... | @@ -41,14 +41,14 @@ A domain is defined as |
... | @@ -41,14 +41,14 @@ A domain is defined as |
|
domain = dG3DDomain(tag,phys,sp,lnum, fdg, dim, nonLocalVar, nonConstitutiveExtraDOFVar, nonCurlData)
|
|
domain = dG3DDomain(tag,phys,sp,lnum, fdg, dim, nonLocalVar, nonConstitutiveExtraDOFVar, nonCurlData)
|
|
```
|
|
```
|
|
where
|
|
where
|
|
- *tag*: a positive integer associated to this domain. If only one domain is available, this number can be choose freely as a positive integer. When multiple domains are presents, this number provides different possibility:
|
|
- **tag**: a positive integer associated to this domain. If only one domain is available, this number can be choose freely as a positive integer. When multiple domains are presents, this number provides different possibilities:
|
|
* If *fdg=False*, and *tag* is the same for all domains, a conventional constitous Galerkin FE (CG) is found.
|
|
* If **fdg=False**, and **tag** is the same for all domains, a conventional constitous Galerkin FE (CG) is found.
|
|
* If *fdg=False*, and *tag* is different between two domains, CG is used for these domains but an DG interface between these two domains must be provided to guarantee the continuity of the fields accross this interface. This option is useful when considering multiple domains when only decohesion between them is interested.
|
|
* If **fdg=False**, and *tag* is different between two domains, CG is used for these domains but an DG interface between these two domains must be provided to guarantee the continuity of the fields accross this interface. This option is useful when considering multiple domains when only decohesion between them is interested.
|
|
* If *fdg=True*, *tag* can be assigned to an arbitrary positive integer as a full discontinuous Galerkin (DG) is considered.
|
|
* If **fdg=True**, **tag** can be assigned to an arbitrary positive integer as a full discontinuous Galerkin (DG) is considered.
|
|
- *phys*: Finite elemet support of this domain. The value of *phys* corresponds to the *Physical* number of the domain defined in the geometry with Gmsh.
|
|
- **phys**: Finite elemet support of this domain. The value of **phys** corresponds to the *Physical* number of the domain defined in the geometry with Gmsh.
|
|
- *sp*: function space type governs the finite element approximation:
|
|
- **sp**: function space type governs the finite element approximation. There are two possibilities:
|
|
* *sp=0*: Lagrange-based shape functions
|
|
* **sp=0**: Lagrange-based shape functions
|
|
* *sp=1*: Hierarchical shape functions
|
|
* **sp=1**: Hierarchical shape functions
|
|
|
|
|
|
|
|
|
|
```python
|
|
```python
|
... | | ... | |