Tesis doctorales: TESEO

Ayuda Salir

Ficha de Tesis

Imprimir

Ministerio de Universidades - Gobierno de España

Título: A COMPREHENSIVE METHODOLOGY TO PREDICT FOREST FIRES BEHAVIOR USING COMPLEMENTARY MODELS
Autor: Brun Sorribas, Carlos
Universidad: Universidad Autónoma de Barcelona
Departamento: ARQUITECTURA DE COMPUTADORES Y SISTEMAS OPERATIVOS
Fecha de Lectura: 18/07/2014
Programa de doctorado: Computación de altas prestaciones
Dirección:
- MARGALEF BURRULL, TOMAS MANUEL (DIRECTOR)
Tribunal:
- San Miguel Ayanz, Jesús (PRESIDENTE)
- FRANCO PUNTES, DANIEL (SECRETARIO)
- Hanzich Estévez, Maurizio (VOCAL)
Descriptores:
- INCENDIOS FORESTALES
- SIMULACION
- PREDICCION NUMERICA METEOROLOGICA
- SISTEMAS EN TIEMPO REAL
Localización: BIBLIOTECA DE COMUNICACIÓN Y HEMEROTECA GENERAL UAB

2014brunsacomp.pdf

Marcador: https://www.educacion.gob.es/teseo/mostrarRef.do?ref=1099377Añadir a favoritos.
Resumen:
Large scale forest fires suppose a great challenge due to their impact on society on many levels. This physical phenomenon is an interdisciplinary problem that requires the efforts of researchers across various fields. Therefore, predicting the behavior of forest fires and minimizing the effects of this hazard is the main goal of this work. One of the major problems with this kind of hazard is the uncertainty and imprecision of the input parameters. We rely on a prediction strategy based on calibration techniques that try to tune these parameters to produce a more accurate prediction and reduce the uncertainty of the parameters. In this work, we propose strategies to solve some of the restrictions associated with this technique. The spatial distribution of the parameters has been considered uniform along the terrain, and we introduce a complementary model to simulate the wind behavior over complex terrains, due to the leading role of wind in fire spreading. We also suggest coupling a meteorological model to enrich the knowledge of the system of the temporal evolution of the weather variables. All these components are integrated in a simulation and prediction framework, and a methodology to simulate real forest fire scenarios is defined in which the steps, processes, models and tools to build an effective real-time forest fire assessment system are specified. This system relies on High Performance Computing resources and paradigms to provide a response in an acceptable time due to the high computational cost of evaluating several scenarios and integrating complementary models.

Imprimir

Modificar Consulta Nueva Consulta

Versión 4.4