 |
|||||||
 |
|
||||||
|
||||||||||||||||
 |
|
 |
|
|
Many disciplines and applications need to simulate real-world processes to better understand, analyze, and predict their behaviors. Natural disaster prevention and management (e.g. Flood, Forest fire) and Environmental protection (e.g. natural resources; like ground water, oil) are examples of these applications. When dealing with simulation of a phenomenon, several factors must be taken into account such as its dynamic behavior and its geometrical and topological representation (spatial modeling). This is where a Geographic Information System (GIS) is a valuable tool. Geographic Information System (GIS), through its powerful capabilities to process spatial data, provides modelers with strong computing platforms for data management, integration, visualization, querying, and analysis and thus greatly facilitate the implementation of various models. However, regarding the nature of real-world phenomena which are 3D and dynamic (they change with respect to space and time) from one side and the data structures of GIS that are usually 2D and static on the other side, several complexities and limitations are added to the entire process of simulation using existing GIS. This research is an attempt to resolve the current GIS limitations to simulate a dynamic phenomenon by improving a dynamic 3D data structure. This structure is based on 3D Delaunay tetrahedralization that deals with discrete objects and continuous phenomena (filed) and generates an on-the-fly interactive topological mesh for numerical simulation. In addition, a Free-Lagrange method based on 3D kinetic Voronoi diagram is developed as solution of numerical model in which topological relations between the mesh elements as well as the physical parameters are locally updated. Leila Hashemi Beni, PhD Student
|
