Higher Institute on Territorial Systems for Innovation
PORTO@iris (Publications Open Repository TOrino - Politecnico di Torino)Not a member yet
146173 research outputs found
Sort by
Antropocene o tecnocene? Finalità della tecnica ed effetti socio-economici e ambientali
Get PDFIl sapere tecnico accompagna gli esseri umani fin dai primordi costituendo un indubbio vantaggio evolutivo per la nostra specie. Negli ultimi decenni tuttavia l’evoluzione tecnologica, accelerata dalla rivoluzione informatica e digitale, attraverso la globalizzazione dei mercati ha mutato forma e fini, con conseguenze preoccupanti sulla tenuta della Biosfera terrestre, per il crescente intreccio tra economia, innovazione tecnologica e industria bellica e per la tenuta delle istituzioni democratiche occidentali. Il successo di una tecnologia appare essenzialmente determinato dalle logiche del profitto e del mercato, in una spirale distruttiva finalizzata alla crescita senza limiti. Tecnologia ed economia paiono condividere gli stessi mezzi e fini
Supporting Lean Analysis Through IIoT in Digitalized Human-Robot Collaborative Environments
No full textControl of networked cyber–physical–human systems
No full textCyber–physical–human systems (CPHSs) — characterized by the organic integration of physical components, a computation and communication cyber layer, and humans — are becoming an integral part of daily life, with applications ranging from assistive robots to smart buildings and modern logistics systems. The key role of humans in these complex systems calls for a paradigm shift, from classical machine-oriented control methods to approaches that focus on the explicit modelling and control of the human layer of a CPHS. In this Review, we showcase state-of-the-art research in mathematical modelling and control of CPHSs. We discuss established control-theoretic approaches to model and control cyber–physical systems, explore two mathematical approaches to modelling human behaviour and social influence, utilizing tools from game theory and opinion dynamics, and show how these models are integrated into CPHSs. Moving to control, we focus on the potential and the challenges that are associated with the human layer. We present approaches for control on different layers, paying attention to the fact that humans can typically only be guided or nudged, and then discuss control across layers. Finally, we describe some major open questions in controlling CPHSs, including integrating data-driven approaches and bridging the gap between theoretical and experimental studies
A fluid–structure interaction framework for mechanical aortic valves: analyzing the effects of valve design and aortic curvature on hemodynamics
Get PDFAortic mechanical heart valves (MHVs) have been implanted for decades to treat aortic valve disease and remain a viable option when valve durability is prioritized. However, the non-physiological hemodynamics induced by MHVs may lead to adverse clinical outcomes. Fluid–structure interaction (FSI) simulations enable the analysis of the biomechanical interaction between MHVs and blood flow. This study presents a strongly coupled, boundary-fitted FSI framework for aortic MHVs, used to assess the impact of MHV design and aortic curvature on hemodynamics. Nine simulation scenarios were investigated, considering three commercially available MHVs and three idealized aortic geometries (one straight and two curved models). Overall, the framework proved to provide results for flow-rate waveforms, velocity fields, and leaflet kinematics aligning well with previous experimental and computational studies. The framework highlighted that: (i) MHV design influences velocity fields and large-scale vorticity transport in the aorta, with systolic differences among the three devices of up to 41% and 133% in average swirling strength and stretching, respectively; (ii) the straight aortic model underestimates systolic swirling strength (up to 56%) and stretching (up to 91%) compared to curved models. This FSI framework can support MHV development by analyzing different device designs and anatomical scenarios
Sky cooling-driven radiant-capacitive hydronic system for all-day building cooling
Get PDFDaytime Radiative Cooling (DRC) surfaces enable heat rejection by emitting infrared radiation to the sky while reflecting solar radiation, allowing for sub-ambient cooling even under direct sunlight. This study develops and validates a transient numerical model of a DRC-based hydronic cooling system designed for building applications. The system integrates ceiling-mounted radiant capacitive modules (RCMs) with outdoor sky radiators (SRs) that dissipate indoor heat to outer space, cooling down a heat transfer fluid. The model is validated using experimental data from a full-scale demonstrator using a commercially available DRC emitter and is employed to assess system performance for a single-family building during a typical cooling season in the cities of Madrid and Rome. Compared to a system limited to nighttime radiative cooling, the DRC-enhanced setup delivers seasonal energy performance improvements of +6.2 % with a commercial DRC material and +10.3 % with an ideal broadband emitter. The study further investigates the effects of varying the surface area ratio between SRs and RCMs and alternative SR placements (rooftop vs. external surface). A fully passive building model with a DRC roof is also considered for comparison. Results show that the DRC-hydronic system can consistently maintain indoor thermal comfort throughout the cooling season, achieving seasonal energy efficiency ratios (SEER) up to 35 times higher than those of conventional air conditioning systems for the case studies analyzed, although the two systems differ in controllability and application scenarios. These findings highlight the strong potential of DRC-integrated hydronic cooling as a highly energy-efficient and sustainable alternative for the climate control of residential buildings
A novel prior-informed machine learning model for diesel engine emission estimation
Get PDFThe increasing complexity of internal combustion engines, coupled with stringent emission regulations, has made virtualization essential for efficient and effective exploration of engine design spaces. This evolution demands predictive models that balance accuracy with computational efficiency, particularly when applied to emission estimation tasks. Data-driven approaches have been widely adopted in this field due to their flexibility and strong predictive capabilities. However, achieving high accuracy with these methods typically requires large training datasets. To overcome this limitation, this study explores the application of Gradient-Informed Neural Networks (GradINNs) for diesel-engine emission prediction. GradINN combines a primary neural network, responsible for the emission estimates, with an auxiliary network that encodes prior beliefs about the gradients of the output with respect to the model’s input parameters. A specialized loss function enforces consistency between the predicted gradients and these prior beliefs. The proposed model is benchmarked against traditional data-driven approaches, specifically Neural Networks (NNs) and Gaussian Process Regression (GPR), using data from both a Design of Experiments (DoE) campaign and an engine map. Results demonstrate that GradINN consistently outperforms both benchmark methods across key emission targets, including nitrogen oxides, particulate matter, unburned hydrocarbons, and carbon monoxide. The proposed approach achieves lower prediction errors and improved generalization, notably maintaining comparable accuracy with up to 25 % fewer training samples compared to the best-performing benchmark, highlighting its potential to reduce experimental effort without compromising accuracy
Just Green Transitions - An Introduction
No full textJust Green Transitions (JGTs) address the shift to sustainable, low-carbon economies while ensuring fairness and inclusivity. Central to global policies like the Paris Agreement and the European Green Deal, JGTs remain multidimensional, diverse, and complex in their interpretation and application. This handbook explores JGTs with a focus on the Western Balkans (WB), examining their potential to address regional disparities, socio-economic challenges, and governance gaps. Highlighting the EU’s Just Transition Mechanism and Green Agenda for the Western Balkans (GAWB), it showcases strategies and case studies demonstrating the interplay of technical, economic, social, and environmental dimensions. Providing actionable frameworks and offering insights for policymakers and practitioners to design inclusive and context-specific transitions. The handbook emphasises collaboration, citizen engagement, and adaptive governance as essential for equitable and resilient green futures
Functionalized carbon nanotubes tape for energy harvesting from salinity gradients through asymmetric capacitive mixing
Get PDFIn this work, we investigated the possibility of using a commercial tape entirely made of continuous carbon nanotubes (CNTs) for blue energy harvesting application. The tape was used to build the electrodes of a device harvesting energy from salinity gradient based on the capacitive mixing (CapMix) technique. The tape was used as it is or functionalized to enhance its storage properties and to obtain an asymmetric device. The electrodes underwent a full set of electrochemical characterizations to test the impact of the functionalization. Thanks to high electrical conductivity, remarkable specific capacitance and good chemical stability, the tape acted both as active material and current collector, eliminating the need for metallic current collectors, reducing the mass of the system and avoiding possible corrosion of the metals due to close contact with saline solutions. This approach provided a simple and easily scalable device able to produce electrical power from the mixing of two solutions at different salinities. The achieved power output stands at 75 μW m−2 in artificial seawater/freshwater and 1.2 mW m−2 in artificial Mediterranean brine/seawater. These results contribute to the broader understanding of energy harvesting from salinity gradients, extending the technological application of CNT tape across the renewable energy field
SlideInspect: From Pixel-Level Artifact Detection to Actionable Quality Metrics in Digital Pathology
Get PDFThe presence of artifacts in whole slide images (WSIs), such as tissue folds, air bubbles, and out- of-focus regions, can significantly
impact WSI digitization, pathologists' evaluation, and the accuracy of downstream analyses. We present SlideInspect, a novel
AI-based framework for comprehensive artifact detection and quality control in digital pathology. Our system leverages deep
learning techniques to segment multiple artifact types across diverse tissue types and staining methods. SlideInspect provides a
hierarchical output: a color-coded slide quality indicator (green, yellow, red) with recommended actions (no action, re-scan, re-
mount, re-cut) based on artifact type and extent, and pixel-level segmentation masks for detailed analysis. The system operates at
multiple magnifications (1.25× for tissue segmentation, 5× for artifact detection) and also incorporates stain quality assessment
for histological stain evaluation. We validated SlideInspect on a large, multi-centric, multi-scanner dataset of over 3000 WSIs,
demonstrating robust performance across different tissue types, staining methods, and scanning platforms. The system achieves
high segmentation accuracy for various artifacts while maintaining computational efficiency (average processing time: 72.7 s per
WSI). Pathologist evaluations confirmed the clinical relevance and accuracy of SlideInspect's quality assessments. By providing
actionable insights at multiple levels of granularity, SlideInspect significantly improves the efficiency and standardization of
digital pathology workflows. Its vendor-agnostic design and multi-stain capability make it suitable for integration into diverse
clinical and research settings
Analytical Assessment of Pre-Trained Prompt-Based Multimodal Deep Learning Models for UAV-Based Object Detection Supporting Environmental Crimes Monitoring
Get PDFIllegal dumping poses serious risks to ecosystems and human health, requiring effective and timely monitoring strategies. Advances in uncrewed aerial vehicles (UAVs), photogrammetry, and deep learning (DL) have created new opportunities for detecting and characterizing waste objects over large areas. Within the framework of the EMERITUS Project, an EU Horizon Europe initiative supporting the fight against environmental crimes, this study evaluates the performance of pre-trained prompt-based multimodal (PBM) DL models integrated into ArcGIS Pro for object detection and segmentation. To test such models, UAV surveys were specially conducted at a semi-controlled test site in northern Italy, producing very high-resolution orthoimages and video frames populated with simulated waste objects such as tyres, barrels, and sand piles. Three PBM models (CLIPSeg, GroundingDINO, and TextSAM) were tested under varying hyperparameters and input conditions, including orthophotos at multiple resolutions and frames extracted from UAV-acquired videos. Results show that model performance is highly dependent on object type and imagery resolution. In contrast, within the limited ranges tested, hyperparameter tuning rarely produced significant improvements. The evaluation of the models was performed using low IoU to generalize across different types of detection models and to focus on the ability of detecting object. When evaluating the models with orthoimagery, CLIPSeg achieved the highest accuracy with F1 scores up to 0.88 for tyres, whereas barrels and ambiguous classes consistently underperformed. Video-derived (oblique) frames generally outperformed orthophotos, reflecting a closer match to model training perspectives. Despite the current limitations in performances highlighted by the tests, PBM models demonstrate strong potential for democratizing GeoAI (Geospatial Artificial Intelligence). These tools effectively enable non-expert users to employ zero-shot classification in UAV-based monitoring workflows targeting environmental crime