International Business Machines Corporation

Siavash N Meshkat

James M Ruppert

Vijay Srinivasan

Lee Richard Nackman

Susan A Meshkat

A method is provided for producing a mesh representation of an arbitrarily shaped object, for applications such as finite element analysis. The method includes the identifying elements of a mesh, such as a tetrahedral mesh, which are suitable, based on predetermined criteria, for merging into one of a predetermined set of target elements. For instance, if the predetermined set of target elements includes pentahedra and hexahedra, and the method operates on a tetrahedral mesh, then tetrahedra of the mesh are identified based on known ways of decomposing the target elements into tetrahedra. The groups of tetrahedra are identified based on whether they share faces in common, and whether they have faces which share a common edge and which are either coplanar or have an angle between the faces which satisfies a predetermined condition. The latter faces are referred to as quadrilateral pairs. A graph representation of the mesh is used, preferably including nodes representing regions of the mesh (i.e., tetrahedra) and edges between the nodes representing faces shared in common and quadrilateral pairs of faces. Using a search tree representation of a graph of a target element, the graph of the mesh is searched to identify subsets of the graph of the mesh, which represent groups of the tetrahedra of the mesh which are suitable for merging into the target element. By merging the tetrahedra this way, the total number of elements of the mesh is reduced, compared with the original tetrahedral mesh.

Method and system for producing mesh representations of objects

A method and system are provided for subdividing an arbitrarily shaped object into a collection of geometric elements ('cells') having predefined, simple topologies which facilitate further subdivision into tetrahedra, and which are well suited for further applications, such as finite element calculations. Initially, a representation of the object includes one or more regions, at least one of the regions not meeting the definition of 'cell', each region having vertices, edges, and faces. An edge is selected according to a priority scheme, and slices which run through the edge and which also either run through other edges or vertices of the representation, are coplanar with other faces of the representation, or have other predetermined attributes, are considered for use in subdividing the representation. A score, which was initially calculated for the representation, is recalculated for hypothetical subdivisions of the representation incorporating each respective one of such slices. The scores are related to predetermined factors, such as the number of regions which are cells, relative to the total number of regions, the number of faces whose topologies correspond with the topologies of faces of cells, relative to the total number of faces, and factors relating to the shapes of the regions and faces. For one of the slices, the score is maximized. The representation is then subdivided, using that slice. This process is repeated until no regions of the representation remain which do not meet the definition of a cell. The method or the invention is preferably practiced using parallel processing at one or more levels. That is, either (i) different regions of the object are subdivided in parallel, (ii) for a given region, interesting slices are scored for different high-priority edges in parallel, (iii) for a given high-priority edge or a given region, interesting slices through that edge are scored in parallel, or a suitable combination thereof.

Parallel method for subdivision of arbitrary curved solids

A method for producing a set of points for a mesh of finite elements for a body to be analyzed, the body exhibiting boundaries and faces. The method comprises the steps of: producing an initial mesh of elements for the body, each element having a plurality of vertices and edges; for each vertex in a first subset of the vertices, finding points of intersection (POI) between a sphere of radius R, centered at a vertex being considered, and the boundaries and edges; determining if a found POI is either on or within the boundaries of the body and exhibits more than a minimum clearance distance from the boundaries, and vertices, any such POI being classified as a selected POI; inserting a selected POI in the initial mesh; and repeating the aforementioned steps for all POI's in the first subset and then proceeding to succeeding subsets of vertices until all POI's in all subsets have been considered.

Point placement method for use in a three-dimensional automatic mesh

A method is described for producing a mesh of finite elements that are entirely within a body to be analyzed, the body exhibiting edges and surfaces. The method comprises the steps of: producing an initial mesh of elements for the body, each element having a plurality of vertices and edges; finding any body edge segment which is not coincident with a finite element edge and adding a vertex on that segment, the added vertex positioned to minimize the number new vertex points needed on the body edge segment. Additional finite elements are then created by connecting the new vertex to adjacent close vertices. A finite element is now identified which has both an edge coincident with an edge segment of the body and another edge segment that passes through a face of the body. A new vertex is placed on the body face in the vicinity of an intersection of the intersecting element edge so as to minimize the number of added required vertices, and the method then creates additional finite elements by connecting the added vertex to adjacent close vertices.

Face and edge trim method for an automatic mesh generation system

A method for fine decomposition in finite element mesh generation, in which a polygonal boundary of a domain is input into the system by an analyst and the domain is automatically divided into rough elements generally corresponding to Voronoi regions, that is, regions which are closer to respective ones of the polygonal line segments or reflex vertices therebetween. Any arc portion of these regions is converted to a straight line. Additional lines are formed between interior vertices of the rough regions so that all rough regions are either triangles or trapezoids. Adjacent rough regions are then paired across internal boundaries and are classified into four types. The rough regions are then subdivided into fine regions of triangular shape according to rules associated with each of the four types. The degree of fine subdivision can be controlled according to known equations providing the total number of fine elements. Finite difference equations are solved on the so generated fine regions and a physical variable is displayed in pictorial form. The analyst then readjusts the domain boundary and repeats the process.

Growing community of inventors

San Jose, CA, United States of America

Siavash N Meshkat