186698 research outputs found
Sort by
1. Scalable CVD Graphene Field-Effect Transistor Platform for Viral Detection: Application to COVID-19
No full textStructural health monitoring of piezoelectric structures by estimating energy conversion efficiency with capacitance measurements
No full textIn structures equipped with piezoelectric materials, the quantification of the efficiency of the energy conversion between mechanical and electrical domain (and vice versa) was already demonstrated to be a reliable feature for determining the presence of a structural damage. Usually, this efficiency is determined by means of vibration measurements to derive the value of the modal electro-mechanical coupling coefficient, which is the damage feature employed to calculate damage indexes. Nevertheless, in some cases, the estimation of this damage feature can be challenging because of different reasons, such as, e.g., non sufficient external excitation to the system, low values of the modal electro-mechanical coupling coefficient which makes its estimation uncertain and implies the need of shunting the piezoelectric elements with tailored electrical impedances. The present paper aims at showing that similar damage indexes can be estimated also by measuring capacitance trend as function of frequency for each piezoelectric element in the structure. This allows monitoring the structure without any need of vibration measurements and addition of external electric circuits used to shunt the piezoelectric elements, with significant simplification of the estimation procedure. Three different damage indexes are proposed, evidencing which of them are the most reliable for detecting the presence of the damage and getting information about its location. The method and the indexes are studied through numerical analyses and validated by means of an experimental campaign on a tailored set-up
On the robustness of adversarial training against uncertainty attacks
No full textIn learning problems, the noise inherent to the task at hand hinders the possibility to infer without a certain degree of uncertainty. Quantifying this uncertainty, regardless of its wide use, assumes high relevance for security-sensitive applications. Within these scenarios, it becomes fundamental to guarantee good (i.e., trustworthy) uncertainty measures, which downstream modules can securely employ to drive the final decision-making process. However, an attacker may be interested in forcing the system to produce either (i) highly uncertain outputs jeopardizing the system's availability or (ii) low uncertainty estimates, making the system accept uncertain samples that would instead require a careful inspection (e.g., human intervention). Therefore, it becomes fundamental to understand how to obtain robust uncertainty estimates against these kinds of attacks. In this work, we reveal both empirically and theoretically that defending against adversarial examples, i.e., carefully perturbed samples that cause misclassification, additionally guarantees a more secure, trustworthy uncertainty estimate under common attack scenarios without the need for an ad-hoc defense strategy. To support our claims, we evaluate multiple adversarial-robust classification models from the publicly available benchmark RobustBench on the CIFAR-10 and ImageNet datasets, and on a robust semantic segmentation model evaluated on Pascal-VOC. The code for the reproducibility of the experiments is available at the following link: https://github.com/pralab/UncertaintyAdversarialRobustness
Feasibility Study of a Brayton-Based High Temperature Heat Pump for Waste Heat Recovery in Industrial Applications
No full textElectrification of thermal users through heat pumps can be a promising way to enhance the exploitation of increasing renewable electrical capacity, offering significant opportunities for decarbonizing the industrial sector. For this purpose, since commercial vapor compression cycles are not readily viable to displace fossil fuel boilers employed in industrial thermal processes, interest is growing toward high temperature heat pumps (supply temperature > 160 degrees C) and, among them, reverse Brayton cycles. This work proposes an innovative Brayton-based open heat pump cycle applied to a relevant industrial case study, with the aim of upgrading the available waste heat to the required process temperature levels. The on-design performance analysis of the reverse Brayton cycle is conducted using the modular in-house tool WTEMP-EVO. Subsequently, a sensitivity analysis is performed on temperature levels, heat sink, and compressor isentropic efficiency. Finally, an off-design model integrating existing machinery with their characteristic curves is developed to evaluate different system operating conditions, as well as possible solutions to improve system rangeability, establishing the groundwork for the implementation of an experimental prototype. Results show that the analyzed cycle can provide heat at temperatures above 200 degrees C with a coefficient of performance higher than 1.5 and a temperature lift of more than 100 degrees C, demonstrating its potential in the industrial sector
Roman Republican Prosopography: the Sources
No full textThis paper examines the evidentiary foundations for prosopographical research on the Roman Republic. It surveys the principal categories of evidence used to reconstruct individual identities, careers, and social relationships, including literary texts, epigraphic material, official lists such as the fasti, and archaeological data. It emphasizes the uneven survival, chronological bias, and ideological agendas that shape these sources, highlighting the methodological problems posed by fragmentation, retrospective reconstruction, and elite self-representation. Particular attention is given to the interaction between narrative historiography and documentary evidence in tracing magistracies, priesthoods, kinship structures, and political networks. The paper also situates traditional source criticism alongside modern scholarly tools, including prosopographical corpora and digital databases, assessing their impact on the field. By clarifying both the potential and the limits of the available evidence, it provides a methodological framework for using prosopography as a critical instrument in the study of Roman Republican political, social, and institutional history
Growth performance, lipid metabolism, gut histoarchitecture and immune and antioxidant related gene expression in juvenile Asian sea bass, Lates calcarifer fed peroxidized lipids with or without dietary selenium nanoparticles
No full textThis study evaluated the effects of dietary recovered frying soybean oil (RFSBO) and selenium nanoparticles (SeNPs) on growth performance, hepatic metabolism, intestinal morphology, and the expression of antioxidant, immune, and growth-related genes in juvenile Asian sea bass (Lates calcarifer, 41.5 ± 0.1 g) reared under high temperature (32–33 °C) and high salinity (38–40 ppt). Six diets were formulated: fresh soybean oil (FSBO), FSBO + SN (4 mg/kg SeNPs), 50 % RFSBO, 50 % RFSBO + SN, 100 % RFSBO, and 100 % RFSBO + SN. Fish (n = 450) were randomly assigned to 18 tanks and fed to apparent satiation three times daily for eight weeks. Fish fed 50 % RFSBO + SN achieved similar final weights to the FSBO group but with significantly better feed conversion ratio, improved gut wall, epithelial, and villus height, and lower malic enzyme activity, indicating reduced metabolic stress. Hepatic triglycerides were significantly lower in this group than in FSBO-fed fish, while glycogen content was maintained. In contrast, 100 % RFSBO caused histological damage, oxidative stress, elevated isocitrate dehydrogenase activity, and lipid imbalance, with SeNPs offering only partial mitigation. SeNP supplementation upregulated gpx1, lyz, il-1β, and igf1 expression under moderate oxidative stress but had limited effects under severe conditions. Overall, RFSBO can replace up to 50 % of dietary FSBO without compromising growth or intestinal health when combined with SeNPs, but higher levels reduce SeNP efficacy. These findings support the use of moderate RFSBO inclusion with SeNP supplementation to sustain fish health and performance under challenging environmental conditions
BEWT: A Benchmark for End-to-End Web Testing
No full textWeb applications are critical to modern life and require rigorous End-to-End (E2E) testing to ensure reliability across front-end and back-end components. While recent work has improved E2E testing—reducing cost, flakiness, and increasing robustness—a common benchmark is missing, hindering fair comparison and progress. This work introduces the first E2E benchmark dataset to address that gap: 12 Selenium WebDriver test suites for 8 web applications, packaged in Docker for easy deployment. It supports test evolution, automation, and flakiness studies, offering 389 Gherkin-based test cases, 283 Page Objects, 1,364 locators, and over 19k lines of code. By providing a reproducible, diverse foundation, this benchmark enables consistent evaluation of testing techniques and fosters advancement in E2E testing research