SpECTRE
v2025.03.17
|
Control error in the for the domain::CoordinateMaps::TimeDependent::Skew
map.
More...
#include <Skew.hpp>
Public Types | |
using | object_centers = domain::object_list<> |
using | options = tmpl::list<> |
Public Member Functions | |
Skew (Skew &&rhs)=default | |
Skew & | operator= (Skew &&rhs)=default |
Skew (const Skew &) | |
Skew & | operator= (const Skew &) |
std::optional< double > | get_suggested_timescale () const |
Returns the internal suggested timescale. A std::nullopt means that no timescale is suggested. | |
void | reset () |
Resets the internal suggested timescale to nullopt. | |
void | pup (PUP::er &p) |
template<typename Metavariables , typename... TupleTags> | |
DataVector | operator() (const ::TimescaleTuner< true > &, const Parallel::GlobalCache< Metavariables > &cache, const double time, const std::string &function_of_time_name, const tuples::TaggedTuple< TupleTags... > &measurements) |
Static Public Attributes | |
static constexpr Options::String | help |
Control error in the for the domain::CoordinateMaps::TimeDependent::Skew
map.
Computes the error in the map parameters \(F_y(t)\) and \(F_z(t)\) of the Skew using a modified version of Eq. (71) from [94]. This modified control error is
\begin{equation} Q_{F_j} = g\left(\frac{w_A \Theta_A^j + w_B \Theta_B^j}{w_A + w_B}\right) - (1-g)F_j \end{equation}
where the falloff function is assumed to be \(W(\vec{x}) \approx 1\) from the Skew map, \(\Theta_H^j\) are the inclination angles between the \(x\)-axis and the normal to the horizon at the intersection point \(x_H^\textrm{Int}\) between the \(x\)-axis and the horizon, \(w_A\) and \(w_B\) are averaging weights defined by
\begin{equation} w_H = \exp{\left(-\frac{x^0_C - x_H^\textrm{Int}}{x^0_C - C^0_H}\right)} \end{equation}
with \(C^0_H\) being the centers of the excision boundaries, and finally with
\begin{equation} g = \frac{1}{2}\left(1-\tanh\left(10\frac{x_A^\textrm{Int} - x_B^\textrm{Int}}{C^0_A - C^0_B} - 5\right)\right). \end{equation}
This transition function \(g\) is meant to only activate skew control when the black holes are close to merger to avoid adverse effects with junk radiation. If \(g < 0.0025\), then we turn skew control off completely and the control error just becomes
\begin{equation} Q_{F_j} = -F_j. \end{equation}
This threshold value of \(g = 0.0025\) was chosen in SpEC and seems to work well.
Requirements:
|
staticconstexpr |