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
Adversarial Bandits and which Leader to Follow
Get PDFL'abstract è presente nell'allegato / the abstract is in the attachmen
Non‐Linear Influence of Reservoir Initial Condition on Flood Reduction
Get PDFReservoirs play a crucial role in modifying natural flow regimes and mitigating flood peaks, yet their effectiveness depends heavily on operational strategies, particularly the initial storage level at the onset of a flood event. This study investigates, for the first time, the non-linear effects of reduced initial storage on the relationship between flood peak attenuation efficiency and flood return period for about 250 large dams in Italy. We estimate flood hydrographs via a simplified hydrological model and apply full hydraulic routing under different scenarios of initial reservoir storage, informed by historical reservoir time series and regional flood seasonality. Our findings reveal that flood peak attenuation is highly sensitive to the initial storage level, with dam performance deteriorating sharply as flood return periods increase, especially when initial storage is high. Seven distinct classes of dams are identified based on their flood attenuation capacity relative to flood severity, highlighting non-linear and threshold effects that are often overlooked in regional dam safety assessments. Notably, the commonly assumed full-reservoir condition yields overly conservative estimates: under this assumption, approximately 20% of the dams reach their maximum allowed water level for return periods of 100 years or less. This national-scale analysis provides new insights into regional differences in reservoir operation, particularly between hydropower-oriented dams in the Alps and water supply reservoirs in southern Italy. By explicitly quantifying how reduced initial storage can enhance flood mitigation, the study offers practical recommendations for optimizing reservoir operations under current and future climatic conditions
Experimental investigation on the effects of CO2 and acid exposure on the hydro-mechanical behaviour of a clayey caprock
Get PDFL'abstract è presente nell'allegato / the abstract is in the attachmen
Viability and control of a delayed SIR epidemic with an ICU state constraint
No full textThis paper studies viability and control synthesis for a delayed SIR epidemic. The model integrates a constant delay representing an incubation/latency time. The control inputs model nonpharmaceutical interventions, while an intensive care unit (ICU) state-constraint is introduced to reflect the healthcare systema's capacity. The arising delayed control system is analyzed via functional viability tools, providing insights into fulfilling the ICU constraint through feedback control maps. In particular, we consider two scenarios: first, we consider the case of general continuous initial conditions. Then, as a further refinement of our analysis, we assume that the initial conditions satisfy a Lipschitz continuity property, consistent with the considered model. The study compares the (in general, sub-optimal) obtained control policies with the optimal ones for the delay-free case, emphasizing the impact of the delay parameter. The obtained results are supported and illustrated, in a concluding section, by numerical examples
Tecnologie immersive e prompting AI:il futuro della rappresentazione visiva e verbale
Get PDFll contributo esplora il concetto di ekphrasis come pratica contemporanea di descrizione generativa, applicata allo spazio digitale della rappresentazione. Attraverso laintersezione tra Disegno e Intelligenza Artificiale (AI), la ricerca analizza come il prompting e le metodologie computazionali traducano descrizioni testuali in rappresentazioni visive complesse, offrendo nuove prospettive teoriche e applicative. LAapproccio proposto integra gemelli digitali, AI e tecnologie immersive di Realta Virtuale (VR) e Aumentata (AR) per ottimizzare laefficienza energetica e la gestione delle risorse. Tre fasi principali guidano il processo: il modello BIM, utilizzato come strumento di comunicazione e gestione dei dati; laelaborazione verbale dei dati tramite AI per migliorare laaccessibilita delle informazioni; la visualizzazione immersiva, che sfrutta VR e AR per arricchire lainterazione con i dati. LAimpiego di assistenti virtuali e interfacce basate sul linguaggio naturale rappresenta una sfida innovativa per rendere il BIM piu accessibile anche agli utenti non esperti. Questa sintesi multidisciplinare dimostra come laekphrasis digitale, nella sua accezione contemporanea, possa generare nuove conoscenze e applicazioni attraverso la convergenza di linguaggi verbali, visivi e computazionali. | The contribution explores the concept of ekphrasis as a contemporary practice of generative description, applied to the digital space of representation. Through the intersection between Drawing and Artificial Intelligence (AI), the research analyzes how prompting and computational methodologies translate textual descriptions into complex visual representations, offering new theoretical and applicative perspectives. The proposed approach integrates digital twins, AI, and immersive technologies such as Virtual Reality (VR) and Augmented Reality (AR) to optimize energy efficiency and resource management. Three main phases guide the process: the BIM model, used as a communication and data management tool; the verbal processing of data through AI to improve information accessibility; immersive visualization, which leverages VR and AR to enrich interaction with data. The use of virtual assistants and interfaces based on natural language represents an innovative challenge to make BIM more accessible even to non-expert users. This multidisciplinary synthesis demonstrates how digital ekphrasis, in its contemporary meaning, can generate new knowledge and applications through the convergence of verbal, visual, and computational languages
Analisi di un Problema di Logistica con l'Intelligenza Artificiale
Get PDFIl nostro lavoro è stato un caso di studio su come un problema pratico di logistica possa essere modellato e risolto utilizzando algoritmi ispirati alla fisica e all'intelligenza artificiale. Il problema è la gestione di uno scaffale che ha n piani per ospitare due scatole su ciascun piano. Le scatole sono
recensione a Arianna Carannante. La cattedrale di Lucera e l’architettura angioina del primo Trecento
No full textFast Determination of Feasible Torque-Speed Range for Variable Flux Machines Including Remagnetization Voltage Limit
No full textThis paper presents a computationally efficient method for determining the feasible operating range of VariableFlux Machines (VFMs) under inverter voltage constraints. Remagnetization transients in VFMs demand substantial current and armature flux, resulting in elevated voltage requirements and pronounced speed limitations. To address this, the proposed approach leverages precomputed Finite Element Analysis (FEA) flux and magnetization maps to define torque speed boundaries under voltage-limited conditions while considering the high voltage demand during remagnetization, eliminating the need for iterative field simulations or FEA co-simulation. The method is validated through driving cycle simulations, demonstrating the reduction in available traction and braking torque due to the inability to re-magnetize the machine beyond the voltagespeed limit. The results provide key insights for optimizing magnetization state control and mission profile planning in traction applications using VFMs
QoT Impairments Induced by Statistical Filtering Variations with a Realistic Equalizer
Get PDFWe analyze filtering penalty in ASE-impaired optical links, including transceiver noise effects. A mathematical model for FSE equalization is validated via time-domain simulation and exploited to perform Monte Carlo analyses, incorporating stochastic variations in optical filters' bandwidth and central frequency