Infolytica's simulation software utilizes the Finite Element Method to accurately model and predict the results for 2D and 3D models. In the finite element method of analysis, the model is divided into a mesh of elements. The field inside each element is represented by a polynomial with unknown coefficients. The finite element analysis is the solution of the set of equations for the unknown coefficients.

Automatic mesh generation

Built into all of Infolytica's software products, our 2D and 3D automatic mesh generation is designed for users who don't want to concern themselves with the construction of the mesh. This type of user can rely on Infolytica's adaption process to get the accurate solutions they require. Please see below for more information on adaption.

For users who want more control of the mesh, especially how it is defined in certain areas of interest, tools are provided that allow the refinement of the mesh. More information on these tools can be found in Refined Meshing Tools, below.

Whichever method you choose, automation or control, Infolytica offers great flexibility in determining the configuration of the mesh.

Adaption

Infolytica's adaption process automatically identifies the areas of the mesh most in need of improvement and refines the mesh (h-adaption), or increases the polynomial order of the elements (p-adaption), in those areas.

Our proven algorithms monitor both h- and p-adaption until the desired accuracy is achieved. The p-adaption is based on the hierarchical element methods developed and improved by Infolytica Corporation.

The following adaption types are available to the engineer to select the best adaption for a given design problem:

• h-adaption (2D and 3D solutions) - subdivides elements into smaller elements
• p-adaption (3D solutions): - polynomial order is increased
• h-p adaption (3D solutions): - both methods may be used for optimum performance

Adaption is controlled by options that are set before solving begins. These options determine:

• the type of adaption method (h-adaption or p-adaption)
• the percentage of elements of the mesh that are refined during each adaption step
• the adaption convergence tolerance
• the maximum number of adaption steps

During adaption, the change in the stored energy, between solver runs, is compared to the specified adaption convergence tolerance. If the change in the value of the stored energy is greater than the tolerance, the mesh is refined and another solution is run. The model is re-solved until the solution converges to the requested tolerance, or until the maximum number of adaption steps is reached.

Adaption can be specified on a per component basis or globally (entire mesh).

Mesh Layers (3D only)

The mesh layer feature is used to create highly anisotropic volume elements either adjacent to a selected surface or within the volume of a selected component.

Mesh layering immensely simplifies the setup of 3D solid model problems. It is a necessity for performing skin depth analysis, but more generally, allows users to easily and quickly specify regions where an extremely refined mesh is needed.

This feature also improves the remeshing efficiency in 3D motion problems, particularly when the separation between the moving component and a stationary part of the model is very thin.

The following mesh layer controls are available:

• Height of each layer - uniform, logarithmic or user-defined
• Thickness of the layers
• Total thickness of the mesh layer region
• Number of layers

Refined meshing tools

Since smaller elements allow the true solution field to be better approximated, all of Infolytica's software products provide you with control over the mesh density. You can change the size of the elements for the entire model, or only in the areas of interest, using the edge subdivision methods (uniform, logarithmic, or double logarithmic) and/or the maximum element edge length feature.

Edge subdivisions

The user assigns a numerical value to indicate the number of segments, and then selects one of the following methods:

• Uniform Subdivision tool - subdivides the component's edges into equal segments
• Double logarithmic ends to center - subdivides the component's edges into segments concentrated at both ends
• Double logarithmic center to ends - subdivides the component's edges into segments concentrated at the center
• Logarithmic from the end to the start - subdivides the component's edges into segments concentrated at the end
• Logarithmic from the start to the end - subdivides the component's edges into segments concentrated at the start

Changes you have made to the mesh can be viewed using the Initial Mesh features, which show the mesh at the solving plane (XY plane, Z=0) for 2D solutions, or for the entire 3D model.

Maximum element edge length

Specifying a maximum element edge length limits the size of the mesh elements, thus increasing (or decreasing) the mesh density. The value can be set either for individual components or for the entire mesh.

For an individual component, the maximum element edge length can be applied to the component's volume or on specified surfaces, edges and vertices.