OOF: Finite Element Analysis of Microstructures
The process of using ppm2oof can be broken up conceptually into two separate phases, although the actual pieces can be sometimes intertwined. The first phase consists of identifying features in the image and the assigning material properties to them. The second phase consists of creating and refining the finite element mesh. This section summarizes the relevant sets of commands for each phase, and directs you to more detailed descriptions elsewhere.
The object of phase one is to assign a material type to every pixel in the image, and, optionally, to assign pixels to pixelgroups. Pixels may belong to more than one group, but may not have more than one material.
Most of the operations used in this phase work on the set of currently selected pixels. Although ppm2oof can work on many images at once, the set of selected pixels is shared among all the images. The Select dashboard contains tools for selecting pixels. The /pixelgroups menu holds commands for assigning the selected pixels to groups. These groups can later be re-selected with the /select menu, which also contains commands for modifying the currently selected set. Material properties are assigned to the selected pixels with the /materials menu. Information about the pixels in an image, including their groups and materials, may be obtained from the Pixel dashboard.
The Display and Coordinates dashboards control what is drawn in a graphics window and how much of the image is visible.
Some selection operations may work better on modified images. The Modify_Image dashboard contains simple tools for image manipulation. The modified images can be saved and restored from the Image_Gallery dashboard. If it is necessary to modify an image in an external program, the /ppmfile command can be used to load a new ppm file, as long as the modified image is the same size as the original.
When working on an image, it is often desirable to restrict all operations to a specified region of the image. The active_area menu allows you to define the active area, and store and retrieve it for later use.
In phase two of using ppm2oof you create a finite element mesh from the material image. There are two menus for creating meshes: /simple_mesh and /adaptive_mesh. The simple mesh divides each pixel into two triangular elements. This is quick and easy, but it has two flaws. It creates too many elements in regions where a few large elements would suffice, and it creates meshes that resolve all of the details of an image, including pixelization artifacts. Meshes created by the adaptive mesh menu, on the other hand, have large triangles in large homogeneous regions. The edges of the elements follow material and group boundaries, and by choosing the element size you can control which features of the image are resolved by the mesh. The Mesh dashboard provides information about the mesh and lets you select triangles and move nodes manually.
All of the features that were mentioned above, and all of those that weren't, are described in more detail below. In particular, the introduction to each submenu gives an overview of the uses of that menu. So, for example, the details of the adaptive meshing procedure are described in the section on the /adaptive_mesh menu.