Electrostatic PDE¶
Governing Equations¶
The governing equation can easily be derived by considering Faraday's law in its stationary form (no time-derivative of the magnetic flux density) and law of Gauss. Introducing the electric potential \mathbf{E} = -\nabla{\phi}, we obatain \begin{equation} -\nabla \cdot {\mathbf{\epsilon}\nabla{\phi}} = 0, \end{equation} where \epsilon is the permittivity. The boundary conditions can be given as \begin{equation} \phi = \phi_{\textrm e} \qquad \textrm{on }\Gamma_{\textrm e}, \end{equation}
\begin{equation} \mathbf{D}\cdot\mathbf{n} = -\mathbf{\epsilon}\nabla{\phi} \cdot\mathbf{n} = q_{\textrm s} \qquad \textrm{on }\Gamma_{\textrm n}, \end{equation} with \phi_{\textrm e} as the Dirichlet (essential) BC-value, \Gamma_{\textrm e} the essential boundary, \mathbf{n} the normal vector on Neumann (natural) boundary \Gamma_{\textrm n} and q_{\textrm s} the surface charge density.
Boundary conditions¶
As described above, there are two types of BC's, Neumann and Dirichlet. However, since openCFS is physics-based, there are several variations and combinations of these. They can be defined in the simulation-xml file in the BC-section
<bcsAndLoads>
<potential name="" value=""/>
<ground name=""/>
<charge name="" value=""/>
<chargeDensity name="" value=""/>
<fluxDensity name=""/>
<fieldParallel name=""/>
<constraint name="Electrodes" quantity="elecPotential"/>
</bcsAndLoads>
- Potential
<potential name="" value=""/>
: Inhomogeneous Dirichlet value \phi_{\textrm{e}}=\textrm{value} - Ground
<ground name=""/>
: Homogeneous Dirichlet value \phi_{\textrm{e}}=0 - Charge
<charge name="" value=""/>
: Surface charge \int_{\textrm{A}}q_s\, ds=\textrm{value} (Neumann-type BC), where ds is the infinitesimal surface element - Charge density
<chargeDensity name="" value=""/>
: Surface charge density q_s=\textrm{value} (Neumann-type BC) - Flux density
<fluxDensity name="" value=""/>
: Defines normal component of dielectric displacement \mathbf{D}\cdot \mathbf{n}=\textrm{value} (similar to a surface charge density) - Field parallel
<fieldParallel name=""/>
: \mathbf{D}\cdot \mathbf{n}=0 on the specified surface - Constraint
<constraint name="Electrodes" quantity="elecPotential"/>
: Puts a constraint on a certain area. This means the specified quantity must be the same for all nodes over the specified area.
Material¶
In general, the permittivity is a rank-2 tensor for anisotropic materials, and the 'total' permittivity can be defined via the relative permittivity \epsilon_{\textrm{r}} as \begin{equation} \epsilon=\epsilon_{\textrm{0}} \epsilon_{\textrm{r}}. \end{equation} It can be specified in the material-xml file e.g. for polyethylene either by defining an insotropic permittivity and/or a tensorial value for the anisotropic version
<material name="polyethylene">
<electric>
<permittivity>
<linear>
<isotropic>
<real> 8.85419E-12 </real>
</isotropic>
<tensor dim1="3" dim2="3">
<real>
1.99200E-11 0.00000E+00 0.00000E+00
0.00000E+00 1.99200E-11 0.00000E+00
0.00000E+00 0.00000E+00 1.99200E-11
</real>
</tensor>
</linear>
</permittivity>
</electric>
</material>
Analysis Types¶
Since in general we are dealing with a time-independent PDE, we just have a static analysis:
- Static:
Postprocessing results (hysteresis excluded)¶
-
Electric Potential (node result, primary solution)
-
Electric Field Intensity (element result) \begin{equation} \mathbf E = -\nabla \phi \end{equation}
-
Electric Energy Density (element result) \begin{equation} e = \epsilon \mathbf{E} \cdot \mathbf{E} \end{equation}
-
** Charge Density** (element result) \begin{equation} q_{{s}} \end{equation}
- Electric Charge (surface region result) \begin{equation} Q = \int_{{A}} q_{{s}} ds \end{equation}
- Electric Energy (region result) \begin{equation} E = \int_{{\Omega}} \mathbf{E} \cdot \mathbf{E} \, dV \end{equation}