1,722,887 research outputs found
Convegno internazionale: "Le Marche e il mare. Arte, architettura, paesaggio"
No full textIl convegno internazionale "Le Marche e il mare. Arte, architettura, paesaggio", a cura di Giuseppe Bonaccorso, Claudio Castelletti e Federico Bulfone Gransinigh con la collaborazione di Flavio Stimilli, aspirava a gettare nuova luce sulla fenomenologia culturale del grande tema del mare nelle Marche, dall’Antichità agli anni 2000, dedicando particolare attenzione scientifica alle interpretazioni degli artisti e alle soluzioni degli architetti nell’ampio contesto storico, geografico e paesaggistico del Medio Adriatico
Receding horizon control for water resources management
No full textIntegrated water resources management (IWRM) is recognized worldwide as the reference paradigm to meet society's long-term needs for water resources while maintaining essential ecological services and economic benefits. In previous publications [A. Castelletti, R. Soncini-Sessa, A procedural approach to strengthening integration and participation in water resource planning, Environmental Modelling & Software 21(10) (2006) 1455 1470; A. Castelletti, F. Pianosi, R. Soncini-Sessa, Integration, participation and optimal control in water resources planning and management, Applied Mathematics and Computation, (2007), doi: 10.1016/j.amc.2007.09.069], the authors have already insisted on the need for a procedural approach to make the IWRM paradigm truly operational; they have emphasized the role played by dynamic optimization in rationalizing and facilitating the selection by the decision maker of a best compromise planning alternative. When planning alternatives also include management policies, as in the case of the water reservoir networks considered in this paper, the best compromise off-line policy resulting from the planning exercise has to be actually implemented in the daily management of the system. Here, again, dynamic optimization may play a central role, as it can be adopted on-line to improve the performance of the off-line policy by exploiting any new useful information available in real-time (e. g., inflow predictions, a power station being temporarily out of service, etc.). In this paper, this approach is explored through a real-world case study of a simple reservoir system. The off-line management policy computed in a previous planning process is refined on-line with a receding horizon control scheme combined with an inflow predictor. The results yield indications that the approach can provide significant advantages to cope with extreme events, particularly those occurring in unusual periods of the year. (C) 2008 Elsevier Inc. All rights reserved.</p
An evaluation framework for input variable selection algorithms for environmental data-driven models
Full text linkAbstract not availableStefano Galelli, Greer B. Humphrey, Holger R. Maier, Andrea Castelletti, Graeme C. Dandy, Matthew S. Gibb
Physical processes in supernova remnants and in their interaction with the interestelar medium
No full textFil: Castelletti, Gabriela. Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales; Argentina
Social surveys to support personal experience in human-water-climate change interactions. A review on farmers' behavior
Full text linkWater resources management and climate change represent two necessarily interdisciplinary topics in which the natural and social sciences must be integrated [1]. Although this nexus was generally overlooked in the accurate statistics and modelling literature by mostly focusing on understanding the natural processes, a paradigm shift is required to put social in the modelling loop [2]. Consequently, water domains (physical, social, political, and symbolic matters) should be entwined in research configurations by considering social learning, personal experience, observations, and human choices. As argued by [3], deepen social perception is fundamental for two main reasons: as a key component of the socio-political context and as the first step for behaviour transformation and attitude change. In this line, social and behavioural sciences have discussed associative processing methods, such as social surveys, to monitor the nature, extent, significance, and influence of personal experience regarding human-nature interactions [4]. Farmers develop their activity supporting the complexity of interrelated nature and human systems characterized by biophysical conditions and social behaviour [5]. Consequently, farmers are in a favourable position to provide first-hand observations and narratives of water resources availability and climate change perceived impacts [6]. Could social surveys contribute to deepening farmers’ behaviour on water supply and climate change impacts while providing new social scenarios to advance understanding of data-mining, processing, and modelling of human-water systems? This contribution provides an upgraded and comprehensive overview of the social surveys added-value in building a methodological approach and defining an intellectual structure to monitoring farmers’ behaviour on water-climate change nexus. The literature review will provide new insides to be discussed for policy formulation and implementation at the local and the regional scale.
[1] G. Escribano-Francés, P. Quevauviller, E. San Martín González, and E. Vargas Amelin, Environmental Science and Policy 69, 1 (2017)
[2] M. Giuliani, A. Castelletti, and C. Gandolfi, Water Resources Research 52: 6928 (2016)
[3] L. Antronico, R. Coscarelli, F. De Pascale, and D. Di Matteo, Sustainability 12: 6985 (2020)
[4] J.R. Marlon, S. van der Linden, P. Howe, A. Leiserowitz, S.H.L, Woo, and K. Broad, Journal of Risk Research 22: 936 (2018)
[5] M. Abid, J. Scheffran, U.A. Schneider, and E. Elahi, Environmental Management 63: 110 (2019)
[6] K. Talanow, E.N. Topp, J. Loos, and B. Martin-Lopez, Journal of Rural Studies 81: 203 (2021
Protohistoire de l'Europe
No full textKruta Venceslas, Castelletti Lanfredo. Protohistoire de l'Europe. In: École pratique des hautes études. Section des sciences historiques et philologiques. Livret-Annuaire 20. 2004-2005. 2006. pp. 97-102
Bayesian Model Selection of Gaussian Directed Acyclic Graph Structures
No full textDuring the last years, graphical models have become a popular tool to represent dependencies among variables in many scientific areas. Typically, the objective is to discover dependence relationships that can be represented through a directed acyclic graph (DAG). The set of all conditional independencies encoded by a DAG determines its Markov property. In general, DAGs encoding the same conditional independencies are not distinguishable from observational data and can be collected into equivalence classes, each one represented by a chain graph called essential graph (EG). However, both the DAG and EG space grow super exponentially in the number of variables, and so, graph structural learning requires the adoption of Markov chain Monte Carlo (MCMC) techniques. In this paper, we review some recent results on Bayesian model selection of Gaussian DAG models under a unified framework. These results are based on closed-form expressions for the marginal likelihood of a DAG and EG structure, which is obtained from a few suitable assumptions on the prior for model parameters. We then introduce a general MCMC scheme that can be adopted both for model selection of DAGs and EGs together with a couple of applications on real data sets
Learning Bayesian networks: a copula approach for mixed-type data
No full textEstimating dependence relationships between variables is a crucial issue in
many applied domains, such as medicine, social sciences and psychology. When
several variables are entertained, these can be organized into a network which
encodes their set of conditional dependence relations. Typically however, the
underlying network structure is completely unknown or can be partially drawn
only; accordingly it should be learned from the available data, a process known
as structure learning. In addition, data arising from social and psychological
studies are often of different types, as they can include categorical, discrete
and continuous measurements. In this paper we develop a novel Bayesian
methodology for structure learning of directed networks which applies to mixed
data, i.e. possibly containing continuous, discrete, ordinal and binary
variables simultaneously. Whenever available, our method can easily incorporate
known dependence structures among variables represented by paths or edge
directions that can be postulated in advance based on the specific problem
under consideration. We evaluate the proposed method through extensive
simulation studies, with appreciable performances in comparison with current
state-of-the-art alternative methods. Finally, we apply our methodology to
well-being data from a social survey promoted by the United Nations, and mental
health data collected from a cohort of medical students
Balassi Szép magyar komédiája és olasz forrása, Castelletti Amarilli című pásztorjátéka
No full textCristoforo Castelletti a 16. századi Rómában élt, és egyházi pályafutása mellett olasz nyelvű verseket és színdarabokat, elsősorban komédiákat írt. Egyetlen pásztorjátéka, az Amarilli- komédiáihoz hasonlóan- rendkívül népszerű volt a maga korában. Többször megjelent nyomtatásban, Castelletti kétszer jelentősen átdolgozta, az utolsó változatot pedig a szerző halála után is számos alkalommal publikálták a 17. század első évtizedeiben. Később elfeledték, csupán egyetlen komédiájának jelent meg modern kritikai kiadása, elsősorban nyelvészeti szempontok miatt, az egyik szereplő ugyanis római nyelvjárásban beszél. Tasso Amintájára emlékeztető pásztorjátéka azonban különös jelentőséggel bír a magyar irodalomtörténet számára, hiszen ez Balassi Bálint egyetlen drámai művének elsődleges forrása. Kötetünkben az Amarilli és a Szép magyar komédia egymás mellett szerepel, hogy jól követhetőek legyenek Balassi változtatásai és egyéni újításai Castelletti pásztorjátékához képest. Az olasz szöveg megértését magyar nyersfordítás könnyíti meg az olvasóközönség számára
Feature Representation Learning in complex water decision making problems
Full text linkIl successo di una politica di controllo dipende fortemente dalla sua rappresentazione, ovvero dall'insieme di variabili con cui è informata.
In problemi di controllo nel mondo reale, la definizione di una rappresentazione appropriata è un compito complesso, data la coesistenza di più processi interagenti la cui rilevanza per il problema di controllo è spesso poco chiara.
In questa tesi, affrontiamo il problema di controllo dei sistemi di risorse idriche, in cui una politica di rilascio della diga è progettata tenendo conto di molteplici domande idriche. Questo problema decisionale è complicato dalla presenza di non linearità, forti disturbi, possibili formulazioni alternative del problema, e molteplici obiettivi contrastanti.
Attualmente, le regole di controllo della maggior parte dei bacini idrici sono condizionate su sistemi informativi basilari che considerano l'invaso del serbatoio e un indice del tempo, d'altra parte, il valore di una rappresentazione della politica più ricca e informativa è generalmente indiscusso.
Sfruttiamo i recenti progressi nel monitoraggio e nella previsione della disponibilità di acqua per sviluppare nuove strategie di apprendimento della rappresentazione della politica per migliorare la resilienza dei sistemi idrici rispetto a vulnerabilità cruciali tra cui siccità , fasi critiche nello sviluppo di serbatoi (ad esempio costruzione e riempimento), e conflitti tra diversi settori.
Inoltre, nei sistemi caratterizzati da molteplici usi della risorsa idrica, diversi obiettivi di controllo potrebbero essere eterogenei nelle loro dinamiche e vulnerabilità, e dunque trarre vantaggio da una rappresentazione delle caratteristiche su misura che varia a seconda dei diversi obiettivi.
Analizziamo la letteratura recente sull'apprendimento della rappresentazione della politica, e proponiamo una tassonomia che comprende approcci a priori, a posteriori, e online.
Per ogni approccio, proponiamo contributi metodologici originali, mirati al problema del controllo dei sistemi idrici a molti-obiettivi.
Tra i contributi metodologici inclusi in questa tesi, (1) proponiamo FRIDA, una procedura basata sull'estrazione di variabili per progettare indici di siccità di bacino su misura, e (2) impieghiamo l'indice FRIDA per informare le operazioni di gestione di una diga; (3) estendiamo i concetti di apprendimento della rappresentazione della politica oltre applicazioni di puro controllo a un problema di pianificazione e riempimento di un serbatoio idrico artificiale; (4) utilizziamo tecniche di Intelligenza Artificiale per analizzare lo stato di diversi segnali climatici per migliorare le previsioni stagionali, secondo una procedura originale chiamata CSI; (5) proponiamo un nuovo algoritmo neuro-evolutivo a molti obiettivi, NEMODPS, che evolve un' architettura della politica su misura per gli obiettivi e relativi compromessi; (6) combiniamo NEMODPS con una strategia di selezione di variabili che apprende una rappresentazione della politica online, e dinamica rispetto agli obiettivi.
Un filo conduttore dei risultati generati in questa raccolta di lavori è che l'apprendimento di un adeguato set di informazioni per informare la politica si configura come una valida risorsa per migliorare le prestazioni del sistema idrico, in particolare rispetto alle sue vulnerabilità più critiche. Nello specifico, mitigando i danni associati agli estremi idrologici (ad esempio le siccità), fasi critiche di sviluppo del serbatoio (costruzione e riempimento), e tensioni sociali derivanti da conflitti tra i diversi utenti idrici.
Parte della ricerca presentata in questa tesi è apparsa, o apparirà, nelle seguenti pubblicazioni: (1) Zaniolo, M., Giuliani, M., Castelletti, A.F., Pulido-Velazquez, M., 2018b. Automatic
design of basin- specific drought indexes for highly regulated water
systems. Hydrology and Earth System Sciences 22, 2409-2424. (Capitolo 2);
(2) Zaniolo, M., Giuliani, M., Castelletti, A., 2019. Data-driven modeling and
control of droughts. IFAC- Papers On Line 52, 54-60. (Chapter 3);.
(3) Zaniolo, M., Giuliani, M., Burlando, P., Castelletti, A., 2020a When timing
matters - misdesigned dam filling impacts hydropower sustainability. Nature
Sustainability (under review). (Capitolo 4);
(4) Giuliani, M., Zaniolo, M., Castelletti, A., Davoli, G., Block, P., 2019. Detecting
the state of the climate system via artificial intelligence to improve seasonal
forecasts and inform reservoir operations. Water Resources Research
55, 9133-9147. (Capitolo 5);
(5) Zaniolo, M., Giuliani, M., Castelletti, A., 2020b. Neuro-evolutionary direct
policy search for multi-objective optimal control. IEEE transactions
on neural networks and learning systems (under review). (Capitolo 6);
(6) Zaniolo, M., Giuliani, M., Castelletti, A., 2020c. Dynamic retrieval of informative
inputs for multi-sector reservoir policy design with diverse spatiotemporal
objective scales. Environmental Modeling and Software (in preparation).
(Capitolo 7).The success of a control policy highly relies by its feature representation, i.e., the information set it is conditioned upon.
In real world control problems, defining an appropriate feature representation is a complex task, given the coexistence of multiple interacting processes whose relevance for the control task is often unclear.
In this thesis, we address the control problem of water resources systems, where a dam release policy is designed accounting for multiple water demands. This decisional problem is challenged by the presence of non-linearities, strong disturbances, possible alternative problem framings, and multiple conflicting objectives.
Currently, the control rules of most water reservoirs are conditioned upon basic information systems comprising reservoir storage and time index, however, the value of a more informative feature representation is generally undisputed.
We capitalize on recent advances in monitoring and forecasting water availability to develop novel feature representation learning strategies to enhance water systems resilience towards their crucial vulnerabilities, including droughts, critical phases in reservoir development (i.e., construction and filling), and multisectoral conflicts.
Additionally, in multi-purpose systems, different control targets might be heterogeneous in their dynamics and vulnerabilities, and likely benefit from a tailored feature representation that varies across different objectives tradeoffs.
We revise current literature on feature representation learning, and propose a taxonomy comprising a priori, a posteriori, and online approaches. For each approach, we propose novel contributions targeting the control problem of multipurpose water systems.
Among the methodological contributions included in this thesis, (1) we propose FRIDA, a feature extraction-based framework to design basin-tailored drought indexes, and (2) we employ FRIDA index to inform water reservoir operations; (3) we extend the concepts of feature representation learning beyond pure control applications to a problem of dam planning and filling; (4) we use Artificial Intelligence to capture the state of multiple climate signals to improve seasonal forecast in a framework named CSI; (5) we propose an original multi-objective neuro-evolutionary algorithm, NEMODPS, that evolves tradeoff-tailored policy architectures, and (6) we combine it with a feature selection routine to learn a policy representation online and tradeoff-dynamically.
A common thread of the outcomes generated in this collection of works is that learning an appropriate policy information set is an asset to improve water system performance, especially by targeting its most critical failures. Specifically, by mitigating the damages associated with hydrological extremes (e.g., drought emergencies), critical stages reservoir development (i.e., construction and filling), and social tensions deriving from conflicts between different users and their demands.
Part of this research has appeared (or will appear) in the following journal publications:(1) Zaniolo, M., Giuliani, M., Castelletti, A.F., Pulido-Velazquez, M., 2018b. Automatic
design of basin- specific drought indexes for highly regulated water
systems. Hydrology and Earth System Sciences 22, 2409-2424. (Capitolo 2);
(2) Zaniolo, M., Giuliani, M., Castelletti, A., 2019. Data-driven modeling and
control of droughts. IFAC- Papers On Line 52, 54-60. (Chapter 3);.
(3) Zaniolo, M., Giuliani, M., Burlando, P., Castelletti, A., 2020a When timing
matters - misdesigned dam filling impacts hydropower sustainability. Nature
Sustainability (under review). (Capitolo 4);
(4) Giuliani, M., Zaniolo, M., Castelletti, A., Davoli, G., Block, P., 2019. Detecting
the state of the climate system via artificial intelligence to improve seasonal
forecasts and inform reservoir operations. Water Resources Research
55, 9133-9147. (Capitolo 5);
(5) Zaniolo, M., Giuliani, M., Castelletti, A., 2020b. Neuro-evolutionary direct
policy search for multi-objective optimal control. IEEE transactions
on neural networks and learning systems (under review). (Capitolo 6);
(6) Zaniolo, M., Giuliani, M., Castelletti, A., 2020c. Dynamic retrieval of informative
inputs for multi-sector reservoir policy design with diverse spatiotemporal
objective scales. Environmental Modeling and Software (in preparation).
(Capitolo 7).DIPARTIMENTO DI ELETTRONICA, INFORMAZIONE E BIOINGEGNERIASystems and Control32DERCOLE, FABIOPERNICI, BARBAR
- …
