temporal_discretisation.rng

<?xml version="1.0" encoding="UTF-8"?>
<grammar xmlns:a="http://relaxng.org/ns/compatibility/annotations/1.0" xmlns="http://relaxng.org/ns/structure/1.0">
  <define name="temporal_discretisation_options">
    <element name="temporal_discretisation">
      <element name="theta">
        <a:documentation>Implicit/explicit control (TTHETA)
 =0.  -- explicit
 =0.5 -- Crank-Nicolson
 =1.  -- implicit</a:documentation>
        <ref name="real"/>
      </element>
      <optional>
        <ref name="temporal_control_volume_options"/>
      </optional>
      <optional>
        <ref name="temporal_discontinuous_galerkin_options"/>
      </optional>
    </element>
  </define>
  <define name="temporal_control_volume_options">
    <element name="control_volumes">
      <a:documentation>Temporal discretisation options that are only relevant if a control volume or coupled_cv spatial discretisation is selected for this field.</a:documentation>
      <optional>
        <element name="number_advection_iterations">
          <a:documentation>Number of iterations within an advection solve.
This increases the accuracy of the face values and ensures that
the pivoted solution is cancelled out.
Defaults to 1 if unselected.</a:documentation>
          <ref name="integer"/>
          <optional>
            <element name="tolerance">
              <a:documentation>Cut short advection_iterations if the specified tolerance
is reached.
This only works for pure control volume discretisations.</a:documentation>
              <ref name="real"/>
              <choice>
                <element name="infinity_norm">
                  <a:documentation>Select the norm with which you want the tolerance to be tested.

The infinity norm.</a:documentation>
                  <empty/>
                </element>
                <element name="l2_norm">
                  <a:documentation>Select the norm with which you want the tolerance to be tested.

The l2 norm.</a:documentation>
                  <empty/>
                </element>
                <element name="cv_l2_norm">
                  <a:documentation>Select the norm with which you want the tolerance to be tested.

The l2 norm evaluated on a control volume mesh.</a:documentation>
                  <empty/>
                </element>
              </choice>
            </element>
          </optional>
        </element>
      </optional>
      <optional>
        <choice>
          <element name="maximum_courant_number_per_subcycle">
            <a:documentation>Use timestep subcycling to solve this equation.
Specify the maximum courant number per subcycle.
This only works for pure control volume discretisations.</a:documentation>
            <ref name="real"/>
            <ref name="field_based_cfl_number_options"/>
          </element>
          <element name="number_advection_subcycles">
            <a:documentation>Use timestep subcycling to solve this equation.
Specify the number of subcycles.
This only works for pure control volume discretisations.</a:documentation>
            <ref name="integer"/>
          </element>
        </choice>
      </optional>
      <optional>
        <element name="limit_theta">
          <a:documentation>Only works if a control volume or coupled_cv spatial discretisation is selected.
If not active then the theta specified above will be used.
Otherwise use variable limited theta on individual faces.</a:documentation>
          <empty/>
        </element>
      </optional>
      <optional>
        <element name="pivot_theta">
          <a:documentation>Only works if a control volume or coupled_cv spatial discretisation is selected.
Time discretisation of upwind discretisation off which the
higher order solution is pivotted.
 - pivot_theta = 1 - implicit pivot (default if not set and 
                     best choice if not intentionally modifying
                     scheme to be explicit)
 - pivot_theta = 0 - explicit pivot</a:documentation>
          <ref name="real"/>
        </element>
      </optional>
    </element>
  </define>
  <define name="temporal_discontinuous_galerkin_options">
    <element name="discontinuous_galerkin">
      <a:documentation>This enables DG-specific timestepping options, such as
explicit advection subcycling. </a:documentation>
      <optional>
        <choice>
          <element name="maximum_courant_number_per_subcycle">
            <a:documentation>Use timestep subcycling to solve this equation.
Specify the maximum courant number per subcycle.</a:documentation>
            <ref name="real"/>
            <ref name="dg_field_based_cfl_number_options"/>
          </element>
          <element name="number_advection_subcycles">
            <a:documentation>Use timestep subcycling to solve this equation.
Specify the number of subcycles.</a:documentation>
            <ref name="integer"/>
          </element>
        </choice>
      </optional>
    </element>
  </define>
</grammar>