2D Meshing in the presence of curved edges

Flexible meshes such as those used within Infoworks ICM/CS and RS allow geometric flexibility to represent complex shapes in more details.  However, when using GIS layers as breaklines, porous walls,mesh zones, porous polygons, roughness zones etc… this can lead to the creation of extremely small mesh triangles which can have a computational overhead with both meshing and simulation times.  One particular problem which can lead to many small triangles is the presence of round and curved edges within specified GIS layers.
Round and curved edges within GIS packages are specified with a number of closely spaced vertices which define a series of minute edges to define the curved edge.  Where these exist within the 2D zone, the meshing algorithm will create a triangle at every vertex slowing the meshing process down.  As a first step we would recommend that you clean your data in GIS using the available tools before importing into ICM/CS/RS to thin out some of these vertices.  There are also tools within Infoworks ICM/CS/RS under Model->Geometry which can help clean up features imported from GIS.
However, it is also possible to reduce the Minimum Angle (degree) parameter in the 2D Zones property sheet to improve meshing times especially around these curved shapes.

Figure 1: Minimum Angle Parameter in the 2D Zone Property Sheet.

At present the default of 25˚ tries to keep uniform and equilateral triangles, where round edges occur this can create a large number of very small triangles shown in the image below by the dense collection of triangles.

Figure 2: Example of round or curved shapes within the 2D mesh with a minimum angle of 25˚.

Reducing the minimum angle to say 5-10˚ allows the triangles to be less equilateral which will results in fewer triangles around curved edges and quicker mesh times.

Figure 3: Example of round or curved shapes within the mesh with a minimum angle of 10˚.

In some tests, meshing times have been reduced by up to 75%.  The skew of the triangles will not lead to any numerical instability due to the aggregation of these triangles into elements which is based on the Minimum Element Area.

Share this post!

    About Duncan Kitts

    Duncan Kitts is a Senior Support Engineer with Innovyze in the United Kingdom, specializing in River modelling and 2D modelling. He has over 9 years experience of modeling the key hydraulic processes involved in both fluvial systems and urban drainage environments. Duncan is responsible for providing support of both infoworks ICM and Infoworks RS.

    This entry was posted in InfoWorks CS, InfoWorks ICM, InfoWorks RS and tagged . Bookmark the permalink.