The use of computational modeling has become very popular and important in many engineering and physical fields, as it is considered a fast and inexpensive technique to support and often substitute experimental analysis. In fact system design and analysis can be carried out through computational studies instead of experiments, that are typically demanding in terms of cost and technical resources; sometimes the systems characteristics and the technical problems make the experiments impossible to perform and the use of computational tools is the only feasible option. Demand of resources for realistic simulation is increasing due to the interest in studying complex and large systems. In these framework smart modeling approaches and model reduction techniques play a crucial role for making complex and large system suitable for simulations. Moreover, it should be considered that often more than one simulation is requested in order to perform an analysis. For instance, if a heuristic method is applied to the optimization of a component, the model has to be run a certain number of times. The same problem arises when a certain level of uncertainty affect the system parameters; in this case also many simulation are required for obtaining the desired information. This is the reason why the use of technique that allows to obtain compact model is an interesting topic nowadays. In this PhD thesis different reduction approaches and strategies have been used in order to analyze three energetic systems involving large domain and long time, one for each reduction approach categories. In all the topic considered, a smart model has been adopted and, when data were available, tested using experimental data. All the model are characterized by large domain and the time involved in the analysis are high in all the cases, therefore a method for compact model achievement is used in all the cases. The considered topics are: • Groundwater temperature perturbations due to geothermal heat pump installations, analyzed trough a multi-level model. • District heating networks (DHN), studied from both the fluid-dynamic and thermal point of view and applied to one of the larger network in Europe, the Turin district heating system (DHS), trough a Proper Orthogonal Decomposition - Radial Basis Function model. • Forest fire propagation simulation carried out using a Proper Orthogonal Decomposition projection model.

Modeling strategies for multiple scenarios and fast simulations in large systems: applications to fire safety and energy engineering / Guelpa, Elisa. - (2016). [10.6092/polito/porto/2643992]

Modeling strategies for multiple scenarios and fast simulations in large systems: applications to fire safety and energy engineering

GUELPA, ELISA
2016

Abstract

The use of computational modeling has become very popular and important in many engineering and physical fields, as it is considered a fast and inexpensive technique to support and often substitute experimental analysis. In fact system design and analysis can be carried out through computational studies instead of experiments, that are typically demanding in terms of cost and technical resources; sometimes the systems characteristics and the technical problems make the experiments impossible to perform and the use of computational tools is the only feasible option. Demand of resources for realistic simulation is increasing due to the interest in studying complex and large systems. In these framework smart modeling approaches and model reduction techniques play a crucial role for making complex and large system suitable for simulations. Moreover, it should be considered that often more than one simulation is requested in order to perform an analysis. For instance, if a heuristic method is applied to the optimization of a component, the model has to be run a certain number of times. The same problem arises when a certain level of uncertainty affect the system parameters; in this case also many simulation are required for obtaining the desired information. This is the reason why the use of technique that allows to obtain compact model is an interesting topic nowadays. In this PhD thesis different reduction approaches and strategies have been used in order to analyze three energetic systems involving large domain and long time, one for each reduction approach categories. In all the topic considered, a smart model has been adopted and, when data were available, tested using experimental data. All the model are characterized by large domain and the time involved in the analysis are high in all the cases, therefore a method for compact model achievement is used in all the cases. The considered topics are: • Groundwater temperature perturbations due to geothermal heat pump installations, analyzed trough a multi-level model. • District heating networks (DHN), studied from both the fluid-dynamic and thermal point of view and applied to one of the larger network in Europe, the Turin district heating system (DHS), trough a Proper Orthogonal Decomposition - Radial Basis Function model. • Forest fire propagation simulation carried out using a Proper Orthogonal Decomposition projection model.
2016
File in questo prodotto:
File Dimensione Formato  
TESI DOTTORATO ELISA GUELPA.pdf

accesso aperto

Tipologia: Tesi di dottorato
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 38.76 MB
Formato Adobe PDF
38.76 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2643992
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo