Solver configurations#

This tutorial will focus on configuring the solver with different stages. We will use a simple 3-reaction 3-species mechanism. The setup here is the same in Multiple Grid Cells, except only one grid cell is used.

\[\begin{split}A &\longrightarrow B, &k_{1, \mathrm{user\ defined}} \\ 2B &\longrightarrow B + C, &k_{2, \mathrm{user\ defined}} \\ B + C &\longrightarrow A + C, \qquad &k_{3, \mathrm{user\ defined}} \\\end{split}\]

If you’re looking for a copy and paste, choose the appropriate tab below and be on your way! Otherwise, stick around for a line by line explanation.

#include <micm/process/user_defined_rate_constant.hpp>
#include <micm/solver/rosenbrock.hpp>
#include <micm/solver/solver_builder.hpp>

#include <chrono>
#include <iomanip>
#include <iostream>

// Use our namespace so that this example is easier to read
using namespace micm;

void test_solver_type(auto& solver)
{
  auto state = solver.GetState();

  // mol m-3
  state.variables_[0] = { 1, 0, 0 };

  double k1 = 0.04;
  double k2 = 3e7;
  double k3 = 1e4;
  state.SetCustomRateParameter("r1", k1);
  state.SetCustomRateParameter("r2", k2);
  state.SetCustomRateParameter("r3", k3);

  double temperature = 272.5;  // [K]
  double pressure = 101253.3;  // [Pa]
  double air_density = 1e6;    // [mol m-3]

  state.conditions_[0].temperature_ = temperature;
  state.conditions_[0].pressure_ = pressure;
  state.conditions_[0].air_density_ = air_density;

  // choose a timestep and print the initial state
  double time_step = 200;  // s

  state.PrintHeader();
  state.PrintState(0);

  SolverStats total_stats;
  std::chrono::duration<double, std::nano> total_solve_time = std::chrono::nanoseconds::zero();

  // solve for ten iterations
  for (int i = 0; i < 10; ++i)
  {
    // Depending on how stiff the system is
    // the solver integration step may not be able to solve for the full time step
    // so we need to track how much time the solver was able to integrate for and continue
    // solving until we finish
    double elapsed_solve_time = 0;

    while (elapsed_solve_time < time_step)
    {
      auto start = std::chrono::high_resolution_clock::now();
      solver.CalculateRateConstants(state);
      auto result = solver.Solve(time_step - elapsed_solve_time, state);
      auto end = std::chrono::high_resolution_clock::now();

      total_solve_time += std::chrono::duration_cast<std::chrono::nanoseconds>(end - start);
      total_stats.function_calls_ += result.stats_.function_calls_;
      total_stats.jacobian_updates_ += result.stats_.jacobian_updates_;
      total_stats.number_of_steps_ += result.stats_.number_of_steps_;
      total_stats.accepted_ += result.stats_.accepted_;
      total_stats.rejected_ += result.stats_.rejected_;
      total_stats.decompositions_ += result.stats_.decompositions_;
      total_stats.solves_ += result.stats_.solves_;
      total_stats.singular_ += result.stats_.singular_;

      elapsed_solve_time = result.final_time_;
    }

    state.PrintState(time_step * (i + 1));
  }
  std::cout << "Total solve time: " << total_solve_time.count() << " nanoseconds" << std::endl;
  std::cout << "accepted: " << total_stats.accepted_ << std::endl;
  std::cout << "function_calls: " << total_stats.function_calls_ << std::endl;
  std::cout << "jacobian_updates: " << total_stats.jacobian_updates_ << std::endl;
  std::cout << "number_of_steps: " << total_stats.number_of_steps_ << std::endl;
  std::cout << "accepted: " << total_stats.accepted_ << std::endl;
  std::cout << "rejected: " << total_stats.rejected_ << std::endl;
  std::cout << "decompositions: " << total_stats.decompositions_ << std::endl;
  std::cout << "solves: " << total_stats.solves_ << std::endl;
  std::cout << "singular: " << total_stats.singular_ << std::endl;
}

int main()
{
  auto a = Species("A");
  auto b = Species("B");
  auto c = Species("C");

  Phase gas_phase{ std::vector<Species>{ a, b, c } };

  Process r1 = Process::Create()
                   .SetReactants({ a })
                   .SetProducts({ Yields(b, 1) })
                   .SetRateConstant(UserDefinedRateConstant({ .label_ = "r1" }))
                   .SetPhase(gas_phase);

  Process r2 = Process::Create()
                   .SetReactants({ b, b })
                   .SetProducts({ Yields(b, 1), Yields(c, 1) })
                   .SetRateConstant(UserDefinedRateConstant({ .label_ = "r2" }))
                   .SetPhase(gas_phase);

  Process r3 = Process::Create()
                   .SetReactants({ b, c })
                   .SetProducts({ Yields(a, 1), Yields(c, 1) })
                   .SetRateConstant(UserDefinedRateConstant({ .label_ = "r3" }))
                   .SetPhase(gas_phase);

  auto system = System(SystemParameters{ .gas_phase_ = gas_phase });
  auto reactions = std::vector<Process>{ r1, r2, r3 };

  auto two_stage = micm::CpuSolverBuilder<micm::RosenbrockSolverParameters>(micm::RosenbrockSolverParameters::TwoStageRosenbrockParameters())
                       .SetSystem(system)
                       .SetReactions(reactions)
                       .Build();

  auto three_stage = micm::CpuSolverBuilder<micm::RosenbrockSolverParameters>(micm::RosenbrockSolverParameters::ThreeStageRosenbrockParameters())
                         .SetSystem(system)
                         .SetReactions(reactions)
                         .Build();

  auto four_stage = micm::CpuSolverBuilder<micm::RosenbrockSolverParameters>(micm::RosenbrockSolverParameters::FourStageRosenbrockParameters())
                        .SetSystem(system)
                        .SetReactions(reactions)
                        .Build();

  auto four_stage_da = micm::CpuSolverBuilder<micm::RosenbrockSolverParameters>(micm::RosenbrockSolverParameters::FourStageDifferentialAlgebraicRosenbrockParameters())
                            .SetSystem(system)
                            .SetReactions(reactions)
                            .Build();

  auto six_stage_da = micm::CpuSolverBuilder<micm::RosenbrockSolverParameters>(micm::RosenbrockSolverParameters::SixStageDifferentialAlgebraicRosenbrockParameters())
                            .SetSystem(system)
                            .SetReactions(reactions)
                            .Build();

  std::cout << "Two stages: " << std::endl;
  test_solver_type(two_stage);

  std::cout << std::endl << "Three stages: " << std::endl;
  test_solver_type(three_stage);

  std::cout << std::endl << "Four stages: " << std::endl;
  test_solver_type(four_stage);

  std::cout << std::endl << "Four stages differential algebraic: " << std::endl;
  test_solver_type(four_stage_da);

  std::cout << std::endl << "Six stages differential algebraic: " << std::endl;
  test_solver_type(six_stage_da);
}

Line-by-line explanation#

There are a total of five different configurations for the Rosenbrock solver. Each corresponds to a different number of stages. What each configuration means is beyond the scope of this tutorial. However, references are included for further reading.

  • Two stage
    • An L-stable method, 2 stages, order 2. [HW96]

  • Three stage
    • An L-stable method, 3 stages, order 3. [SVB+97]

  • Four stage
    • An L-stable method, 4 stages, order 4. [HW96]

  • Four stage, differential algebraic
    • A stiffly stable method, 4 stages, order 3, [HW96]

  • Six stage, differential algebraic
    • Stiffly stable rosenbrock method of order 4 with 6 stages, [HW96]

Configuring the rosenbrock solver is as easy as providing the solver with a set of micm::RosenbrockSolverParameters

                         .SetSystem(system)
                         .SetReactions(reactions)
                         .Build();

  auto four_stage = micm::CpuSolverBuilder<micm::RosenbrockSolverParameters>(micm::RosenbrockSolverParameters::FourStageRosenbrockParameters())
                        .SetSystem(system)
                        .SetReactions(reactions)
                        .Build();

  auto four_stage_da = micm::CpuSolverBuilder<micm::RosenbrockSolverParameters>(micm::RosenbrockSolverParameters::FourStageDifferentialAlgebraicRosenbrockParameters())
                            .SetSystem(system)
                            .SetReactions(reactions)
                            .Build();

After that, the usage is the same as a regular solver. A templated method was used here to run the same solving code for each of the different solver configurations.

using namespace micm;

void test_solver_type(auto& solver)
{
  auto state = solver.GetState();

  // mol m-3
  state.variables_[0] = { 1, 0, 0 };

  double k1 = 0.04;
  double k2 = 3e7;
  double k3 = 1e4;
  state.SetCustomRateParameter("r1", k1);
  state.SetCustomRateParameter("r2", k2);
  state.SetCustomRateParameter("r3", k3);

  double temperature = 272.5;  // [K]
  double pressure = 101253.3;  // [Pa]
  double air_density = 1e6;    // [mol m-3]

  state.conditions_[0].temperature_ = temperature;
  state.conditions_[0].pressure_ = pressure;
  state.conditions_[0].air_density_ = air_density;

  // choose a timestep and print the initial state
  double time_step = 200;  // s

  state.PrintHeader();
  state.PrintState(0);

  SolverStats total_stats;
  std::chrono::duration<double, std::nano> total_solve_time = std::chrono::nanoseconds::zero();

  // solve for ten iterations
  for (int i = 0; i < 10; ++i)
  {
    // Depending on how stiff the system is
    // the solver integration step may not be able to solve for the full time step
    // so we need to track how much time the solver was able to integrate for and continue
    // solving until we finish
    double elapsed_solve_time = 0;

    while (elapsed_solve_time < time_step)
    {
      auto start = std::chrono::high_resolution_clock::now();
      solver.CalculateRateConstants(state);
      auto result = solver.Solve(time_step - elapsed_solve_time, state);
      auto end = std::chrono::high_resolution_clock::now();

      total_solve_time += std::chrono::duration_cast<std::chrono::nanoseconds>(end - start);
      total_stats.function_calls_ += result.stats_.function_calls_;
      total_stats.jacobian_updates_ += result.stats_.jacobian_updates_;
      total_stats.number_of_steps_ += result.stats_.number_of_steps_;
      total_stats.accepted_ += result.stats_.accepted_;
      total_stats.rejected_ += result.stats_.rejected_;
      total_stats.decompositions_ += result.stats_.decompositions_;
      total_stats.solves_ += result.stats_.solves_;
      total_stats.singular_ += result.stats_.singular_;

      elapsed_solve_time = result.final_time_;
    }

    state.PrintState(time_step * (i + 1));
  }
  std::cout << "Total solve time: " << total_solve_time.count() << " nanoseconds" << std::endl;
  std::cout << "accepted: " << total_stats.accepted_ << std::endl;
  std::cout << "function_calls: " << total_stats.function_calls_ << std::endl;
  std::cout << "jacobian_updates: " << total_stats.jacobian_updates_ << std::endl;
  std::cout << "number_of_steps: " << total_stats.number_of_steps_ << std::endl;
  std::cout << "accepted: " << total_stats.accepted_ << std::endl;
  std::cout << "rejected: " << total_stats.rejected_ << std::endl;
  std::cout << "decompositions: " << total_stats.decompositions_ << std::endl;
  std::cout << "solves: " << total_stats.solves_ << std::endl;
  std::cout << "singular: " << total_stats.singular_ << std::endl;
}

int main()
{
  auto a = Species("A");
  auto b = Species("B");
  auto c = Species("C");

Running this program should give an output similar to this:

Two stages:
time,         A,         B,         C
   0,  1.00e+00,  0.00e+00,  0.00e+00
 200,  5.37e-01,  4.50e-06,  4.63e-01
 400,  4.51e-01,  3.23e-06,  5.49e-01
 600,  4.01e-01,  2.64e-06,  5.99e-01
 800,  3.65e-01,  2.28e-06,  6.35e-01
1000,  3.38e-01,  2.02e-06,  6.62e-01
1200,  3.16e-01,  1.83e-06,  6.84e-01
1400,  2.97e-01,  1.68e-06,  7.03e-01
1600,  2.82e-01,  1.56e-06,  7.18e-01
1800,  2.68e-01,  1.46e-06,  7.32e-01
2000,  2.56e-01,  1.37e-06,  7.44e-01
Total solve time: 135748 nanoseconds
accepted: 178
function_calls: 356
jacobian_updates: 178
number_of_steps: 178
accepted: 178
rejected: 0
decompositions: 178
solves: 356
singular: 0

Three stages:
time,         A,         B,         C
   0,  1.00e+00,  0.00e+00,  0.00e+00
 200,  5.35e-01,  4.50e-06,  4.65e-01
 400,  4.50e-01,  3.23e-06,  5.50e-01
 600,  4.00e-01,  2.63e-06,  6.00e-01
 800,  3.64e-01,  2.27e-06,  6.36e-01
1000,  3.37e-01,  2.01e-06,  6.63e-01
1200,  3.15e-01,  1.82e-06,  6.85e-01
1400,  2.96e-01,  1.67e-06,  7.04e-01
1600,  2.81e-01,  1.55e-06,  7.19e-01
1800,  2.67e-01,  1.45e-06,  7.33e-01
2000,  2.55e-01,  1.37e-06,  7.45e-01
Total solve time: 110002 nanoseconds
accepted: 127
function_calls: 254
jacobian_updates: 127
number_of_steps: 127
accepted: 127
rejected: 0
decompositions: 127
solves: 381
singular: 0

Four stages:
time,         A,         B,         C
   0,  1.00e+00,  0.00e+00,  0.00e+00
 200,  5.36e-01,  4.49e-06,  4.64e-01
 400,  4.50e-01,  3.20e-06,  5.50e-01
 600,  4.00e-01,  2.62e-06,  6.00e-01
 800,  3.64e-01,  2.26e-06,  6.36e-01
1000,  3.37e-01,  2.01e-06,  6.63e-01
1200,  3.15e-01,  1.82e-06,  6.85e-01
1400,  2.97e-01,  1.67e-06,  7.03e-01
1600,  2.81e-01,  1.55e-06,  7.19e-01
1800,  2.67e-01,  1.45e-06,  7.33e-01
2000,  2.56e-01,  1.37e-06,  7.44e-01
Total solve time: 127291 nanoseconds
accepted: 127
function_calls: 381
jacobian_updates: 127
number_of_steps: 127
accepted: 127
rejected: 0
decompositions: 127
solves: 508
singular: 0

Four stages differential algebraic:
time,         A,         B,         C
   0,  1.00e+00,  0.00e+00,  0.00e+00
 200,  5.36e-01,  4.49e-06,  4.64e-01
 400,  4.51e-01,  3.23e-06,  5.49e-01
 600,  4.00e-01,  2.63e-06,  6.00e-01
 800,  3.64e-01,  2.27e-06,  6.36e-01
1000,  3.37e-01,  2.01e-06,  6.63e-01
1200,  3.15e-01,  1.82e-06,  6.85e-01
1400,  2.97e-01,  1.68e-06,  7.03e-01
1600,  2.81e-01,  1.55e-06,  7.19e-01
1800,  2.68e-01,  1.45e-06,  7.32e-01
2000,  2.56e-01,  1.37e-06,  7.44e-01
Total solve time: 126334 nanoseconds
accepted: 128
function_calls: 384
jacobian_updates: 128
number_of_steps: 128
accepted: 128
rejected: 0
decompositions: 128
solves: 512
singular: 0

Six stages differential algebraic:
time,         A,         B,         C
   0,  1.00e+00,  0.00e+00,  0.00e+00
 200,  5.36e-01,  4.49e-06,  4.64e-01
 400,  4.51e-01,  3.22e-06,  5.49e-01
 600,  4.00e-01,  2.63e-06,  6.00e-01
 800,  3.64e-01,  2.27e-06,  6.36e-01
1000,  3.37e-01,  2.01e-06,  6.63e-01
1200,  3.15e-01,  1.82e-06,  6.85e-01
1400,  2.97e-01,  1.67e-06,  7.03e-01
1600,  2.81e-01,  1.55e-06,  7.19e-01
1800,  2.67e-01,  1.45e-06,  7.33e-01
2000,  2.56e-01,  1.37e-06,  7.44e-01
Total solve time: 174959 nanoseconds
accepted: 127
function_calls: 762
jacobian_updates: 127
number_of_steps: 127
accepted: 127
rejected: 0
decompositions: 127
solves: 762
singular: 0