Análisis visual de la evolución de zonas de comportamiento similar en fluidos dependientes del tiempo
Archivos
Fecha
2014-07-04
Autores
Oves García, Reinier
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Central “Marta Abreu” de Las Villas. Facultad de Matemática, Física y Computación. Departamento Ciencias de la Computación
Resumen
La visualización de fluidos se basa en el estudio de los métodos de visualización que permiten
representar las propiedades de los líquidos y los gases sobre una región determinada. Extraer zonas
de comportamiento similar en campos vectoriales dependientes del tiempo es una tarea no trivial
que requiere de herramientas sofisticadas y grandes requerimientos de hardware.
En la presente investigación se desarrolla un mecanismo que permite identificar de manera
efectiva las zonas de comportamiento similar en campos vectoriales estacionarios. Este mecanismo
está implementado sobre la base del algoritmo de agrupamiento basado en particiones kmenas y
permite determinar y explorar la evolución de una zona de comportamiento similar en el tiempo. Se
definen dos formas de representar la evolución de las zonas de comportamiento similar. La primera
se realiza a través de la visualización del elemento más representativo de la zona y la segunda y
a través de un grafo que describe las bifurcaciones de las regiones en el tiempo. Los resultados
prácticos y teóricos de dicha investigación se encuentran implementados en lenguaje C++ en el
paquete de herramientas FlowVisTools, el cual ofrece la posibilidad de extensión e integración en
otros sistemas de visualización.
Fluid visualization is based on the study of visualization methods that represent the properties of liquids and gases on a given region. The extraction of areas of similar flow behavior in unsteady vector fields is a non-trivial task that requires of sophisticated tools and big hardware requirements. In this work is developed a mechanism to identify areas of similar behavior in steady vector fields in an effective way. The implementation of this mechanism depends on both, the kmeans clustering algorithm and the visual exploration of areas with similar flow behavior in each time step. We define two forms to represent the evolution of similar flow behavior in unsteady vector fields. The first way shows the evolution of the most representative element in each region and the second way represents the entire region with a graph visualization technique where nodes represent each zone in time and edges represents the evolution process. The practical and theoretical results of this research are implemented in C++ language as a library and each independent program is integrated in the toolkit FlowVisTools which offers the possibility of extension and integration in other display systems.
Fluid visualization is based on the study of visualization methods that represent the properties of liquids and gases on a given region. The extraction of areas of similar flow behavior in unsteady vector fields is a non-trivial task that requires of sophisticated tools and big hardware requirements. In this work is developed a mechanism to identify areas of similar behavior in steady vector fields in an effective way. The implementation of this mechanism depends on both, the kmeans clustering algorithm and the visual exploration of areas with similar flow behavior in each time step. We define two forms to represent the evolution of similar flow behavior in unsteady vector fields. The first way shows the evolution of the most representative element in each region and the second way represents the entire region with a graph visualization technique where nodes represent each zone in time and edges represents the evolution process. The practical and theoretical results of this research are implemented in C++ language as a library and each independent program is integrated in the toolkit FlowVisTools which offers the possibility of extension and integration in other display systems.
Descripción
Palabras clave
Sistemas, Métodos de visualización, Visualización, Campos vectoriales, Visualización de fluidos