facilitating early diagnosis and enabling potential therapeutic interventions. Additionally, the study of canine urothelial carcinoma revealed novel insights into neoangiogenesis and biomarker expression, offering new prognostic tools for treatment stratification. Collectively, these findings detailed the transformative potential of NGS in veterinary diagnostics, enabling more accurate disease identification, earlier therapeutic interventions, and enhanced understanding of host-pathogen interactions. Integrating advanced sequencing technologies into routine veterinary practice holds the promise of improving animal health, productivity, and genetic conservation, with broader implications for public health and food security
Real-time analysis of gamma-ray transients: developing software and statistical methods for the CTAO
No full textThe Cherenkov Telescope Array Observatory (CTAO) is going to be the leading observatory for very-high-energy astrophysics in the next decades. It is going to be a powerful instrument for the exploration of the violent, variable Universe thanks to its unprecedented sensitivity to short timescale phenomena. The CTAO will be equipped with a real-time-analysis pipeline, the Science-Alert-Generation (SAG) pipeline, which will perform scientific monitoring of the observed field of view and generate candidate Science Alerts when detecting transient phenomena.
We developed the scientific pipeline of SAG, SAG-SCI, that performs high-level analysis of gamma-ray data using traditional analysis methods. We developed SAG-SCI respecting CTAO software quality requirements and scientific strategies to optimise the analyses on different types of targets and generate candidate Science Alerts. We used SAG-SCI to analyse CTAO simulated data under several observational scenarios, to validate the software algorithms, investigate systematic uncertainties and present an overview of the most relevant applications of SAG-SCI.
When the analyses must be performed in a low-photon-statistics regime, the traditional methods are not valid, thus we developed a Burst Analysis statistically valid in this regime. We applied the Burst Analysis to observations of SGR 1935+2154, a Galactic magnetar, performed with the CTAO Large-Sized Telescope Prototype, LST-1. We searched for possible TeV counterpart of its X-ray persistent and burst emission. While neither persistent nor transient emission was detected in our search, we could provide, for the first time, TeV upper limits to the emission of a short magnetar burst simultaneous to its soft gamma-ray signal.
Our development of SAG-SCI will allow the CTAO to become a fundamental facility for time-domain astrophysics, and enable multi-wavelength and multi-messenger approaches leading to a deeper understanding of the Universe. Our development of the Burst Analysis will allow CTAO to analyse several interesting classes of fast transients in the low-photon-statistics regime
Like a river. Adolescence and Youth Fiction: portraits of metamorphic identities
No full textLa ricerca indaga le rappresentazioni di adolescenza nella Youth Fiction contemporanea, prendendo avvio da un problema di ricerca che si riferisce alla possibilità di individuare una specificità adolescenziale in letteratura, ovvero al recupero di indizi letterari riferiti all’adolescenza come stato dell’essere caratterizzato da un sentire e da un percepirsi-nel-mondo differenti rispetto alle condizioni delle altre età della vita umana. Il suddetto problema di ricerca prende a sua volta le mosse dalla rilevazione dell’esistenza di un immaginario sociale crisiologico e patologizzante costruito attorno al concetto di adolescenza (Barone 2005, 2009), un immaginario permeato da metafore quali “l’adolescenza è un problema” e “l’adolescenza è una malattia”. A partire da un problema di ricerca così formulato, l’indagine si interroga su quali rappresentazioni dell’essere adolescente nel contesto attuale dell’Occidente fornisca la letteratura Youth Fiction contemporanea, proponendosi di indagare se tale letteratura delinei un’alterità dell’adolescenza che non sia posta unicamente in termini di problema, di disagio e di malattia. Inscritto nella cornice teorica del problematicismo pedagogico (Bertin 1951), questo studio ricostruisce la storia, l’evoluzione e gli sviluppi contemporanei della letteratura Young Adult (genere letterario all’interno del quale si colloca la Youth Fiction), per quindi accostarsi alle singole opere di narrativa mediante un paradigma indiziario (Ginzburg 2023; Faeti 2001). La ricerca individua, così, indizi decifratori della condizione adolescenziale, ovvero elementi narrativi e formali ricorrenti, costanti della Youth Fiction contemporanea che fungono da metafore di adolescenza, suggerendone una nuova leggibilità operante nella direzione di uno smantellamento di vecchi pregiudizi. A sostenerne l’analisi e interpretazione sono, in particolare, gli studi sulla letteratura per adolescenti (Chambers 2020; Trites 2000, 2014), la linguistica cognitiva (Lakoff e Johnson 2022), la semiotica interpretativa (Eco 1980, 2005, 2016) e gli approcci di seconda generazione allo studio cognitivo della letteratura (Herman 2000, 2013, 2014; Newen 2018; Malafouris 2018; Silva 2022, 2024).The research investigates the representations of adolescence in contemporary Youth Fiction, focusing specifically on the literature which has been published from the 1990s to the present. The research problem refers to the possibility of identifying literary clues related to adolescence as a state of being characterized by its own way of feeling. This problem in turn derives from the detection of a pathologizing social imagery around the concept of adolescence (Barone 2005, 2009), an imaginary permeated by metaphors like “adolescence is a problem” and “adolescence is an illness”. The research thus questions what representations of being an adolescent in Western societies are provided by contemporary Youth Fiction, aiming to investigate whether this literature depicts a diversity of adolescence which is not only presented in terms of crisis, unease, and problems. Inscribed in the theoretical framework of pedagogical problematicism (Bertin 1951), this study reconstructs the history, evolution, and contemporary developments of Young Adult Literature (the literary genre of which Youth Fiction is part), to then approach the single youth novels by means of close readings. Using the lens of the evidential paradigm (Ginzburg 2023; Faeti 2001), the study identifies recurring narrative and formal elements which act as metaphors for adolescence and suggest a new readability of this stage of life. The analysis and interpretation of these metaphors are informed by an interdisciplinary framework based on Youth Fiction studies (Chambers 2020; Trites 2000, 2014), cognitive linguistics (Lakoff e Johnson 2022), interpretative semiotics (Eco 1980, 2005, 2016), cognitive narratology integrated with the 4E approaches to cognition (Herman 2000, 2013, 2014; Newen 2018; Silva 2022, 2024), and material engagement theory (Malafouris 2018)
Transfer learning for neural network surrogates of flow and transport
No full textThis thesis introduces a multi-fidelity approach to train deep neural network surrogates for efficient and accurate modeling of high-dimensional fluid dynamic systems. Due to their nonlinear, dynamic, and multi-scale nature, these complex physical phenomena require substantial computational resources. While various surrogate modeling techniques exist to help alleviate this computational burden, constructing data-driven surrogates requires several thousand high-fidelity simulations to generate adequate training samples. To tackle this challenge, we present a framework that leverages multi-fidelity simulations to reduce data generation costs and employs transfer learning to train deep Convolutional Neural Networks (CNNs). We first explore the possibility of using multi-model data sources to train a CNN to replicate blood flow dynamics within an aorta geometry while capturing the variability of the governing parameters across individual patient-specific cases. Then, we investigate the use of transfer learning on two levels of data to train an inverse CNN to solve high-dimensional inverse problems in subsurface hydrology, i.e., heterogeneity field reconstruction and contaminant source identification. Our approach optimally balances computational speed-up and predictive accuracy where traditional high-fidelity models are computationally prohibitive. To mitigate the effects of the curse of dimensionality, which can impact other model reduction techniques such as Polynomial Chaos Expansion (PCE), we propose a multi-fidelity framework combined with global sensitivity analysis as a dimensionality reduction method. This approach aims to extend the applicability of the PCE technique as an alternative to neural networks in applications with a reasonable number of parameters
Nanofibrous-based composite materials for energy harvesting and storage applications
No full textThis study presents the development and characterization of advanced nanofibrous materials for energy harvesting and storage, with a focus on piezoelectric nanofibers and innovative separators for lithium-ion batteries (LIBs). Electrospun P(VDF-TrFE) nanofibrous mats were fabricated and subjected to various poling techniques to enhance their macroscale piezoelectric response through dipole alignment. Among the methods investigated, AC contact poling and corona poling were the most effective, yielding a β-phase content of 94% and piezoelectric strain coefficients (d33) of -27.2 ± 0.6 pC/N and -20.8 ± 1.7 pC/N, respectively—significantly surpassing conventional DC poling. These nanofibers were integrated into a piezoelectric energy harvesting system coupled with a pullulan-based ionic liquid micro-supercapacitor, achieving efficient energy conversion and storage under mechanical stimuli, with a maximum energy output of 211 mJ over 5 hours. This modular integration offers enhanced design flexibility and energy efficiency, indicating strong potential for wearable electronics. In parallel, novel electrospun PVDF-HFP separators for LIBs were developed and enhanced via plasma treatments and inorganic nanofillers. Plasma pre-treatment improved fiber uniformity, reducing average diameters from 557 ± 161 nm to 493 ± 71 nm, while post-treatment significantly boosted electrolyte uptake to 612%—a sixfold increase over commercial Celgard separators (110%). Incorporating nanofillers such as ZrO₂, SnO₂, and SiO₂ notably improved the mechanical and thermal performance of the separators. ZrO₂, for example, increased the elastic modulus from 141 ± 5 MPa to 237 ± 7 MPa and minimized thermal shrinkage. Electrochemical analyses revealed superior ionic conductivity and charge/discharge performance, with ZrO₂-based separators offering the best overall balance of mechanical, thermal, and electrochemical properties. These findings underscore the promise of piezoelectric nanofibers and nanocomposite LIB separators in next-generation energy systems, providing scalable, high-performance solutions for flexible and wearable electronics
Research on a comprehensive evaluation method for asphalt pavement skid resistance based on tire–pavement contact characteristics and contaminant influence
No full textThis study investigates skid resistance of asphalt pavements through experiments, theory, and simulations, focusing on tire–pavement interaction, surface texture, and contaminant effects. Tire–pavement friction is composed of adhesion, elastic deformation, micro-cutting, and intermolecular forces, with adhesion dominating under dry conditions but greatly reduced by water or pollutants. A high-precision pressure film system was used to analyze tire–pavement stress, revealing that contact area and stress distribution depend on pavement texture, load, tire pressure, and camber angle. A Boltzmann distribution model showed strong correlation between stress distribution and macrotexture depth. Texture analysis using 3D scanning and image processing found that open-graded pavements, coarse aggregates, and deeper, more uniform textures enhance skid resistance, though long-term wear reduces performance until it stabilizes. Hazardous spill tests demonstrated severe impacts: engine oil reduced friction by up to 71% due to lubrication and retention, while gasoline and diesel dissolved asphalt, degrading texture; brake fluid strongly adsorbed to aggregates. Molecular dynamics simulations explained how pollutants disrupt interfacial adhesion. Based on these findings, a predictive evaluation model was developed, integrating contact mechanics, texture features, and contaminant effects. The model accurately predicts friction coefficients under various conditions, providing practical value for skid resistance assessment and contamination risk management