One of the ways you can simplify and reduce the size of a finite element model is by using any symmetries present in your model. Whether you are building the geometry natively in COMSOL Multiphysics® or importing a design from an external file, there are modeling strategies as well as features in the software that you can use to take advantage of such symmetries. In this article, we will explain how you can take advantage of symmetry in your simulation to reduce its size and discuss the functionality in the software that enables you to easily do so.
Reducing the size of your model is advantageous because of the amount of computational time and memory that can be saved when solving the equations for your model. These savings can be significant when working with large models and multiphysics models. It is also beneficial regardless of the phase of model development that you are currently working in. If you are starting a model from scratch, by using a reduced, simplified version of your design, you can take less time to set up and build your model and reach results. From there, you can continue to expand the scope of your simulation and increase complexity.
If you are working with a geometry that has already been fully built, reducing its size can help you reach solutions for the model equations more quickly. You can then perform postprocessing in order to visualize the solution for the entire model geometry. Additionally, reducing your model size enables you to catch and resolve any errors in your model setup more quickly.
By using the symmetries in a model, you can reduce its size by half or more, making this an efficient method for solving large models. This applies to cases where both the geometry and modeling assumptions include symmetries. It can also be the case that neglecting, or omitting, some geometric or other modeling features will allow for additional symmetry to be used. Some common types of symmetries are axial symmetry as well as symmetric and antisymmetric planes and lines. Modeling in axial symmetry in particular is very efficient, but often involves assuming that some modeling features can be neglected.
The geometry for a flange, as shown in different sizes: half, a quarter, an eighth, and a cross section.
The geometry for a flange. In lieu of modeling the entire flange, the axisymmetric nature of the design can be used to reduce its model size by half, a quarter, an eighth, and down to a cross section. Reducing to axial symmetry is valid if it can be assumed that the bolt holes do not significantly affect the solution.
A 3D model of a purple gear with two gray work planes.
An example geometry that exhibits cyclic symmetry. The symmetry planes are indicated by two gray work planes.
Axial symmetry is common for cylindrical and similar 3D geometries. If the geometry is axisymmetric, there are variations in the radial (r) and vertical (z) direction only, i.e., not in the angular () direction. You can then solve a 2D problem in the rz-plane instead of the full 3D model, which can save significant computational memory and time. Most physics interfaces in COMSOL Multiphysics® are available in axisymmetric versions and take the axial symmetry into account. In addition, several interfaces allow for an assumed form of azimuthal variation, including:
Symmetry planes and lines are common in both 2D and 3D models. Symmetry means that a model is identical on either side of a dividing line or plane. For a scalar field, the normal flux is zero across the symmetry line. In structural mechanics, the symmetry conditions are different. Many physics interfaces have symmetry conditions directly available as features.
Antisymmetry planes and lines means that the loading of a model is oppositely balanced on either side of a dividing line or plane. For a scalar field, the dependent variable is 0 along the antisymmetry plane or line. Structural mechanics applications have other antisymmetry conditions. Many physics interfaces have symmetry conditions directly available as features.
Partition Operations enable you to split up any geometric entity, whether it be objects, domains, boundaries, or edges. You can divide up your geometry using another geometric object or a work plane, and then use the Delete operation to remove the unnecessary entities.
The Cross Section operation enables you to go from modeling a solid to a plane. A work plane is used in combination with this operation to extract a cross section from a 3D geometry, which can be used within a 2D or 2D axisymmetric model component.
The Projection operation enables you to project 3D objects in order to create 2D curves. The geometry can be projected onto a work plane in a 3D model component or in a 2D model component. This is useful for complex geometries, as you can specify what 3D entities (objects, domains, boundaries, edges, points) you want to project into 2D. This is an alternative to having a work plane cut through the geometry and the entire geometric object, such as what is done for the Cross Section operation.
As noted earlier, there are a plethora of physics interfaces that are available in 2D and 2D axisymmetric versions. Additionally, there are many physics interfaces that have different types of symmetry conditions available as physics feature nodes. These include Symmetry, Symmetry Plane, and Antisymmetry boundary conditions, among other types of symmetry boundary conditions (e.g., Sector Symmetry), which enable you to specify the symmetry planes or lines of symmetry in your model.
A screenshot of the Model Builder showing half of the Shell-and-Tube Heat Exchanger model in the Graphics window and the Symmetry settings. The Symmetry boundary condition that is used in the Shell-and-Tube Heat Exchanger tutorial model.
A screenshot of the Model Builder showing the Permanent Magnet tutorial model in the Graphics window and the Symmetry Plane settings. The Symmetry Plane boundary condition that is used in the Permanent Magnet tutorial model, where we specify the antisymmetry of the magnetic field. There are also several other ways you can use symmetry to simplify magnetic field modeling.
There are some cases where the results of structural mechanics models are not purely symmetric, even though the problem may appear so at first. You should note such cases, which include the following:
There are several blog posts that discuss taking advantage of the symmetries in a model for different application areas. There are also several tutorial models available in the Application Libraries that show different implementations of taking advantage of the symmetries in a geometry. From these you can explore, learn from, and understand the logic that was used for reducing the computational domain as a result of any symmetries. A small sample of the available examples are listed below.
このページに関するフィードバックを送信, または サポートに連絡 してください.