| dc.description.abstract | Solid meshes play a very important role in finite element analysis. For the
time being, the complete automatic construction technology of tetrahedral meshes
and boundary layer meshes is mature enough. However, it is easy to generate
narrow meshes on the corners, resulting in poor quality. It will directly affect the
analysis results and efficiency. In order to improve this situation, our laboratory
is committed to developing the automatic construction of hexahedral meshes and
prism meshes.
Taking the most common thin shell component in injection molding as an
example, there are various protrusion-features on the shell, including tubes,
columns, ribs and other complex structures. In the past, the meshes of those
structures were built manually by engineers. The process was repetitive and
boring, and it wasted a lot of time. Therefore, if such protrusions can be
recognized automatically and the correlation between the protrusions can be
recorded, it will assist in the construction of solid meshes.
For the reason, this research focuses on chamfer recognition, inner and outer
face recognition, and outer auxiliary face recognition. Develop related algorithms,
hoping to record the results of inner and outer face types data during the
recognition process, so as to provide relevant information for subsequent feature
decomposition and meshes construction. | en_US |