Add BNS initial data executable and SpEC initial Data #6422

isaaclegred · 2024-12-18T19:09:18Z

Proposed changes

Add an elliptic executable to solve for the velocity potential of a BNS system for use in initial data to evolutions. Additionally add SpEC analytic data for initializing a spacetime/matter distribution for the velocity potential solve.

Upgrade instructions

Code review checklist

The code is documented and the documentation renders correctly. Run
make doc to generate the documentation locally into BUILD_DIR/docs/html.
Then open index.html.
The code follows the stylistic and code quality guidelines listed in the
code review guide.
The PR lists upgrade instructions and is labeled bugfix or
new feature if appropriate.

Further comments

nilsvu

I found a few potential bugs in the fixed sources that might explain the deviations from SpEC that you see.

Please add a test input file that has the Check: parse line so the executable gets compiled on CI.

Also, it would be nice to add a test for the SpecData that's similar to tests/Unit/PointwiseFunctions/AnalyticData/GrMhd/Test_FukaInitialData.cpp in that you can supply SpEC sample data as an environment variable, and then it checks that the data fulfills the equations (see verify_solution from tests/Unit/Helpers/PointwiseFunctions/AnalyticSolutions/FirstOrderEllipticSolutionsTestHelpers.hpp for an example).

nilsvu · 2024-12-18T21:53:17Z

src/Elliptic/Executables/BnsInitialData/SolveBnsInitialData.hpp

+  static constexpr Options::String help{
+      "Find the solution for the velocity potential of"
+      "an irrotational BNS system using fixed"
+      "problem on fixed space-time and with a fixed"


Missing space at the beginning or end of each line. Also, this sentence seems a little off, maybe delete "using fixed problem".

nilsvu · 2024-12-18T22:06:58Z

src/Elliptic/Systems/BnsInitialData/FirstOrderSystem.hpp

@@ -76,6 +77,7 @@ struct FirstOrderSystem
      ::Tags::Flux<velocity_potential, tmpl::size_t<3>, Frame::Inertial>>;

  using background_fields = tmpl::list<
+      hydro::Tags::RestMassDensity<DataVector>,


where is this used?

It's been used diagnostically, since otherwise we can't write out the density along with the rest of the data.

nilsvu · 2024-12-18T22:20:57Z

src/PointwiseFunctions/AnalyticData/BnsInitialData/Factory.hpp

+
+#include "PointwiseFunctions/AnalyticData/BnsInitialData/SpecInitialData.hpp"
+namespace BnsInitialData::InitialData {
+using all_initial_data = tmpl::list<TovStar, SpecInitialData<1>>;


Where do you get the TovStar from? The TOV solution is just the trivial $\Phi=0$, isn't it?

I'm just going to delete the TOV star, it's not useful anymore now that we have SpEC data

Ope sorry forgot I already deleted it, needed to do some cleanup apparently

nilsvu · 2024-12-18T22:25:16Z

src/PointwiseFunctions/AnalyticData/BnsInitialData/Factory.hpp

+#pragma once
+
+#include "PointwiseFunctions/AnalyticData/BnsInitialData/SpecInitialData.hpp"
+namespace BnsInitialData::InitialData {


name this namespace AnalyticData please for consistency with the other systems

nilsvu · 2024-12-18T22:25:31Z

src/PointwiseFunctions/AnalyticData/BnsInitialData/Factory.hpp

+
+#include "PointwiseFunctions/AnalyticData/BnsInitialData/SpecInitialData.hpp"
+namespace BnsInitialData::InitialData {
+using all_initial_data = tmpl::list<TovStar, SpecInitialData<1>>;


and name this all_analytic_data

nilsvu · 2024-12-19T00:12:32Z

src/PointwiseFunctions/AnalyticData/BnsInitialData/SpecInitialData.cpp

+  data_directory_ = rhs.data_directory_;
+  equation_of_state_ = rhs.equation_of_state_->get_clone();
+  density_cutoff_ = rhs.density_cutoff_;
+  orbital_angular_velocity_ = rhs.orbital_angular_velocity_;
+  euler_enthalpy_constant_ = rhs.euler_enthalpy_constant_;
+  spec_exporter_ =
+      std::make_unique<spec::Exporter>(sys::procs_on_node(sys::my_node()),
+                                       data_directory_, vars_to_interpolate_);


Replace with:

*this = rhs;

nilsvu · 2024-12-19T00:14:10Z

src/PointwiseFunctions/AnalyticData/BnsInitialData/SpecInitialData.hpp

+ * constant read from the input file.
+ */
+template <size_t ThermodynamicDim>
+class SpecInitialData : public elliptic::analytic_data::Background,


Name this maybe just SpecData

nilsvu · 2024-12-19T00:15:20Z

src/PointwiseFunctions/AnalyticData/BnsInitialData/SpecInitialData.hpp

+ *
+ * This class loads numerical data written out by the SpEC initial data solver.
+ * It uses the `spec::Exporter` linked in from SpEC to interpolate to arbitrary
+ * grid points. The coordinates are assumed to be in SpEC's "grid" frame.


We should check if SpEC's grid frame is the one that is deformed to the surface of the star

nilsvu · 2024-12-19T00:16:16Z

src/PointwiseFunctions/AnalyticData/BnsInitialData/SpecInitialData.hpp

+  }
+  // The velocity potential is used in the initial guess
+  template <typename DataType>
+  tuples::TaggedTuple<Tags::VelocityPotential<DataType>> variables(


Move the definition of these variables functions to the cpp file. You can also remove the templating on DataType and just use DataVector (if you want).

nilsvu · 2024-12-19T00:17:34Z

src/PointwiseFunctions/Hydro/InitialData/IrrotationalBns.cpp

@@ -116,10 +116,9 @@ void spatial_rotational_killing_vector(
    const Scalar<DataType>& sqrt_det_spatial_metric) {


remove unused argument

nilsvu · 2025-02-19T23:06:56Z

src/PointwiseFunctions/AnalyticData/BnsInitialData/SpecData.hpp

+        step_function(get(rest_mass_density) - 0.0),
+        get(equation_of_state_->pressure_from_density(rest_mass_density)) /
+                get(rest_mass_density) +
+            *(1.0 +


This * seems off

nilsvu · 2025-02-19T23:08:08Z

tests/Unit/Elliptic/Systems/BnsInitialData/Test_Equations.cpp

+          background_values),
+      get<gr::Tags::InverseSpatialMetric<DataVector, 3>>(background_values),
+      velocity_profile);
+  const auto flux_gradient = partial_derivative(flux, mesh, inv_jacobian);


Use divergence()

nilsvu · 2025-02-19T23:23:37Z

src/PointwiseFunctions/AnalyticData/BnsInitialData/SpecData.hpp

+        make_not_null(
+            &get<::Tags::FixedSource<Tags::VelocityPotential<DataType>>>(
+                result)),
+        euler_enthalpy_constant_ *


Overall minus sign is missing

nilsvu · 2025-02-19T23:25:14Z

src/PointwiseFunctions/AnalyticData/BnsInitialData/SpecData.hpp

+             1.0 / square(lapse()) * deriv_of_shift(ti::i, ti::I) +
+             // Christoffel terms
+             1.0 / square(lapse()) *
+                 (rotational_shift(ti::I) *


this should only be the shift

nilsvu · 2025-02-19T23:38:53Z

src/PointwiseFunctions/AnalyticData/BnsInitialData/SpecData.hpp

+    const auto deriv_log_lapse_times_density_over_specific_enthalpy =
+        partial_derivative(
+            Scalar<DataType>{
+                get(get<gr::Tags::Lapse<DataType>>(interpolated_vars)) *


nilsvu · 2025-02-19T23:52:21Z

src/PointwiseFunctions/AnalyticData/BnsInitialData/SpecData.hpp

+             1.0 / square(lapse()) *
+                 (rotational_shift(ti::I) *
+                      spatial_christoffel_second_kind_contracted(ti::i) -
+                  divergence_spatial_rotational_killing_vector())));


delete this divergence because it should be zero

isaaclegred · 2025-02-21T04:21:44Z

src/PointwiseFunctions/AnalyticData/BnsInitialData/SpecData.cpp

+      density_cutoff_(density_cutoff),
+      orbital_angular_velocity_(orbital_angular_velocity),
+      euler_enthalpy_constant_(euler_enthalpy_constant),
+      spec_exporter_(std::make_unique<spec::Exporter>(1, data_directory_,


Hmm @nilsvu, it seems that if this is set to 1 then my test runs, but the executable gives some weird runtime error and if it's set to sys::props_on_node(sys::my_node()) then the test segfaults but the executable runs no problem

isaaclegred added 2 commits December 18, 2024 11:02

Add BNSInitialData Executable

29320b3

fix breaking change to elliptic analytic data

c4c7e0e

isaaclegred requested a review from nilsvu December 18, 2024 19:09

isaaclegred added the in progress Don't review, used for sharing code and getting feedback label Dec 18, 2024

Fixup remove TOV BNS_ID analytic data

b285226

nilsvu requested changes Dec 19, 2024

View reviewed changes

isaaclegred added 3 commits January 27, 2025 09:49

fixup comments and add broken test of system with spec initial data

54361ec

fixup make SpEC data test work

b567a5b

fixup update test

71efb53

nilsvu reviewed Feb 19, 2025

View reviewed changes

isaaclegred commented Feb 21, 2025

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add BNS initial data executable and SpEC initial Data #6422

Add BNS initial data executable and SpEC initial Data #6422

isaaclegred commented Dec 18, 2024

nilsvu left a comment

nilsvu Dec 18, 2024

nilsvu Dec 18, 2024

isaaclegred Jan 18, 2025

nilsvu Dec 18, 2024

isaaclegred Jan 18, 2025

isaaclegred Jan 18, 2025

nilsvu Dec 18, 2024

nilsvu Dec 18, 2024

nilsvu Dec 19, 2024

nilsvu Dec 19, 2024

nilsvu Dec 19, 2024

nilsvu Dec 19, 2024

nilsvu Dec 19, 2024

nilsvu Feb 19, 2025

nilsvu Feb 19, 2025

nilsvu Feb 19, 2025

nilsvu Feb 19, 2025

nilsvu Feb 19, 2025

nilsvu Feb 19, 2025

isaaclegred Feb 21, 2025

		@@ -116,10 +116,9 @@ void spatial_rotational_killing_vector(
		const Scalar<DataType>& sqrt_det_spatial_metric) {

Add BNS initial data executable and SpEC initial Data #6422

Are you sure you want to change the base?

Add BNS initial data executable and SpEC initial Data #6422

Conversation

isaaclegred commented Dec 18, 2024

Proposed changes

Upgrade instructions

Code review checklist

Further comments

nilsvu left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment